ADVISORY BOARDS KEN BUCKLE University of New South Wales, Australia
MARY ELLEN CAMIRE University of Maine, USA
ROGER CLEMENS University of Southern California, USA
HILDEGARDE HEYMANN University of California, Davis, USA
ROBERT HUTKINS University of Nebraska, USA
RONALD JACKSON Quebec, Canada
HUUB LELIEVELD Global Harmonization Initiative, The Netherlands
DARYL B. LUND University of Wisconsin, USA
CONNIE WEAVER Purdue University, USA
RONALD WROLSTAD Oregon State University, USA
SERIES EDITORS GEORGE F. STEWART
(1948–1982)
EMIL M. MRAK
(1948–1987)
C. O. CHICHESTER
(1959–1988)
BERNARD S. SCHWEIGERT (1984–1988) JOHN E. KINSELLA
(1989–1993)
STEVE L. TAYLOR
(1995–
)
Academic Press is an imprint of Elsevier 30 Corporate Drive, Suite 400, Burlington, MA 01803, USA 525 B Street, Suite 1900, San Diego, CA 92101-4495, USA 32 Jamestown Road, London NW1 7BY, UK Radarweg 29, PO Box 211, 1000 AE Amsterdam, The Netherlands Linacre House, Jordan Hill, Oxford OX2 8DP, UK First edition 2009 Copyright # 2009 Elsevier Inc. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise without the prior written permission of the publisher. Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone (+44) (0) 1865 843830; fax (+44) (0) 1865 853333; email:
[email protected]. Alternatively you can submit your request online by visiting the Elsevier web site at http://elsevier.com/locate/permissions, and selecting Obtaining permission to use Elsevier material. Notice No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made. ISBN: 978-0-12-374439-5 ISSN: 1043-4526 For information on all Academic Press publications visit our website at elsevierdirect.com Printed and bound in USA 09 10 11 12 10 9 8 7 6 5 4 3 2 1
CONTRIBUTORS Numbers in parentheses indicate the pages on which the authors' contributions begin.
Abby K. van den Berg
Proctor Maple Research Center, University of Vermont, Harvey Road, Underhill Center, Vermont, USA (101) Sinead C. Corr
Department of Biochemistry and Immunology, Trinity College Dublin, Ireland (1) Susan E. Duncan
Virginia Polytechnic Institute and State University (Virginia Tech), Blacksburg, Virginia, USA (17) A. DunnGalvin
Department of Paediatrics and Child Health, Cork University Hospital, Ireland (65) Cormac G. M. Gahan
Department of Microbiology, School of Pharmacy and Alimentary Pharmabiotic Centre, University College Cork, Ireland (1) Colin Hill
Department of Microbiology, University College Cork, Ireland (1) J’ O. B. Hourihane
Department of Paediatrics and Child Health, Cork University Hospital, Ireland (65) Timothy D. Perkins
Proctor Maple Research Center, University of Vermont, Harvey Road, Underhill Center, Vermont, USA (101) R. T. Riley
Toxicology & Mycotoxin Research Unit, USDA Agricultural Research Service, Athens, Georgia, USA (145) M. C. Speer
Center for Human Genetics, Duke University Medical Center, Durham, North Carolina, USA (145)
vii
viii
Contributors
V. L. Stevens
Department of Epidemiology and Surveillance Research, American Cancer Society, Atlanta, Georgia, USA (145) K. A. Voss
Toxicology & Mycotoxin Research Unit, USDA Agricultural Research Service, Athens, Georgia, USA (145) J. Gelineau-van Waes
Department of Genetics, Cell Biology & Anatomy, University of Nebraska Medical Center, Omaha, Nebraska, USA (145) Janet B. Webster
Virginia Polytechnic Institute and State University (Virginia Tech), Blacksburg, Virginia, USA (17)
CHAPTER
1 Understanding the Mechanisms by Which Probiotics Inhibit Gastrointestinal Pathogens Sinead C. Corr,* Colin Hill,† and Cormac G. M. Gahan‡
Contents
Abstract
2 4
I. Introduction II. Evidence for Potential Mechanisms of Action A. Epithelial barrier function and probiotic signaling B. Production of acid and secretion of inhibitory substances C. Immunomodulation D. Inhibition of virulence factor expression III. Conclusions Acknowledgment References
4 7 8 10 11 12 12
In recent years, there has been a growing interest in the use of probiotic bacteria for the maintenance of general gastrointestinal health and the prevention or treatment of intestinal infections. Whilst probiotics are documented to reduce or prevent specific infectious diseases of the GI tract, the mechanistic basis of this effect remains unclear. It is likely that diverse modes-of-action contribute to inhibition of pathogens in the gut environment and proposed mechanisms include (i) direct antimicrobial activity through production of bacteriocins or inhibitors of virulence gene
* Department of Biochemistry and Immunology, Trinity College Dublin, Ireland { {
Department of Microbiology, University College Cork, Ireland Department of Microbiology, School of Pharmacy and Alimentary Pharmabiotic Centre, University College Cork, Ireland
Advances in Food and Nutrition Research, Volume 56 ISSN 1043-4526, DOI: 10.1016/S1043-4526(08)00601-3
#
2009 Elsevier Inc. All rights reserved.
1
2
Sinead C. Corr et al.
expression; (ii) competitive exclusion by competition for binding sites or stimulation of epithelial barrier function; (iii) stimulation of immune responses via increases of sIgA and anti-inflammatory cytokines and regulation of proinflammatory cytokines; and (iv) inhibition of virulence gene or protein expression in gastrointestinal pathogens. In this review, we discuss the modes of action by which probiotic bacteria may reduce gastrointestinal infections, and highlight some recent research which demonstrates the mechanistic basis of probiotic cause and effect.
I. INTRODUCTION The gastrointestinal tract is a complex ecosystem which can be a reservoir of both beneficial and harmful bacteria. Recently, there has been interest in the role of the gut microbiota in health, and also in the deliberate use of bacterial supplements to influence this microbial community in a manner which could potentially assist in maintaining health and in disease prevention (Holzapfel et al., 1998; Senok et al., 2005). Probiotics (defined as any live microorganism, that when administered to human or animal hosts, has health-promoting benefits) could potentially offer an alternative to conventional therapies such as antibiotics for the prophylaxis or treatment of intestinal infections (Bourlioux et al., 2003; Rolfe, 2000). From various in vitro and in vivo studies to date, it is clear that probiotics offer great potential in prevention and treatment of infections (Table 1.1). However, a thorough understanding of their mechanisms of action is required to ensure their efficient use. Probiotics are presumed to modulate the indigenous intestinal flora and improve health via a plethora of potential mechanisms of action, such as immunomodulation, direct antagonism, or competitive exclusion (summarized in Fig. 1.1) (Gotteland et al., 2006; Sartor, 2004; Venturi et al., 1999). Probiotics can inhibit growth of enteric pathogens by decreasing luminal pH, the secretion of bactericidal peptides/proteins, or the stimulation of defensin production by epithelial cells (Toure et al., 2003; Zhu et al., 2000). Probiotics can also block attachment to or invasion of the intestinal epithelium by pathogens through blocking of epithelial surface receptors or induction of mucins, large carbohydrate molecules which form a barrier along the epithelial monolayer (Mack et al., 1999; Mattar et al., 2002). A number of in vivo studies have been performed which have determined the probiotic capabilities of such strains. While these studies have been important in demonstrating probiotic efficacy against various infectious diseases, few have specifically identified the mechanistic basis behind the observed benefits, and many rely on in vitro data to decipher the possible mechanism of action. However, what has emerged to date is that
TABLE 1.1 A list of potential probiotic strains and their observed beneficial effects Organism
Effect
Mechanism
Reference
E. coli strain Nissle 1917
Increases tight junction protein expression, ZO-2 Prevents the pathogen-induced drop in transepithelial resistance Increases extracellular secretion of mucin, MUC3 Induces human b-defensin, hBD-2 gene expression Production of acetic acid and thus lowering of pH Production of hydrogen peroxide
Zyrek et al. (2007)
L. casei NCDO1205 L. rhamnosus GG
Improves epithelial barrier function Improves epithelial barrier function Improves epithelial barrier function Improves epithelial barrier function Secretion of inhibitory substances Secretion of inhibitory substances Secretion of inhibitory substances Immunomodulation Immunomodulation
B. lactis Bb-12
Immunomodulation
Decrease IL-8 and increase IL-10 response Activates NF-kB and regulates inflammatory response in macrophages Stimulates sIgA
L. rhamnosus GG
Inhibition of virulence factor expression Inhibition of virulence factor expression
Reduces expression of genes encoding shiga toxin Inhibition of urease activity in Y. enterocolitica
L. rhamnosus R0011 L. plantarum 299v VSL#3 probiotic mixture B. breve Yakult L. johnsonii NCC533 L. salivarius UCC118
L. plantarum ITM21B
Production of bacteriocin
Sherman et al. (2005) Mack et al. (2003) Schlee et al. (2008) Asahara et al. (2004) Pridmore et al. (2008) Corr et al. (2007b) Corr et al. (2007a) Miettinen et al. (2000) Fukushima et al. (1998) Carey et al. (2008) Lavermicocca et al. (2008)
4
Sinead C. Corr et al.
Probiotic Transient flora (including probiotics and pathogens)
Pathogen 1
3
Mucus
Commensal microbiota M Epithelium (including immune cells)
4
E 2
T DC
FIGURE 1.1 Probiotics may protect against infection by pathogens through (1) Direct antagonism via bacteriocin production. (2) Immunomodulation via immune cell (T-cell, Dendritic cell) activation. (3) Improvement of epithelial barrier function and competitive exclusion via induction of mucus and blocking of epithelial binding receptors. (4) Strengthening of epithelial tight junctions by increased expression of tight junction proteins, or by a combination of these mechanisms.
the inhibition of pathogens by specific probiotics may represent a highly specific commensal–pathogen interaction. It is clear that further understanding of this phenomenon is required in order to specifically target gastrointestinal pathogens through the use of appropriate probiotic strains.
II. EVIDENCE FOR POTENTIAL MECHANISMS OF ACTION A. Epithelial barrier function and probiotic signaling A key mechanism by which probiotics are thought to exert anti-invasive activity is via induction of conformational changes within the epithelial monolayer (Mack et al., 1999). In a recent study of barrier disruption in T84 epithelial cells by infection with enteropathogenic Escherichia coli, coincubation with the probiotic E. coli strain Nissle 1917 (EcN) or addition of the probiotic after infection abolished this disruption and restored barrier integrity (Zyrek et al., 2007). DNA-microarray analysis identified more than 300 genes exhibiting altered expression following incubation of the epithelial cells with EcN, including expression and distribution of zonula occludens-2 (ZO-2), a tight-junction protein. Further studies have shown that pretreatment of epithelial monolayers with probiotic bacteria, Lactobacillus acidophilus R0052 and Lactobacillus
Mechanisms of Probiotic Action
5
rhamnosus R0011, reduces epithelial injury following exposure to E. coli O157:H7 and E. coli O127:H6 by preventing the pathogen-induced drop in transepithelial resistance, a measure of barrier integrity (Sherman et al., 2005). These probiotics also reduced the number of foci of rearrangements of a-actinin, indicative of reduced number of attaching and effacing lesions formed in response to E. coli O157:H7. In this study, viable lactic acid-producing bacteria were necessary to mediate the observed effects. In another recent study preincubation of Hep-2 cell monolayers with two strains of lactobacilli, Lactobacillus delbrueckii subsp. lactis CIDCA 133 and Lactobacillus plantarum CIDCA 83114 prior to infection with enterohaemorrhagic E. coli (EHEC) minimized F-actin rearrangements and morphological alterations in the cell monolayers (Hugo et al., 2008). These studies collectively indicate that lactobacilli are capable of directly triggering cellular responses in host cells that may impede virulence mechanisms of EHEC. The exact molecular mechanisms by which probiotics stimulate alterations in epithelial cell function are currently under investigation. Studies have shown that probiotic strains such as the VSL#3 probiotic compound (Bifidobacterium longum, Bifidobacterium infantis, Bifidobacterium breve, L. acidophilus, Lactobacillus casei, L. delbrueckii subsp. bulgaricus, L. plantarum, Streptococcus salivarius subsp. thermophilus) can improve epithelial and mucosal barrier function through production of specific metabolites (Madsen et al., 2001). These include production of short-chain fatty-acids (SCFAs) as byproduct of microbial fermentation, such as butyrate which induces epithelial cell differentiation and increases barrier integrity (Cook and Sellin, 1998). L. acidophilus has been shown to improve gut barrier function in rats by improving microflora disturbance, increasing occludin expression, and maintaining the gut epithelial tight junction (Qin et al., 2005). Another physiological change potentially induced by probiotics in the host involves induction or overexpression of mucin. GI tract mucins are large, carbohydrate-rich, high-molecular-weight glycoproteins which are the major components of the mucous layer overlying the intestinal epithelium (Mattar et al., 2002). Mucin forms a physicochemical barrier which protects epithelial cells from chemical, enzymatic, mechanical, and microbial damage, and limits microbial adherence and subsequent invasion (Mack et al., 2003). At least 12 mucin genes have been identified, and of these MUC2 and MUC3 are the predominant ileocolonic mucins (Mack et al., 2003). The MUC2 gene is expressed in goblet cells of the small and large intestine and is the major secreted mucin of the colon (Mack et al., 1999). The membrane-associated mucin MUC3 is not highly expressed in the colon but is expressed on both goblet cells and enterocytes of the small intestine (Chang et al., 1994). Adherence of selected Lactobacillus strains (L. plantarum 299v, L. rhamnosus GG) to the human intestinal HT29 epithelial-cell line induces up-regulation of mucin gene expression, and correlates with increased extracellular secretion of MUC3
6
Sinead C. Corr et al.
(Mack et al., 2003). L. plantarum 299v and L. rhamnosus GG inhibit the adherence of enteropathogenic E. coli to HT29 intestinal epithelial cells via induction or overexpression of mucin (Mack et al., 1999). In an in vitro Caco-2 cell model, L. casei LGG up-regulates MUC2 expression and has an inhibitory effect on bacterial translocation of the intestinal epithelium (Mattar et al., 2002). Thus, increased expression of intestinal mucin in response to lactobacilli mediates inhibition of adherence of pathogens to intestinal cells. However, analysis of this phenomenon using in vivo infection models has not yet been implemented. Interestingly, Collado and colleagues (2008) have recently shown that specific probiotic strains have the capacity to prevent adhesion of the opportunistic pathogen Enterobacter sakazakii to immobilized human mucous in vitro. These studies indicate that in addition to inducing upregulation of mucous secretion by the epithelia, specific probiotic strains also have the capacity to competitively exclude or displace pathogens from human mucous as a mechanism for preventing the transient colonization of gastrointestinal pathogens. Potential probiotic strains can also induce the release of defensins from epithelial cells. These small peptides/proteins are active against bacteria, fungi, and viruses and also stabilize gut barrier function (Furrie et al., 2005). It has been shown that E. coli Nissle 1917 induces human b-defensin-2 (hBD-2) gene expression in Caco-2 intestinal epithelial cells (Wehkamp et al., 2004). This induction was mediated by NF-kB and AP-1 signaling pathways. Recently, several strains including E. coli Nissle 1917, L. acidophilus, Lactobacillus fermentum, Lactobacillus paracasei subsp. paracasei, Pediococcus pentosaceus, and the VSL#3 probiotic mixture were found to induce hBD-2 gene expression in Caco-2 cells (Schlee et al., 2008). This was also dependent on mitogen-activated protein kinase (MAPK), NF-kB, and AP-1 signaling pathways (Schlee et al., 2008). This induction of hBD-2 may also enhance mucosal barrier function. The adhesion ability of some probiotic strains affords probiotic bacteria the capacity to compete with pathogenic bacteria for receptors expressed on epithelial cells, thus blocking contact between epithelial cells and pathogenic bacteria (Sherman et al., 2005; Tsai et al., 2005). In a recent study, BALB/c mice were fed L. acidophilus LAP5 or L. fermentum LF33 originally isolated from swine and poultry for seven consecutive days before oral challenge with Salmonella enterica serovar Typhimurium (Tsai et al., 2005). Numbers of Salmonella invading livers and spleens of probiotic-fed mice were significantly lower than placebo controls, and it was thought that the adhesiveness of Lactobacillus cells to mouse intestinal epithelium may be an important factor for their antagonistic activity against Salmonella invasion in vivo. However, this inference was based upon in vitro assessment of adherence to intestinal cell lines and was not proven in vivo (Tsai et al., 2005).
Mechanisms of Probiotic Action
7
B. Production of acid and secretion of inhibitory substances Lactobacillus and Bifidobacterium spp. are capable of producing organic acids as end products of metabolism. Selected Bifidobacterium species, including B. breve strain Yakult, display anti-infectious activity against Shiga toxin-producing E. coli (STEC) O157:H7 in mice (Asahara et al., 2004). In this study, B. breve Yakult was administered to mice daily for three consecutive days and mice were infected with STEC on day 3. A dramatic decrease in bodyweight and subsequent death was observed in placebo-fed mice, while bodyweight was maintained and no fatalities were observed in B. breve-fed mice. This anti-infective activity was thought to be due to production of acetic acid by B. breve and lowering of intestinal pH, which had the combined effect of inhibiting Shiga-like toxin (Stx) production (Asahara et al., 2004). Lactobacillus and Bifidobacterium spp. have been shown to impede infection of human intestinal cells by enterohemorrhagic E. coli O157:H7 by the combined action of lactic acid and proteinaceous substances (Gopal et al., 2001). An in vitro study of the ability of L. rhamnosus DR20 and Bifidobacterium lactis DR10 to impede infection of differentiated human intestinal cell-lines by E. coli O157:H7 found that pretreatment of E. coli with concentrated cell-free culture supernatants from these probiotic bacteria significantly reduced numbers of culturable E. coli and the invasiveness of this strain (Gopal et al., 2001). The probiotic E. coli strain Nissle 1917 interferes with S. Typhimurium invasion of human embryonic intestinal epithelial INT407 cells via secretion of inhibitory substances, as shown when the probiotic was separated from the bacteria by a nonpermeable membrane (Altenhoefer et al., 2004). In a previous study, we utilized a similar transwell chamber system to demonstrate that lactobacilli and bifidobacteria (L. casei, L. acidophilus, Lactobacillus salivarius, B. breve, B. infantis, B. longum) are capable of inhibiting Listeria monocytogenes invasion of C2Bbe1 epithelial cells in the absence of direct contact through secretion of proteinaceous molecule(s), active at low pH in the case of the lactobacilli strains tested (Corr et al., 2007a). However, the nature of the proteinaceous agent needs to be identified. Recently, Pridmore and co-workers (2008) have examined the production of hydrogen peroxide by the human gastrointestinal isolate Lactobacillus johnsonii NCC533. Through in silico analysis of the genome of this potential probiotic strain they identified the means by which hydrogen peroxide is synthesized. Furthermore, they demonstrated that the strain actively produced hydrogen peroxide in vitro at levels that were inhibitory for S. Typhimurium. Bacteriocins are compounds with potential anti-microbial activity synthesized by many bacterial species, including lactic acid bacteria (Cotter et al., 2005; Gotteland et al., 2006). As the ability of bacteriocins to
8
Sinead C. Corr et al.
inhibit or kill pathogens is well documented, these molecules represent obvious candidates as mediators of an antipathogen effect. Indeed, bacteriocins have been shown to be necessary in vivo for long-term oral colonization by a noncariogenic variant of Streptococcus mutans in a therapeutic approach known as replacement therapy (Smith et al., 2006). In a recent study, we demonstrated the ability of L. salivarius UCC118 to inhibit L. monocytogenes infection of mice, and directly linked this inhibitory effect to production of bacteriocin by L. salivarius (Corr et al., 2007b). We showed that mice orally inoculated with L. salivarius UCC118 were protected from subsequent oral infection by L. monocytogenes. However, a stable mutant of L. salivarius UCC118 that is unable to produce the bacteriocin, Abp118, failed to protect mice confirming that bacteriocin production is the primary mediator of protection against this organism. Furthermore, L. salivarius UCC118 did not offer any protection when mice were infected with a strain of L. monocytogenes expressing the cognate Abp118 bacteriocin immunity protein AbpIM again confirming that the observed protective effect was the result of direct antagonism between L. salivarius and the pathogen, mediated by the bacteriocin Abp118.
C. Immunomodulation Probiotic bacteria are capable of tempering the host inflammatory response to infection and are considered to be important mediators of immune-regulation in the gastrointestinal environment (Corr et al., 2007a; O’Hara et al., 2006). It is likely that this immunomodulatory role is an important factor governing the immune clearance of gastrointestinal pathogens and in preventing the establishment of postinfectious inflammatory conditions (including irritable bowel syndrome, IBS) in the GI tract. Furthermore, chronic inflammatory diseases of the GI tract (including Crohn’s disease) are postulated to be linked to underlying infections (by Mycobacterium avium subsp. paratuberculosis or specific E. coli strains) (Darfefeuille-Michaud et al., 2004; Sechi et al., 2004). Probiotic treatment raises the possibility that such chronic infections may be amenable to noninvasive intervention in order to limit the cause of the underlying inflammation. Probiotic bacteria regulate mucosal immune responses through induction of anti-inflammatory cytokines such as IL-10 and TGF-b, while decreasing expression of proinflammatory cytokines, such as TNF and IFN-g (Corr et al., 2007a; Di Giacinto et al., 2005; Silva et al., 2004). B. breve and Streptococcus thermophilus secrete metabolites which inhibit LPSinduced TNF-a secretion from peripheral blood mononuclear cell (PBMC) monolayers (Menard et al., 2004). We demonstrated a significant reduction in interleukin-8 (IL-8) and an increase in IL-10 cytokines secreted from epithelial cells following pretreatment with probiotics
Mechanisms of Probiotic Action
9
prior to infection with L. monocytogenes (Corr et al., 2007a). A number of commensal strains including L. casei NCDO1205, L. salivarius UCC118, and B. breve UCC2003 were capable of inducing this response. Similarly, both B. infantis 35624 and L. salivarius UCC118 are capable of reducing S. typhimurium-induced proinflammatory responses in vitro (O’Hara et al., 2006). These probiotic commensal strains were capable of blunting IL-8 responses and increasing the IL-10 response in an in vitro model of Salmonella infection. The mechanistic basis of such responses has been examined by Kelly and co-workers (2004). Bacteroides thetaiotaomicron reduces inflammation due to Salmonella–TLR5 interactions (Kelly et al., 2004). The mechanism underpinning this anti-inflammatory response was dependent upon PPAR-g (peroxisome proliferator activated receptor- g)-mediated inhibition of NF-kB and was directly induced by B. thetaiotaomicron. Furthermore, L. rhamnosus GG is capable of activating NF-kB and STATs, latent cytoplasmic transcription factors which regulate transcription of genes encoding proteins involved in cytokine signaling and inflammatory responses in macrophages (Miettinen et al., 2000). Some probiotics also stimulate secretory IgA production and activate regulatory T cells (Fukushima et al., 1998). These effects have been seen in human studies and demonstrate that anti-Polio sIgA is increased in those administered a probiotic preparation viable B. lactis Bb-12. Similarly, an increase in IgAþ cells was witnessed in mice administered L. casei (Galdeano and Perdigo´n, 2006). However, other studies have demonstrated that stimulation of sIgA in humans is stimulated by a prebiotic preparation but not by administration of live probiotic (Bifidobacterium animalis) (Bakker-Zierikzee et al., 2006). Inflammatory conditions of the GI tract may be initiated by a disregulated local immune response to the normal microbiota and are host dependent (Sartor, 2003; Shanahan, 2001). However, a subset of IBS patients experience symptoms following gastrointestinal infection (postinfectious IBS). In addition, underlying infection has been proposed as a possible trigger in Crohn’s disease and both M. avium subsp. paratuberculosis or adherent invasive E. coli (AIEC) have been suggested as possible sources of inflammation (Darfefeuille-Michaud et al., 2004; Sechi et al., 2004). Ingrassia and co-workers have demonstrated that L. casei DN-114 001 is capable of inhibiting AIEC strains isolated from Crohn’s disease patients in cell culture models of infection, suggesting that probiotic intervention may present a future strategy for limiting the pathogenesis of a potential trigger of inflammation in Crohn’s disease (Ingrassia et al., 2005). Indeed, human studies indicate that specific probiotic strains can reduce symptoms of IBS through immunomodulation (Kajander et al., 2008; O’Mahony et al., 2001) and may have promise for the treatment of
10
Sinead C. Corr et al.
inflammatory bowel disease (IBD) although further research is needed (Hedin et al., 2007). Recently, L. acidophilus has been shown to reduce the inflammatory response in gastric epithelial cells via production of conjugated linoleic acids (CLA) (Kim et al., 2008). In this study, conditioned medium containing L. acidophilus-producing CLA interacts with IkB kinase inducing phosphorylation of inhibitory IkBa leading to its dissociation from NF-kB and thus, NF-kB activation. Lactobacillus reuteri has recently been shown to secrete factors which potentiate apoptosis by stabilizing IkBa degradation and inhibiting nuclear translocation of p65, thus leading to suppression of NF-kB-dependent gene products that mediate cell proliferation and cell survival including Cox-2 and Bcl-2, respectively (Iyer et al., 2008). Promotion of cell apoptosis serves as a therapy to prevent colorectal cancer and IBD (Iyer et al., 2008). The VSL#3 probiotic mix which contains viable lyophilized bifidobacteria (B. longum, B. infantis, and B. breve), lactobacilli (L. acidophilus, L. casei, L. delbrueckii subsp. bulgaricus, and L. plantarum), and S. salivarius subsp. thermophilus (VSL Pharmaceuticals, Fort Lauderdale, FL), can significantly modulate the immune response and has been shown to play a role in maintenance of treatment in ulcerative colitis (Venturi et al., 1999). In this study, patients with ulcerative colitis in remission were given VSL#3 for 12 months and it was shown that of those taking the probiotic, the majority remained in remission throughout the study period. Recently, it was shown that culturing human blood dendritic cells with cell-wall components of the probiotic mixture VSL#3 induced dendritic cell maturation and up-regulated production of IL-10 (Hart et al., 2004). Dendritic cells, which play an important role in early bacterial recognition and in T-cell responses, may be central mediators of these probiotic effects. Indeed, administration of VSL#3 is associated with an early increase in IL-10 production and regulatory CD4þ T cells bearing surface TGF-b in murine models of colitis, while human studies have shown increased mucosal regulatory T cells and a reduction in pouchitis disease activity (Di Giacinto et al., 2005; Pronio et al., 2008). L. acidophilus strain L-92 has recently been shown to regulate both Th1 and Th2 cytokine responses in BALB/c mice possibly through modulation of TGF-bassociated activation of T-regulatory cells, suggesting a potential therapy for Th1- and Th2-mediated disease including autoimmune disease and inflammatory diseases (Torii et al., 2007).
D. Inhibition of virulence factor expression A potential mechanism of action by which potential probiotic strains may impede pathogens is through the modulation of gene and/or protein expression patterns through bacterial signaling mechanisms. Interestingly, cell-free supernatants of L. acidophilus have been shown to inhibit
Mechanisms of Probiotic Action
11
quorum sensing and virulence gene expression in E. coli O157:H7 but did not affect expression of shiga toxin in this strain (Medellin-Pen˜a et al., 2007). Other researchers have utilized microarray analyses to investigate the global transcriptional changes in E. coli O157:H7 following coincubation with L. rhamnosus GG (LGG). Results indicated that LGG coincubation reduces expression of the stx genes encoding shiga toxin production in E. coli O157:H7 (Carey et al., 2008). Subsequently, a variety of Lactobacillus, Pediococcus, and Bifidobacterium strains (L. rhamnosus GG, Lactobacillus curvatus, L. plantarum, Lactobacillus jensenii, L. acidophilus, L. casei, L. reuteri, Pediococcus acidilactici, Pediococcus cerevisiae, P. pentosaceus, Bifidobacterium thermophilum, Bifidobacterium boum, Bifidobacterium suis, and B. animalis) were shown to repress stxA expression in this model system, suggesting a global mechanism by which the microbiota could impede virulence factor expression in this pathogen (Carey et al., 2008). Similarly, a recent study examined the ability of a variety of potential probiotic strains to inhibit the ureolytic pathogen Yersinia enterocolitica (Lavermicocca et al., 2008). They determined that one probiotic strain, L. plantarum ITM21B, was capable of inhibiting urease activity in the pathogen. Overall, it is likely that future studies will uncover the regulatory networks that govern signaling mechanisms between pathogens and commensals.
III. CONCLUSIONS There is mounting evidence to support a role for probiotics as an alternative to conventional methods for prevention and treatment of intestinal diseases and inflammatory disorders. The introduction of probiotic organisms has been proposed to improve digestive function (Savaiano et al., 1984), reduce chronic inflammation (Di Giacinto et al., 2005; O’Hara et al., 2006), and improve recovery from foodborne disease (Aiba et al., 1998). Previous work using rodent models of disease has demonstrated a role for probiotics in the amelioration of infections caused by Helicobacter pylori (Gotteland et al., 2006), Citrobacter rodentium (a murine model of Enteropathogenic E. coli (EPEC)) ( Johnson-Henry et al., 2005) and S. Typhimurium (Silva et al., 2004) and clinical trials have shown that administration of probiotics can significantly improve eradication of H. pylori in infected patients (Gotteland et al., 2006). In vitro analyses have indicated that regulation of mucous production by probiotics can prevent colonization by EPEC (Mack et al., 1999) and there is an apparent correlation between immunomodulation by probiotics and elimination of foodborne pathogens (Jijon et al., 2004). Efficient use of probiotic therapies will require that the precise mechanism(s) by which specific probiotic strains exert their effect is identified. While the molecular details underpinning probiotic modes of action remain almost entirely unknown,
12
Sinead C. Corr et al.
recently there has been significant progress towards understanding how probiotics exert their beneficial effects at the molecular level. This suggests that the next phase of therapeutic development will represent a ‘‘bugs to drugs’’ approach whereby probiotic-based therapeutic agents are developed as specific pharmabiotics (O’Hara and Shanahan, 2007).
ACKNOWLEDGMENT The authors wish to acknowledge funding by the Irish Government through the continued support of Science Foundation Ireland for the Alimentary Pharmabiotic Centre, University College Cork (http//apc.ucc.ie).
REFERENCES Aiba, Y., Suzuki, N., Kabir, A. M., Takagi, A., and Koga, Y. (1998). Lactic acid-mediated suppression of Helicobacter pylori by the oral administration of Lactobacillus salivarius as a probiotic in a gnotobiotic murine model. Am. J. Gastroenterol. 93, 2097–2101. Altenhoefer, A., Oswald, S., Sonnenborn, U., Enders, C., Schulze, J., Hacker, J., and Oelschlaeger, T. A. (2004). The probiotic Escherichia coli strain Nissle 1917 interferes with invasion of human intestinal epithelial cells by different enteroinvasive bacterial pathogens. FEMS Immunol. Med. Microbiol. 40, 223–229. Asahara, T., Shimizu, K., Nomoto, K., Hamabata, T., Ozawa, A., and Takeda, Y. (2004). Probiotic Bifidobacteria protect mice from lethal infection with Shiga toxin-producing Escherichia coli O157:H7. Infect. Immun. 72(4), 2240–2247. Bakker-Zierikzee, A. M., Tol, E. A., Kroes, H., Alles, M. S., Kok, F. J., and Bindels, J. G. (2006). Faecal SIgA secretion in infants fed on pre- or probiotic infant formula. Pediatr. Allergy Immunol. 17(2), 134–140. Bourlioux, P., Koletzko, B., Guarner, F., and Braesco, V. (2003). The intestine and its microflora are partners for the protection of the host: Report on the Danone Symposium ‘‘The intelligent Intestine.’’ Am. J. Clin. Nutr. 78, 675–683. Carey, C. M., Kostryznska, M., Ojha, S., and Thompson, S. (2008). The effect of probiotics and organic acids on Shiga-toxin 2 gene expression in enterohemorrhagic Escherichia coli O157:H7. J. Microbiol. Methods [Epub.]. Chang, S. K., Dohrman, A. F., Basbaum, C. B., Ho, S. B., Tsuda, T., Toribara, N. W., Gum, J. R., and Kim, Y. S. (1994). Localization of mucin (MUC2 and MUC3) messenger RNA and peptide expression in human normal intestine and colon cancer. Gastroenterology 107, 28–36. Collado, M. C., Isolauri, E., and Salminen, S. (2008). Specific probiotic strains and their combinations counteract adhesion of Enterobacter sakazakii to intestinal mucus. FEMS Microbiol. Lett. [Epub ahead of print]. Cook, S. I. and Sellin, J. H. (1998). Review article: Short chain fatty acids in health and disease. Aliment. Pharmacol. Ther. 12, 499–507. Corr, S. C., Gahan, C. G., and Hill, C. (2007a). Impact of selected Lactobacillus and Bifidobacterium species on Listeria monocytogenes infection and the mucosal immune response. FEMS Immunol. Med. Microbiol. 50(3), 380–388. Corr, S. C., Li, Y., Riedel, C. U., O’Toole, P. W., Hill, C., and Gahan, C. G. (2007b). Bacteriocin production as a mechanism for the antiinfective activity of Lactobacillus salivarius UCC118. Proc. Natl. Acad. Sci. USA 104(18), 7617–7621.
Mechanisms of Probiotic Action
13
Cotter, P. D., Hill, C., and Ross, R. P. (2005). Bacteriocins: Developing innate immunity for food. Nat. Rev. Microbiol. 3, 777–788. Darfefeuille-Michaud, A., Boudeau, J., Bulois, P., Neut, C., Glasser, A. L., Barnich, N., Bringer, M. A., Swidsinski, A., Beaugerie, L., and Colombel, F. (2004). High prevalence of adherent-invasive Escherichia coli associated with ileal mucosa in Crohn’s disease. Gastroenterology 127(2), 412–421. Di Giacinto, C., Marinaro, M., Sanchez, M., Strober, W., and Boirivant, M. (2005). Probiotics ameliorate recurrent Th1-mediated murine colitis by inducing IL-10 and IL-10-dependent TGF-b-bearing regulatory cells. J. Immunol. 174, 3237–3246. Fukushima, Y., Kawata, Y., Hara, H., Terada, A., and Mitsuoka, T. (1998). Effect of probiotic formula on intestinal immunoglobulin A production in healthy children. Int. J. Food Microbiol. 42(1–2), 39–44. Furrie, E., Macfarlane, S., Kennedy, A., Cummings, J. H., Walsh, S. V., O’Neil, D. A., and Macfarlane, G. T. (2005). Synbiotic therapy (Bifidobacterium longum/Synergy 1) initiates resolution of inflammation in patients with active ulcerative colitis: A randomised controlled pilot trial. Gut 54(2), 242–9. Galdeano, C. M. and Perdigo´n, G. (2006). The probiotic bacterium Lactobacillus casei induces activation of the gut mucosal immune system through innate immunity. Clin. Vaccine Immunol. 13(2), 219–226. Gopal, P. K., Prasad, J., Smart, J., and Gill, H. S. (2001). In vitro properties of Lactobacillus rhamnosus DR20 and Bifidobacterium lactis DR10 strains and their antagonistic activity against an enterotoxigenic Escherichia coli. Int. J. Food Microbiol. 67, 207–216. Gotteland, M., Brunser, O., and Cruchet, S. (2006). Systematic review: Are probiotics useful in controlling gastric colonization by Helicobacter pylori? Aliment. Pharmacol. Ther. 23(8), 1077–1086. Hart, A. L., Lammers, K., Brigidi, P., Vitali, B., Rizzello, F., Gionchetti, P., Campieri, M., Kamm, M. A., Knight, S. C., and Stagg, A. J. (2004). Modulation of human dendritic cell phenotype and function by probiotic bacteria. Gut 53, 1602–1609. Hedin, C., Whelan, K., and Lindsay, J. O. (2007). Evidence for the use of probiotics and prebiotics in inflammatory bowel disease: A review of clinical trials. Proc. Nutr. Soc. 66(3), 307–315. Holzapfel, W. H., Haberer, P., Snel, J., Schillinger, U., and Huis in’t Veld, J. H. J. (1998). Overview of gut flora and probiotics. Int. J. Food Microbiol. 41, 85–101. Hugo, A. A., Kakisu, E., De Antoni, G. L., and Pe´rez, P. F. (2008). Lactobacilli antagonize biological effects of enterohaemorrhagic Escherichia coli in vitro. Lett. Appl. Microbiol. 46(6), 613–619. Ingrassia, I., Leplingard, A., and Darfeuille-Michaud, A. (2005). Lactobacillus casei DN-114 001 inhibits the ability of adherent-invasive Escherichia coli isolated from Crohn’s disease patients to adhere to and to invade intestinal epithelial cells. Appl. Environ. Micorbiol. 71(6), 2880–2887. Iyer, C., Kosters, A., Sethi, G., Kunnumakkara, A. B., Aggarwal, B. B., and Versalovic, J. (2008). Probiotic Lactobacillus reuteri promotes TNF-induced apoptosis in human myeloid leukemia-derived cells by modulation of NF-kappaB and MAPK signalling. Cell Microbiol. 10(7), 1442–1452. Jijon, H., Backer, J., Diaz, H., Yeung, H., Theil, D., McKaigney, C., DeSimone, C., and Madsen, K. (2004). DNA from probiotic bacteria modulates murine and human epithelial and immune function. Gastroenterology 126, 1358–1373. Johnson-Henry, K. C., Nadjafi, M., Avitzur, Y., Mitchell, D. J., Ngan, B. Y., Galindo-Mata, E., Jones, N. L., and Sherman, P. M. (2005). Amelioration of the effects of Citrobacter rodentium infection in mice by pretreatment with probiotics. J. Infect. Dis. 191(12), 2106–2117. Kajander, K., Myllyluoma, E., Rajilic-Stojanovic, M., Kyronpalo, S., Rasmussen, M., Jarvenpaa, S., Zoetendal, E. G., de Vos, W. M., Vapaatalo, H., and Korpela, R. (2008).
14
Sinead C. Corr et al.
Clinical trial: Multispecies probiotic supllementation alleviates the symptoms of irritable bowel syndrome and stabilizes intestinal microbiota. Aliment. Pharmacol. Ther. 27(1), 48–57. Kelly, D., Campbell, J. I., King, T. P., Grant, G., Jansson, E. A., Coutts, A. G., Petersson, S., and Conway, S. (2004). Commensal anaerobic gut bacteria attenuate inflammation by regulating nuclear-cytoplasmic shuttling of PPAR-gamma and RelA. Nat. Immunol. 5(1), 104–112. Kim, J. M., Kim, J. S., Kim, Y. J., Oh, Y. K., Kim, I. Y., Chee, Y. J., Han, J. S., and Jung, H. C. (2008). Conjugated linoleic acids produced by Lactobacillus dissociates IKK-g and Hsp90 complex in Helicobacter pylori-infected gastric epithelial cells. Lab. Invest. 88(5), 541–542. Lavermicocca, P., Valerio, F., Lonigro, S. L., Di Leo, A., and Visconti, A. (2008). Antagonistic activity of potential probiotic Lactobacilli against the ureolytic pathogen Yersinia enterocolitica. Curr. Microbiol. 56(2), 175–181. Mack, D. R., Michail, S., Wel, S., McDougall, L., and Hollingsworth, M. A. (1999). Probiotics inhibit enteropathogenic E. coli adherence in vitro by inducing intestinal mucin gene expression. Am. J. Physiol. 276(4), G941–G950. Mack, D. R., Ahrne, S., Hyde, L., Wei, S., and Hollingsworth, M. A. (2003). Extracellular MUC3 mucin secretion follows adherence of Lactobacillus strains to intestinal epithelial cells in vitro. Gut 52, 827–833. Madsen, K., Cornish, A., Soper, P., McKaigney, C., Jijon, H., Yachimec, C., Doyle, J., Jewell, L., and DeSimone, C. (2001). Probiotic bacteria enhance murine and human intestinal epithelial barrier function. Gastroenterology 121(3), 580–591. Mattar, A. F., Teitelbaum, D. H., Drongowski, R. A., Yongyi, F., Harmon, C. M., and Coran, A. G. (2002). Probiotics up-regulate MUC-2 mucin gene expression in a Caco-2 cell-culture model. Pediatr. Surg. Int. 18, 586–590. Medellin-Pena, M. J., Wang, H., Johnson, R., Anand, S., and Griffiths, M. W. (2007). Probiotics affect virulence-related gene expression in Escherichia coli O157:H7. Appl. Environ. Microbiol. 73(13), 4259–4267. Menard, S., Candalh, C., Bambou, J. C., Terpend, K., Cerf-Bensussan, N., and Heyman, M. (2004). Lactic acid bacteria secrete metabolites retaining anti-inflammatory properties after intestinal transport. Gut 53, 821–828. Miettinen, M., Lehtonen, A., Julkunen, I., and Matikainen, S. (2000). Lactobacilli and streptococci activate NF-kB and STAT signalling pathways in human macrophages. J. Immunol. 164, 3733–3740. O’Hara, A. M. and Shanahan, F. (2007). Gut microbiota: Mining for therapeutic potential. Clin. Gastroenterol. Hepatol. 5(3), 274–284. O’Hara, A. M., O’Regan, P., Fanning, A., O’Mahony, C., MacSharry, J., Lyons, A., Bienenstock, J., O’Mahony, L., and Shanahan, F. (2006). Functional modulation of human intestinal epithelial cell responses by Bifidobacterium infantis and Lactobacillus salivarius. Immunology 118, 202–215. O’Mahony, L., Feeney, M., O’Halloran, S., Murphy, L., Kiely, B., Fitzgibbon, J., Lee, G., O’Sullivan, G., Shanahan, F., and Collins, J. K. (2001). Probiotic impact on microbial flora, inflammation and tumour development in IL-10 knockout mice. Aliment. Pharmacol. Ther. 15, 1219–1225. Pridmore, R. D., Pittet, A. C., Praplan, F., and Cavadini, C. (2008). Hydrogen peroxide production by Lactobacillus johnsonii NCC 533 and its role in anti-Salmonella activity. FEMS Microbiol. Lett. 283(2), 210–215. Pronio, A., Montesani, C., Butteroni, C., Vecchione, S., Mumolo, G., Vestri, A., Vitolo, D., and Boirivant, M. (2008). Probiotic administration in patients with ileal pouch-anal anastomosis for ulcerative colitis is associated with expansion of mucosal regulatory cells. Inflamm. Bowel Dis. 14(5), 662–668. Qin, H. L., Shen, T. Y., Gao, Z. G., Fan, X. B., Hang, X. M., Jiang, Y. Q., and Zhang, H. Z. (2005). Effect of lactobacillus on the gut microflora and barrier function of the rats with abdominal infection. World J. Gastroenterol. 11(17), 2591–2596.
Mechanisms of Probiotic Action
15
Rolfe, R. D. (2000). The role of probiotic cultures in the control of gastrointestinal health. J. Nutr. 130, 396S–402S. Sartor, R. B. (2003). Targeting enteric bacteria in treatment of inflammatory bowel diseases: Why, how and when. Curr. Opin. Gastroenterol. 19(4), 358–365. Sartor, R. B. (2004). Probiotic therapy of intestinal inflammation and infections. Curr. Opin. Gastroenterol. 21, 44–50. Savaiano, D. A., AbouElAnour, A., Smith, D. E., and Levitt, M. D. (1984). Lactose malabsorption from yogurt, pasteurized yogurt, sweet acidophilus milk, and cultured milk in lactase-deficient individuals. Am. J. Clin. Nutr. 40(6), 1219–1223. Schlee, M., Harder, J., Koten, B., Stange, E. F., Wehkamp, J., and Fellermann, K. (2008). Probiotic lactobacilli and VSL#3 induce enterocyte b-defensin 2. Clin. Exp. Immunol. 151 (3), 528–535. Sechi, L. A., Mura, M., Tanda, E., Lissia, A., Fadda, G., and Zanetti, S. (2004). Mycobacterium avium sub. paratuberculosis in tissue samples of Crohn’s disease patients. New Microbiol. 27 (1), 75–77. Senok, A. C., Ismaeel, A. Y., and Botta, G. A. (2005). Probiotics: Facts and myths. Clin. Microbiol. Infect. 11, 958–966. Shanahan, F. (2001). Probiotics in inflammatory bowel disease. Gut 48(5), 609. Sherman, P. M., Johnson-Henry, K. C., Yeung, H. P., Ngo, P. S. C., Goulet, J., and Tompkins, T. A. (2005). Probiotics reduce Enterohemorrhagic Escherichia coli O157:H7and Enteropathogenic Escherichia coli O127:H6-induced changes in polarized T84 epithelial cell monolayers by reducing bacterial adhesion and cytoskeletal rearrangements. Infect. Immun. 73(8), 5183–5188. Silva, A. M., Barbosa, F. H., Duarte, R., Vieira, L. Q., Arantes, R. M., and Nicoli, J. R. (2004). Effect of Bifidobacterium longum ingestion on experimental salmonellosis in mice. J. Appl. Microbiol. 97(1), 29–37. Smith, L., Orugunty, R. S., and Hillman, J. D. (2006). In ‘‘Research and Applications in Bacteriocins’’ (M. A. Riley, and O. Gillor, eds). Horizon Bioscience, Norfolk, UK. Torii, A., Torii, S., Fujiwara, S., Tanaka, H., Inagaki, N., and Nagai, H. (2007). Lactobacillus acidophilus strain L-92 regulates the production of Th1 cytokine as well as Th2 cytokines. Allergol. Int. 56(3), 293–301. Toure, R., Kheadr, E., Lacroix, C., Maroni, O., and Fliss, I. (2003). Production of antibacterial substances by Bifidobacterial isolates from infant stool active against Listeria moncytogenes. J. Appl. Microbiol. 95(5), 1058–1069. Tsai, C. C., Hsih, H. Y., Chiu, H. H., Lai, Y. Y., Liu, J. H., Yu, B., and Tsen, H. Y. (2005). Antagonistic activity against Salmonella infection in vitro and in vivo for two Lactobacillus strains from swine and poultry. Int. J. Food. Microbiol. 102, 185–194. Venturi, A., Gionchetti, P., Rizzello, F., Johansson, R., Zucconi, E., Brigidi, P., Matteuzzi, D., and Campieri, M. (1999). Impact on the composition of the fecal flora by a new probiotic preparation: Preliminary data on maintenance treatment of patients with ulcerative colitis. Aliment. Pharmacol. Ther. 13, 1103–1108. Wehkamp, J., Harder, J., Wehkamp, K., Wehkamp-von Meissner, B., Schlee, M., Enders, C., Sonnenborn, U., Nuding, S., Bengmark, S., Fellermann, K., Schroder, J. M., and Stange, E. F. (2004). NF-kB- and AP-1-mediated induction of human b-defensin-2 in intestinal epithelial cells by Escherichia coli Nissle 1917: A novel effect of a probiotic bacterium. Infect. Immun. 72(10), 5750–5758. Zhu, W. M., Liu, W., and Wu, D. Q. (2000). Isolation and characterization of a new bacteriocin from Lactobacillus gasseri KT7. J. Appl. Micro. 88, 877–886. Zyrek, A. A., Cichon, C., Helms, S., Enders, C., Sonnenborn, U., and Schmidt, M. A. (2007). Molecular mechanisms underlying the probiotic effects of Escherichia coli Nissle 1917 involve ZO-2 and PKCzeta redistribution resulting in tight junction and epithelial barrier repair. Cell Microbiol. 9(3), 804–816.
CHAPTER
2 Sensory Impacts of Food–Packaging Interactions Susan E. Duncan and Janet B. Webster
Contents
I. II. III. IV. V.
Introduction Consumer Perception Threshold Concept Sensory Effects Methods for Examining Taint and Other Sensory Effects from Packaging VI. Taints A. Taints from contact materials B. Taints from additives or noncontacting materials C. Taints from recycled materials VII. Scalping/Sorption VIII. Protection of Sensory Quality by Food Packaging A. Protection against light B. Preventing moisture loss IX. Using Packaging to Improve Sensory Quality A. Sensory impact of novel antimicrobial ingredients in packaging systems B. Flavor and odor absorbers for improved flavor C. Controlling oxidation through timed release of antioxidants X. Conclusions Acknowledgment References
18 20 21 22 26 27 27 45 46 47 49 50 52 53 53 54 55 56 57 57
Virginia Polytechnic Institute and State University (Virginia Tech), Blacksburg, Virginia, USA Advances in Food and Nutrition Research, Volume 56 ISSN 1043-4526, DOI: 10.1016/S1043-4526(08)00602-5
#
2009 Elsevier Inc. All rights reserved.
17
18
Abstract
Susan E. Duncan and Janet B. Webster
Sensory changes in food products result from intentional or unintentional interactions with packaging materials and from failure of materials to protect product integrity or quality. Resolving sensory issues related to plastic food packaging involves knowledge provided by sensory scientists, materials scientists, packaging manufacturers, food processors, and consumers. Effective communication among scientists and engineers from different disciplines and industries can help scientists understand package–product interactions. Very limited published literature describes sensory perceptions associated with food–package interactions. This article discusses sensory impacts, with emphasis on oxidation reactions, associated with the interaction of food and materials, including taints, scalping, changes in food quality as a function of packaging, and examples of material innovations for smart packaging that can improve sensory quality of foods and beverages. Sensory evaluation is an important tool for improved package selection and development of new materials.
I. INTRODUCTION Packaging materials are critical components within packaging systems for improving product sensory integrity and quality. There is no known packaging material, not even glass or metal, which does not interact in some way with the product it protects; these interactions can affect the quality of both the food and the package. Sensory evaluation is an important tool in the detection of packaging interactions with food. Changes in food or beverage color and appearance, flavor, odor, and textural characteristics as affected by packaging material chemistry, processing, and characteristics can make or break product success. Food and beverage packaging functions to contain, protect, communicate, and provide convenience in use of the product (Robertson, 2006). A good packaging system is designed with an appropriate barrier to protect food from external contamination by microorganisms, foreign materials, chemical contaminants, or environmental degradation (Kilcast, 1996; Robertson, 2006). Printed information on the package provides direct communication pertinent to the product in the form of words and graphics, including colors, to the consumer. Indirect communication, whether intentional or unintentional, results when the human sensory response is triggered by unexpected or uncharacteristic product changes as a result of food–packaging interactions (Duncan, 2007). Such changes may noticeably affect product integrity, quality, and shelf-life, resulting in affective (degree of liking or preference) or analytical (recognition of overall or specific product changes) human responses to the contained product.
Sensory Impacts of Food–Packaging Interactions
19
Flavor, odor, appearance, and texture of foods are not static. Both packaging materials and food products may undergo chemical transformations that alter these food characteristics. Reactions such as oxidation, hydrolysis, and vaporization can lead to the production of off-odors and flavors and changes in texture and visual characteristics (Ayhan et al., 2001). These changes may be caused by the packaging materials themselves, by an interaction between the package and the food, and/or because of poor packaging selection (Huber et al., 2002) and may lead to consumer complaints. New packaging materials or modifications that improve product integrity, quality, and shelf-life are valuable if product improvement is detectable by the consumer or end-user. However, selection of an inappropriate packaging material can result in a negative sensory impression of the contained product by the user, leading to dissatisfaction, decreased use, and communication to the food manufacturer of the product problem. Approximately 50% of all off-odor complaints in one company were due to improper packaging, with major problem sources ranging from degradation of the packaging materials to inadequate selection of packaging material (Frank et al., 2001; Huber et al., 2002). Sweets, cakes, and cookies were the primary foods affected by taints in another food company over a 10-year period, accounting for 50% of the reported cases (Lord, 2003). Food product manufacturers are at risk of experiencing significant losses in production, sales, and consumer confidence as a result of detectable sensory changes from food–package interactions, potentially leading to a damaged brand image (Huber et al., 2002; Kilcast, 1996; Lord, 2003). Clear communications among technical representatives within the packaging supply line are critical (Fig. 2.1). Food manufacturers bear the highest burden in this line of communication as they have the financial risk of loss of production from defective products as well as the loss of product, which cannot be reworked, and packages, which cannot be reused (Huber et al., 2002). The challenge of identifying the source of a sensory problem caused by packaging is expensive and time consuming. Communications along the packaging supply line, therefore, should describe the sensory character of the material in its virgin state and after package conversion, with food manufacturers disclosing processing, packaging, and storage conditions and expected duration of contact with the foodstuff (Duncan, 2007). The packaging supply chain includes the materials or resin manufacturer through the packaging supplier/converter to the food processor/manufacturer, each having a vested interest in material and package/product success. Expensive legal litigation resulting from package/product failure is a potential and serious consequence of poor communication. Modern legislation guards against the spoilage of foods by packaging materials by regulating that food packaging materials do not transfer
20
Susan E. Duncan and Janet B. Webster
Consumers
Retail and food service
Food manufacturers
Complaints! Sensory analyses/ flavor and aroma Pr
Suppliers
Packaging manufacturers
ob lem
s.. . Instrumental analyses/ volatile chemistry
Questions? Polymer synthesis/ manufacturers Client interactions
Communication
Problem resolution
FIGURE 2.1 Clear communications among industry players can help prevent or resolve sensory impacts for food–package interactions.
constituents to foods in sufficient quantities to endanger human health or cause deterioration of the sensory characteristics of foods (Huber et al., 2002; Soderhjelm and Eskelinen, 1985). Understanding the sensory impacts related to food or beverage interactions with packaging materials may be helpful in preventing or resolving problems as well as in selecting appropriate materials for maintaining product integrity, quality, and shelf-life. The source of the change in sensory quality associated with a food–packaging interaction may occur at any stage of the food manufacturing or supply chain and from different sources at each stage. Preventing the problem from occurring is best but, when a problem does occur, good communications and good detective work are needed to determine the cause of consumer or client complaints (Duncan, 2007).
II. CONSUMER PERCEPTION Ultimately, the ability of a packaging material to protect food quality depends not on whether an interaction has occurred but on whether or not a consumer can detect that interaction. The repeat purchase of a packaged food product is contingent on many factors, with enjoyment and positive sensory stimulation being among the most important (Kilcast, 1996). Consumer expectations associated with sensory characteristics and
Sensory Impacts of Food–Packaging Interactions
21
quality of a packaged food can be diminished because of unexpected food– packaging interactions. In addition, the perception of contamination caused by unexpected sensory characteristics also raises concerns about safety, even if there is little or no health risk (Anonymous, 1988; Dietrich et al., 2005). However, only a small proportion of consumers purchasing a product exhibiting a food–packaging interaction may call to comment or complain (AFGC, 2007). Therefore, getting sensory observations and qualitative descriptive information from these contacts is very important. Consumer communication of a sensory problem in a packaged food system may result in a variety of descriptions from different people experiencing the same problem, suggesting that these reports are unreliable. The lack of verbal skills or training in analytical descriptive methods, unfamiliarity with chemical species that may be causing the problem, and bias associated with the conditions under which the sensation was experienced make consumer descriptions of sensory changes difficult to act upon (Kilcast, 1996). It is possible that many sensitive people may experience the sensory problem but only a few may report it. This may be why some incidences of sensory problems associated with food–package interactions seem sporadic, limiting perception of the problem scope and making detection and resolution of the problem challenging (AFGC, 2007). Individuals also perceive flavors at differing concentrations, or threshold levels, and these levels can vary by a factor of a billion from one individual to another. A report of a sensory problem by even one consumer of known high sensitivity may be reason enough to determine the cause (Huber et al., 2002; Kilcast, 1996). Consumers are very important sentinels of quality changes and their comments or complaints may represent the beginning of a major problem.
III. THRESHOLD CONCEPT The point at which the interaction between the food and package material causes a change in the sensory response of an individual is an important threshold. At this point, the observer is aware that there is something different about the product. Typically, thresholds are discussed in relation to a given chemical compound that contributes a specific odor, flavor, appearance, or textural change. Thresholds for a given compound vary with the medium (food or drink composition) in which it is present, the temperature at which the product is presented, other stimuli contributing to the sensory character of the product, the methodology used in determining the threshold, and individual sensitivity to the stimulus (Kilcast, 1996; Land, 1989; Meilgaard et al., 2007).
22
Susan E. Duncan and Janet B. Webster
If a high proportion of the population is sensitive to the sensory change and the change is negative in character, there is a high potential for consumer complaints about the product. A sensory threshold is the lowest concentration of a compound detectable by a certain proportion, usually 50%, of a given population, indicating that the stimulus is at a level sufficient to create a sensory perception (AFGC, 2007; Kilcast, 1996; Lawless and Heymann, 1998; Meilgaard et al., 2007). Different types of thresholds are recognized. The detection (or absolute) threshold is the lowest physical intensity at which a stimulus is perceptible (Kilcast, 1996; Lawless and Heymann, 1998; Meilgaard et al., 2007); the recognition threshold is the physical intensity at which a stimulus is correctly identified (Kilcast, 1996; Meilgaard et al., 2007). The range in sensitivity of individuals within a population to a given chemical stimulus is typically 2000-fold (AFGC, 2007).
IV. SENSORY EFFECTS Effects of materials on sensory characteristics and quality of foods and beverages can occur from direct contact, as with a primary package intended for containment, or by indirect means resulting from the environmental conditions (relative humidity, temperature, and air quality) as well as the characteristics of secondary packaging materials. Any material in contact or proximity to a foodstuff may have an effect. This extends to the materials used for water distribution systems, plumbing, gaskets, adhesives, valves, shipping containers, pallets, and other materials, all of which may contribute to perceivable changes in sensory characteristics. There is a large body of research on the effects of food packaging on flavor and odor chemistry of various foods. However, the focus of such research is often targeted toward identifying the chemicals and relative concentrations that occur as a result of food–package interactions. The relationship between food–package interactions and human sensory response has received relatively little scientific attention. There are numerous books, review papers, and research publications describing chemical interactions of packaging materials with foods that affect flavor chemistry. Most explain the interactions as a function of changes in analytical flavor chemistry, focusing on migration of substances from the package into the food, absorption of flavor components from the food into the package, and transfer of odors, gases, and light through the polymer package, which then can affect food quality (Ahvenainen, 2000; Barnes et al., 2007; Katan, 1996; Leland, 1997; Piringer and Ruter, 2000; Risch and Ho, 2000). In many research studies, the sensory impact of these reactions is not directly described or is not evaluated, although there may be reference to the impact. Relying only on
Sensory Impacts of Food–Packaging Interactions
23
analytical evaluation of flavor chemistry can be deceiving because these compounds may or may not have an impact on the sensory quality of the food (Piringer and Ruter, 2000; Torri et al., 2008). Sensory evaluation can help characterize flavors, aromas, appearance effects, and textural changes that might not be able to be determined analytically. Sensory perception is often more sensitive than analytical methods and in some cases is the only way to determine if a change has occurred in a food due to an interaction with its packaging. Human senses can detect changes in volatile chemistry at extremely low concentrations (106–1012 mol/ml) for some chemicals (Kilcast, 1996; Lawless and Heymann, 1998; Meilgaard et al., 2007; Reineccius, 2006). However, there are problems with sensory testing that must be kept in mind. These problems include differing perceptions and thresholds for changes in appearance, texture, flavor, and aroma due to the unique physiological and psychological makeup of individual human subjects. The range of human sensitivity, verbal skills for description of the sensation(s), and affective response to the sensation contribute to the challenges associated with identifying the source of a sensory problem in a packaged food system. The combination of sensory evaluation with analytical approaches is required to identify perceptible changes and to identify the potential chemical changes that may be causing the sensory effect. Low concentrations of compounds responsible for changes in food characteristics may not be detectable by even the most sensitive analytical methods but, in combination with appropriately applied sensory methods, the clues provided by both techniques may help identify the problem, provide indications to the cause, and suggest clues for the source of the problem. Sensory issues relating to food and packaging interactions may be classified based on the sensory quality changes that occur. Four broad categories that infer the direction (positive/negative) of the quality change include: (1) taints resulting from the packaging material or as a functional limitation of the package; (2) scalping of food constituents; (3) packaging function for protection of sensory properties of the food; and (4) improving flavor and odor quality through food–package interaction (Fig. 2.2). Taints are defined as a taste or odor foreign to the product (AFGC, 2007; Kilcast, 1996). Taints typically are unpleasant in character and are initiated by or originate from sources external to the food; they are caused by constituents of the packaging material or the near environment migrating into the product. Common sources of taints related to food packaging include packaging materials, inks and dyes, adhesives, and secondary packaging including pallets, shipping containers, and corrugated cardboard materials. Although taints are contaminants of the food
24
Susan E. Duncan and Janet B. Webster
Taint
Scalping
Light
Moisture
Gases/ volatiles
Food flavorants Colorant
Water
Material molecules
Fat
FIGURE 2.2 Negative sensory impacts occur when tainting or scalping occur or ineffective packaging selections are made.
environment, not all contaminants cause taint (AFGC, 2007). To meet legal and consumer expectations, the goal should be zero levels of tainting species (below sensory threshold) so even the most sensitive members of the population cannot detect the taint (Kilcast, 1996). The concentration of the penetrating molecules may be very low and still create a significant sensory response. Tainting molecules in the headspace of a package will influence the sensory response by the consumer when the package is initially opened and perhaps motivate a complaint action (Kilcast, 1996). The threshold for detection in the air above the sample is typically lower than needed to cause a response in the food medium. The majority of published literature associated with sensory issues of food–package interactions is related to taints and flavor scalping. Scalping (sorption) is also detrimental to consumer perception of food quality. From a sensory perspective, this type of interaction is different from taints. This phenomenon is associated with key food constituents absorbing into a plastic or other material (Brody, 2002). In this situation, desirable food constituents, such as aroma compounds, acids, lipids, and pigments, are removed from the food or beverage by the packaging materials. This is most commonly described in relation to a decrease in flavor intensity or as an alteration of the flavor profile but could also be associated with color or odor changes. Changes or decreases in quality may be perceived by consumers. The exhibition of reduced sensory quality in many shelf-stable foods in plastic packaging is attributed to scalping (Brody, 2002).
Sensory Impacts of Food–Packaging Interactions
25
Off-flavors and off-odors, in contrast to tainting or scalping, are related to changes in food chemistry associated with degradative reactions within the food or deteriorative changes of food components (AFGC, 2007; Kilcast, 1996). The character of these changes is typically unpleasant and undesirable in relation to food quality and consumer satisfaction. In addition, changes in pigment chemistry, moisture loss (or gain), or other reactions within the food system may result as a function of improper material selection. Sensory studies on the protective function of packaging (maintaining/improving) on quality primarily exist as supporting information to the assessment of chemistry, microbiology, processing, or engineering of the food–package system. In many cases, these sensory studies are designed to determine if a difference exists because of the experimental variables but detailed, descriptive studies of sensory impacts of the protective function of packaging on food properties is limited. Novel packaging approaches using active materials or new packaging technologies for a desired sensory effect within the food system, or those that enhance product sensory quality and improve shelf-life, do not fit into any of the prior categories. This area of literature is emerging. The value in understanding the differences in these sources of subtle changes in sensory quality is associated with distinguishing the cause and possible solutions associated with the problem. There is a wide array of potential contaminants associated with taints and identifying the cause as something originating from external contamination or from internal product change helps narrow the search (Kilcast, 1996). Understanding sensory impacts from food–package interactions will provide greater capacity for improving food–package systems. For example, a pineyspruce taint was observed in a routine quality assurance examination of a packaged ready-to-eat breakfast cereal (Heydanek, 1977). With a careful analytical approach that utilized sensory evaluation in the early stages, the source was traced to the resin bonding paper layers in the glassine liner of the package. With the use of gas chromatography–mass spectroscopy (GC–MS), in combination with additional sensory testing of additional glassine materials, the migrant molecules were identified as major terpenes, fenchyl alcohol, and borneol. The purpose of this review is to give an overview of the effect that food packaging has on the sensory quality of food. The focus will be on human response and sensory terminology used in describing these interactions in relation to foods and beverages in contact with different materials. Volatile chemistry will be described only as it relates to the sensory effects. It is not the intent of this review to provide detailed information on sensory evaluation methodology. There are numerous text and reference books for developing a basic knowledge of sensory evaluation methods. However, a brief description of nontraditional sensory methods for examining taint and other sensory effects from packaging is provided.
26
Susan E. Duncan and Janet B. Webster
V. METHODS FOR EXAMINING TAINT AND OTHER SENSORY EFFECTS FROM PACKAGING The selection of an appropriate sensory methodology for examining taints, scalping, off-flavors, or improved product quality is dependent on the project objectives. No single method can answer all questions, such as describing the flavor character and intensity as well as consumer response, so effective communication of the project goal and objectives among the sensory specialist, packaging expert, and project leader is needed (Duncan, 2007). Appropriate selection and training of participants in descriptive evaluation is needed to provide accurate and reliable qualitative and quantitative responses. Descriptive methods, which provide much more information than discrimination or affective methods, are helpful in differentiating how a product–package interaction affects sensory profiles. However, this is a very time intensive effort because of the investment in panel preparation and maintenance. Discrimination testing can be very effective in determining if a difference occurs because of the product–package interaction but these methods do not describe how (change in sensory profile) the product–package interaction affects perception. Such methods require little or no training of panelists but more panelists are needed to increase the power of the test. Untrained panelists representative of the targeted consumer population should be used for estimating affective consumer responses to taints and off-characteristics as well as verifying value-added quality maintenance or improvements of packaged food products. Since a wide range of individual responses can occur in consumer testing, a large number of consumers are needed to verify if the product–package interaction is truly having an impact. For all methods, attention to appropriate environmental, sample, and panelist controls are needed during preparation and presentation. Communication with panelists must be appropriate for instructing panelists in completing the tasks but restricted in details of samples and objectives in order to avoid biases that may affect outcomes or interpretation of the results. Standard methods that apply to detecting taint from plastic and paperboard materials are described in US ASTM Std E 462-84 (tests for odor and taint transfer from packaging film), ASTM E 619-84 (examination of odors from paper packaging), and German Standard DIN 10 955 (testing of taints transferred by direct contact and also by vapor phase transfer) (Kilcast, 1996). Many of these methods are based on accelerated storage conditions and using food simulants or water or on evaluation of volatile compounds emitted from packaging material (Kim-Kang, 1990; Torri et al., 2008). Sensory methodology books (Lawless and Heymann, 1998; Meilgaard et al., 2007) provide guidance for appropriate design and
Sensory Impacts of Food–Packaging Interactions
27
application of sensory methods. Book chapters with specific reference to packaging (Kilcast, 1996, 2003; Lord, 2003) provide some background and general sensory information in the context of the sensory effects of food packaging. Recent German references assess difficulties in standardizing methods for sensory analysis of packaging materials and quality requirements for colorants and additives used in food and beverage packaging (Anonymous, 2007; Buettner et al., 2007).
VI. TAINTS Although there are many routes by which taints may occur in foods, one of the greatest risks is from contact of foods with materials that may contain potential migrants (Baigrie, 2003; Kilcast, 1996; Reineccius, 2006). Many materials have molecules that can migrate from the package into the food product. In addition, polymer materials, such as gaskets used in manufacturing processes, containers used for transport, or packaging materials inappropriately used in a process (i.e., irradiation or hot-fill) may cause taints. Water distribution and plumbing materials, flooring materials, and disinfectants or other environmental contaminants on food packaging surfaces also may be sources of tainting molecules. Long-shelf life foods, which are stored in direct contact with packaging materials for a long period of time, are at great risk of developing taint. Beverages, extended shelf-life (ESL) or shelf-stable milk products, and other liquids have a high risk for tainting chemical species to transfer into the food matrix. Water and many ESL dairy products have a low flavor profile so even a low concentration of migrating molecules from the polymer may impact sensory characteristics in these products. A higher fat content within the food product, such as chocolate, contained in the packaging material can cause an increased taint within the food product. Levels of tainting odors and flavors are highest where there is direct contact between the packaging material and foods with a high fat content on the surface. Sensory descriptors associated with tainting chemicals from packaging sources have been reported (Ewender et al., 1995; Lord, 2003).
A. Taints from contact materials Materials may exhibit a tainting odor from the polymers, monomers, additives, adhesives, or process. Packaging related taints, as reported from the Nestle’s Central Packaging Laboratory over a 4-year period (1996–2000) were related to solvents (28%), degraded polyethylene (PE) (24%), styrene (15%), halogenated phenols (15%), degraded paper (3%),
28
Susan E. Duncan and Janet B. Webster
and other unknown sources (15%) (Huber et al., 2002). The virgin material should not exhibit an odor; however, that will not guarantee that the material will not impact the sensory profile of the food. Many materials used for food and beverage packaging have characteristic odors or sensory active compounds (Torri et al., 2008). The intensity and description of the odor may be affected by the number and type of volatile compounds that are released under environmental conditions at the time of evaluation. Chemical composition of the material and polymer morphology may play a role in the sensory characterization. Sensory descriptors do not define a specific chemical compound but may be related to different compounds, a blend of compounds, and even a limited concentration range of a compound or class of compounds. For example, trans-2-nonenal in water changes in sensory (taste) description from ‘‘plastic’’ (0.2 mg/l) to ‘‘woody’’ (0.4–2.0 mg/l), ‘‘fatty’’ (8–40 mg/l), and ‘‘cucumber’’ (1000 mg/l) (Piringer and Ruter, 2000). Such terms are descriptive of the sensation and perception by human response to the chemical stimuli (Table 2.1). Sensory active components from packaging may influence the perception of product quality (Piringer and Ruter, 2000). Parameters that may contribute to sensory influence include (Granzer et al., 1986): Concentration of the component in the packaging material. Solubility of the component in the packaging material (partition gas
phase/packaging material). Solubility of the component in the food (partition gas phase/food). Sensory threshold level of the component. Type and intensity of the food aroma. Diffusion rate of the component in the packaging material. Diffusion rate of the component in food. Time and temperature of storage. Ratio of the amount of packaging material to the amount of food.
Consideration of these parameters is important when tracking the source of taints and interpreting the relationship between analytical chemistry and sensory impact of the food–package interaction. Evaluating odor and flavor taints is frequently done with water, fatty food simulants (oil, chocolate, unsalted butter), hydrophilic powders (sugar, cornflour), or combined hydrophilic–hydrophobic matrices (milk or cream, biscuits) (Kilcast, 2003). The Robinson test often is used to evaluate materials for tainting potential. This test places the test material in a sealed container separated from the food simulant or test food at a relative humidity between 53% and 75%. After about 48 h, the test food is evaluated for taint compared to a control, using a discrimination method (Lord, 2003). Chocolate is frequently used as the food simulant for this test. Intensity of the taint may be evaluated using a
Sensory Impacts of Food–Packaging Interactions
29
TABLE 2.1 Sensory descriptors associated with taint sources (Baner, 2000; Caul, 1961; Huber et al., 2002; Piringer and Ruter, 2000; Torri et al., 2008) Taint sources
Material
Sensory descriptors
Packaging materialsa
Plastics
Acetic acid, acrid, alcohol, adhesive, burnt, burnt wax, candle-like, cat urine, chemical, lube oil, musty, oxidized polyethylene, paint, paraffins, paste, phenolic, plastic, pungent, rancid, soapy, solvent, stale, styrene, stuffy, vinyl, waxy Almonds, cardboard, fruity, green grassy, pine, rancid Jute sack, petrol Aromatic, camphorated, chemical, fruity, solvent, sweet, toluene Fat, linseed oil, mineral oil, painty, petrol, rancid, varnish Antiseptic, disinfectant, herbicide, hospital, insecticide, medical, metallic, phenolic Moldy, mushrooms, musty Catty, cork, naphthalene, wooden pencil
Paper/board
Printing/ convertingb
Miscellaneous Solvent based
Offset printing
External contaminationc
Chemical
Microbiological Miscellaneous a b c
Material-intrinsic odors Solvent based includes flexo and gravure printing Absorption of foreign odor from other materials in contact with packaging during production and storage as well as food–package interactions
5-point rating scale (0 ¼ not perceptible; 4 ¼ strong). The use of the Robinson-style test method, using the rating scale, is effective for quality control of materials intended for food packaging. Understanding the sensory influence of raw and processed materials on food simulants and complex matrices is helpful in reducing taints in packaged foods. Table 2.2 provides a summary of sensory descriptors associated with foods and beverages in contact with packaging materials.
TABLE 2.2
Sensory descriptors associated with some packaging materials containing different foods and beverages
Broad category
Packaging material
General
Product
Sensory description
References
Glass
Milk
Karatapanis et al. (2006)
Metal
Beer Canned pork products Cakes, biscuits, chocolate confectionary Chocolate-coated cakes
Plastic, oxidized, burnt flavor Musty flavor Catty flavor Off-flavor
Paper
Description
Board stock
Cocoa beans
Mushroom odor Musty, mouldy Halogen-like odor Papery, cardboardy Burnt Piney flavor Stale, fruity flavor Painty, chalky, fruity, rancid, sulfide, and resinous odor Mouldy odor and disinfectant flavor Mouldy flavor
Butter
Refrigerator/stale
Anonymous (1988), Whitfield et al. (1984) Lozano et al. (2007)
Waxy, oily, rancid Spoiled protein (sulphide) Resinous
Caul (1961) Caul (1961) Caul (1961)
Breakfast cereal Milk
Packing sacks
Adhesives
Waxed parchment paper Coatings
Off-flavor
Council AFaG (1997) Kim-Kang (1990) Goldenberg and Matheson (1975) Goldenberg and Matheson (1975)
Cocoa powder
Caul (1961) Caul (1961) Caul (1961) Anonymous (1988) Heydanek (1977) Karatapanis et al. (2006) Anonymous (1988)
Whitfield et al. (1984)
Off-flavor
Goldenberg and Matheson (1975)
Retort pouch
Cakes, biscuits, chocolate confectionary, sugar confectionary Ham products
Cat urine odor
Piringer and Ruter (2000)
Retort pouch
Fruit-flavored soft drinks
Off-flavor
Passy (1983)
Liner
Ready-to-eat breakfast cereals Candy wraps
Pine or spruce-like odor
Heydanek (1977) Kim-Kang (1990)
Water
Bitter, burnt, old rubberlike flavor Stearic, paint, bitter, hay, rancid odor, and flavor grainy Jasmine-like, herbal and floral ‘‘Lube’’ oil, burnt, phenolic Musty odor Candle-grease, musty, rancid, soapy, pungent, acrid, sickly, astringent, synthetic, metallic, and dry flavor Musty
Milk
Stale, fruity flavor
Corn chips (snack foods) Water
Plastic odor Tastes: sweet, metallic, stony, pungent, dusty,
Printing films
Plastics
Polyamide/ionomer laminate Polyester, aluminum foil, polyethylene laminate Glassine
Polypropylene
Oats
Boiled sweets
Orange juice Polyolefins
Low density polyethylene (LDPE)
High density polyethylene (HDPE)
Larsen et al. (2005) Heydanek (1978) Lord (2003) Caul (1961) Feigenbaum et al. (1998) Linssen and Roozen (1994)
Linssen and Roozen (1994) Karatapanis et al. (2006), Moyssiadi et al. (2004) Sander et al. (2005) Villberg et al. (1997)
(continued)
TABLE 2.2 (continued) Broad category
Packaging material
Description
Product
HDPE water pipes
Water
Polyethylene (PE)
Wine Coffee Wine Prawns and ocean fish
PE-coated paperboard Cross-linked PE Water pipes
Milk, water, fruit juices Water
Sensory description stale, plastic, foul, stink bug, candle grease Odors: Sweet, chemical, stale, dirty, foul Earthy-musty flavor Waxy, plastic, citrus
Pungent, musty odors Rio, medicinal, phenolic, or iodine-like flavor Musty cork flavor Iodoform taint Oxidized oil, waxy, rubbery Candle-like, stale, stuffy, musty, soapy, rancid odor Plastic flavor Wax-like odor Alcohol, sweet chemical, plastic, bitter, mechanical, glue, burning, spicy, fruity, almond,rotten swampy, burning plastic pipe
References
Council AFaG (1997) Heim and Dietrich (2007b), Dietrich (2007) Council AFaG (1997) Council AFaG (1997) Council AFaG (1997) Caul (1961) Piringer and Ruter (2000)
Leong et al. (1992) Piringer and Ruter (2000) Durand and Dietrich (2007)
PE þ rubber net
Water
Polyethylene terephthalate (PETE) Milk
PETE þ rubber net Polystyrene
Oat products
Epoxy
Coffee creamer and condensed milk Chocolate and lemon cream cookies Fruit drinks Orange and lemon drinks Packed cheese (with PE/ PETE lid) Drinking water
Cellophane Cellulose film
Sandwiches
PVC
Other
Colas Water
Environmental odorants
Yellowish color, plastic taste/odor Papery, scorched cloth
Kontominas et al. (2006)
Plastic, oxidized, burnt flavor Turpentine-like odor Yellowish or opaque color, taste/odor Insecticide or plastic flavor Woody, sweet Plastic-like chemical odor and flavor Astringent chemical plastic flavor Off-flavor and odor
Karatapanis et al. (2006), Moyssiadi et al. (2004) Kim-Kang (1990) Kontominas et al. (2006)
Catty flavor Off-flavor
Kim-Kang (1990) Kim-Kang (1990)
Pungent, chemical pine odor and flavor Plastic, glue, putty, adhesive, chemical, musty Sweet, woody, rubbery Off-flavor
Lord (2003)
Caul (1961)
Heydanek (1978) Caul (1961) Heydanek (1978) Baner (2000) Passy (1983)
Floral
Heim and Dietrich (2007a), Dietrich (2007) Caul (1961) Goldenberg and Matheson (1975) Caul (1961)
Woody
Caul (1961)
(continued)
TABLE 2.2 Broad category
(continued) Packaging material
Description
Product
Printing Inks
Oils
Rubber hydrochloride (pilofilm) Vinyl
Saran Vinyl chloride
Glued seams
Cocoa powder
Mixture
Maple syrup Granular gelatin
Sensory description
References
Musty, cardboardy, burnt, floral, painty, chalky, fruity, rancid, sulfide, resinous, woody odors Off-odor Musty odor Musty, cardboardy, burnt, painty, chalky, fruity, rancid, sulfide, and resinous odors Oily, fatty, buttery Varnishy, painty Inky, rancid Garbagey Sour milk (Casein)
Caul (1961), Anonymous (1988)
Halogen-like odor Aromatic sweet, chlorine, oxidized oil, or solventy Alcohol, soapy Moldy odor and disinfectant flavor Off-odor Fish odor
Caul (1961) Caul (1961)
Caul (1961) Anonymous (1988) Anonymous (1988)
Caul (1961) Caul (1961) Caul (1961) Caul (1961) Caul (1961)
Caul (1961) Anonymous (1988) Kim-Kang (1990)
Sensory Impacts of Food–Packaging Interactions
35
1. Polystyrene (PS) Residual styrene monomer from polystyrene production has been associated with tainting problems in different food products (Baner, 2000; Heydanek, 1978; Huber et al., 2002; Piringer and Ruter, 2000). Sensory descriptors for styrene monomer include chemical, insecticide, and plastic (Baner, 2000; Caul, 1961; Heydanek, 1978). Thresholds for styrene monomer are very low in water (taste threshold: 0.022–0.37 mg/kg) and air (odor threshold: 0.050 mg/kg). Thresholds in complex foods and beverages range from 0.2 to 0.3 mg/kg for orange fruit juice drink, a 3% oil-in-water emulsion, and skim milk (0% fat), 1–3 mg/kg for whole milk, oil-in-water emulsions of 15–30% oil, cocoa powder (10–20% fat), and greater than 3 mg/kg for condensed milk (10% fat), butter, and cream (33% fat) (Baner, 2000). PS starts to decompose at very low levels after several hours and at temperatures greater than 240 C. Ethylbenzene, which is commonly used to dilute solvents during PS polymerization, is another source of these taints (Baner, 2000). Styrene monomer concentration in foods packaged in 31 different PScontaining food packages and contact materials averaged 224 mg/kg with two products having concentrations between 800 and 1500 mg/kg, well above the sensory threshold limits (Baner, 2000). Strict specifications for styrene monomers as well as for residual solvents, toluene, and odor and taint transfer for supplier materials should be set (Huber et al., 2002). Polystyrene used for plastic cups has been a source of off-flavors (Huber et al., 2002). Oat products stored in polystyrene containers developed an ‘‘insecticide or plastic’’ off-flavor after 6 weeks of storage (Heydanek, 1978). Corn products stored under the same conditions did not develop these taints. Compounds with odors similar to the off-flavor in oats were identified by GC–MS and gas chromatography–olfactometry (GC–O). A similar pattern of volatile peaks were observed in both the PS package and the oat products. The taint was identified as being related to the styrene monomer, which was probably present in the PS feedstock. The level of styrene residual was much greater in oats (146 mg/kg), at 20 times the styrene odor threshold in air (0.73 mg/kg), than in corn (1.5 mg/kg). Styrene levels in corn were at twice the odor threshold but were not high enough to cause a taint in the corn product. It is possible the higher fat content of the oats, 11% as compared to 2% for the corn, increased product affinity for the styrene monomer. Coffee creamers and condensed milk packaged in thermoformed polystyrene single serve (5–10 g product) portion pack containers have demonstrated styrene taint problems (Baner, 2000). These products are typically packaged at ultra-high temperatures (UHT) and, if packaged aseptically, may be stored without refrigeration. The higher storage temperature increases the potential for styrene taint in the product and substantially decreases the potential product shelf-life because of
36
Susan E. Duncan and Janet B. Webster
tainting issues. The use of laminate materials, placing PE and ethylene vinyl alcohol (EVOH) on the food contact surface, reduces the migration of styrene monomers from PS to the food. However, even with these laminate barrier materials there still have been styrene taints resulting from the styrene monomer migrating from the PS layer to the inner PE layer when the material was shipped and stored in roll form prior to forming (Baner, 2000). The estimated styrene concentration migrating from the PS into creamers was 23–31 mg/kg, far exceeding the sensory threshold of 0.1–3 mg/kg in cream and potentially creating a sensory impact when the cream is diluted into a food or beverage, such as tea or coffee. The large ratio of package surface to cream volume in these single serve portion packs contributes to this high concentration. The high fat content of cream also increases the transfer rate whereas nonfat milk would have much lower, probably undetectable, styrene taint under the same conditions (Baner, 2000). However, other research suggests that milkfat masks low levels of taint from other materials in whole milk compared to nonfat milk (Leong et al., 1992; van Aardt et al., 2001a). Storage temperature and changes in product sensory profile with storage also affect the perception of taint from styrene monomer and other tainting compounds. Styrene monomer migration was suggested as a possible contributor to flavor and odor changes of butter packaged in waxed parchment paper during refrigerated storage (Lozano et al., 2007). Butter wrapped in common commercial wrapping foil or wax parchment paper demonstrated different flavor profiles at 6 and 12 months of storage at refrigeration (4 C) and freezer (20 C) temperatures (Lozano et al., 2007). A trained sensory panel (n ¼ 8) evaluated intensity (0 (absence); 15 (high)) of selected sensory characteristics (cooked/nutty, milkfat/ lactone, refrigerator/stale, and salty taste) representing characteristics of fresh and stored butters. Butters stored for 6 months at refrigeration temperature in waxed parchment paper had detectable ‘‘refrigerator/ storage’’ flavor (intensity of 1.0–1.1) compared to fresh and foil-wrapped butters but products in both packaging had lower intensities of positive flavor attributes (‘‘cooked/nutty,’’ ‘‘milkfat/lactone’’). At 12 months, mean intensity values for both parchment and foil wrapped refrigerated products were between 1.6 and 1.9, indicating that this flavor problem had increased and the intensity of ‘‘cooked/nutty’’ notes continued to decrease. Frozen products wrapped in parchment exhibited a very low (0.5) intensity of refrigerator/stale flavor after 12 months of storage as compared to no perception of this characteristic in foil-wrapped butter. Frozen storage helped to maintain levels of milkfat/lactone notes and, while cooked/nutty notes decreased slightly more in parchmentwrapped butter than in foil-wrapped, this characteristic still was observed at higher levels in frozen products than in refrigerated products at 12 months.
Sensory Impacts of Food–Packaging Interactions
37
Styrene monomer migration from the package stored at refrigeration temperatures was detected by GCO but frozen storage can decrease the migration rate (Lozano et al., 2007). Odor threshold levels of styrene, ethylbenzene, and toluene in oil were reported as 3.4, 4.1, and 94.7 mg/kg, respectively, and were higher than the level of these compounds found in all butter samples. Lozano et al. (2007) suggested the additive effect of multiple benzene derivative compounds at subthreshold levels, in combination with declining fresh flavor compounds, altered the perception of the refrigerator/stale flavor intensity as the parchmentwrapped product aged. The authors also suggested that differences in matrix between oil (100% lipid) and butter (80% lipid, 20% water) must be considered when relating thresholds from oil to butter. They hypothesized that styrene monomer thresholds in butter would be lower than those reported in oil. This is an important consideration when applying threshold data of compounds in food simulants to the sensory detection of these compounds in complex food products. Although the common method for assessing thresholds is a 3-sample alternative forced choice method with increasing concentrations, there are some product characteristics that make presentation of more than one sample at a time unsuitable (Lawless and Heymann, 1998). Linssen et al. (1991) evaluated PS packaging material taint in chocolate ingredients (chocolate flakes (15% fat), cocoa power (10% and 20% fat), intended for beverages. Chocolate ingredients were mixed with or without (control) PS sheet pieces (0.5 dm2, 2.0 dm2), and stored in glass jars for 7 days at 30 C. Drinks were prepared from chocolate ingredients, water, and sugar for sensory testing. Untrained panelists (n ¼ 48–50) evaluated the drinks compared to a standard (no PS) for a difference comparison as well as for recognition of styrene. A signal detection discrimination method was selected because product stickiness and lingering taste character limited the number of samples that could be presented simultaneously. Using the signal detection method, panelists responded to each sample with one of four category choices (Linssen et al., 1991). The categories for the difference test and the styrene recognition test were ‘‘(same as the) standard, perhaps standard, perhaps not the standard, not standard’’ and ‘‘styrene recognized, perhaps styrene recognized, perhaps styrene not recognized, styrene not recognized.’’ R-indices, used to express the results of signal detection over the characteristic background, represent the probability of correctly distinguishing (chance ¼ 50%; 100% indicates perfect discrimination) between products or correct recognition of styrene by the panelists. There was no difference in taste, based on R-indices close to 50%, between test samples and the standard for cocoa drinks and chocolate flakes in contact with low styrene exposure (0.5 dm2). Higher styrene exposure (2.0 dm2) in contact with chocolate flakes increased the probability of identifying the difference, with R-indices of 64% and 72%
38
Susan E. Duncan and Janet B. Webster
for the plain and milk chocolate flakes, respectively. Milk chocolate has a lower bitter taste intensity than plain chocolate and is more susceptible to styrene taint. Comparisons to a 200 ppb styrene standard in water increased the probability of identifying the styrene taint in chocolate flakes (R-indices of 62–88%). Providing a recognition standard increases the probability of discriminating the difference. Increasing the contact area and polymer thickness increased the migration concentration of styrene and increased the chance of taint. Because the potential for styrene taint to occur is high, any packaging materials containing PS should be evaluated under accelerated migration testing conditions with the intended product (Baner, 2000). The product should be tested using a sensory discrimination test, such as a triangle test, in comparison with a reference product stored under the same conditions but not in contact with the packaging. Accelerated testing conditions must consider food product quality to avoid significant chemical or microbiological spoilage.
2. Polyamides (PA) Good oxygen barrier properties of a material do not guarantee a good flavor barrier. Polyamides have good oxygen barrier properties but do not provide a good flavor barrier because of their hydrophilic properties (Brody, 2002). Two ham products, cooked and packaged in PA/ionomer laminate films from different film manufacturers, were identified, by consumer complaints, as smelling like cat urine (Piringer and Ruter, 2000). The PA source was different for each manufacturer but the ionomer was traced to one source. Linkage to a specific printing ink on the film was evident and the source was traced to diacetone alcohol (DAA), a chemical precursor of mesityloxide. The cat urine aroma resulted from a reaction by sulfur-containing proteins in the ham that had migrated into the ionomer film with DAA, converting it to mesityloxide (Piringer and Ruter, 2000). Printing inks must be free of mesityloxide and its precursors when used on ionomer-containing laminates or films intended for packaging foods with sulfur-containing proteins (Piringer and Ruter, 2000).
3. Polyolefins This class of materials includes the family of plastics based on ethylene and propylene (Robertson, 2006). Low, linear, and high density PE and polypropylene (PP) materials are common food packaging materials. The use of polyolefins, such as PE and PE terephthalate (PETE) in contact with foods and beverages is common. However, contact with foods, especially under conditions of heat or long duration, can potentially impact sensory characteristics of the contained product. Packaging should be carefully selected, especially for applications that involve heat treatment at high temperatures while in contact with foods.
Sensory Impacts of Food–Packaging Interactions
39
Residual solvents from processing of polyolefins and other materials can affect sensory quality of packaged foods. Entrapment of solvents, inks, or other tainting chemicals within layers of newly processed material on a reel increases the risk of taint into food. Frank et al. (2001) reported that chocolate wrappers (paper/aluminum low density PE (LDPE)/Surlynj) with a high odor score (3.0; 4.0 ¼ very strong difference from reference), based on a trained descriptive panel, demonstrated decreasing odor intensity, from the initial odor score 1.3 when aerated for up to 48 h. However, the score (2.6) at 24 h was still above the maximum odor score for acceptable use (2.5 out of 4) as food packaging. Packaging materials are rolled on large reels for shipping and, if not appropriately aerated prior to use, solvents remaining on the inner layers of the roll can cause taint in wrapped chocolate and other foods (Baner, 2000; Frank et al., 2001). Aeration (gassing-off ) of unrolled materials enables the packaging supplier to reduce the odor through gassing off of solvents, thus, improving the product quality. Food manufacturers should inquire if adequate aeration time was provided prior to package conversion and require such specifications be met. The intended use of materials in contact with foods can be a critical parameter in determining material selection. Heating of materials in contact with foods can increase the risk of taint in some circumstances. Plastic nettings, made of plastic-based threads from PE and PETE, are often used for containing raw meats and vegetables. Vegetables, such as potatoes or carrots, are typically washed and often peeled after removal from the package before raw consumption or cooking, thus, reducing the risk of taints. However, some meat products are cooked in these netting materials at temperatures exceeding 100 C for an hour or more. Migration of molecules from these packaging materials into fatty and aqueous-based food simulants was detected at ranges below and above the European Union established migration limits (60 mg/l) (Kontominas et al., 2006). The impact of migrating molecules was measured by heating different PE or PETE net or threads in potable water for 4 h at 100 C. A panel of five judges evaluated the taste, odor, and color of the aqueous simulant based on a 5-point intensity scale (0 ¼ no difference between experimental and control sample (water); 4 ¼ very large difference). A preset designation established that a score greater than 1 indicated an unacceptable sensory score and evidence of sensory impact from contact of the material with the aqueous simulant. Only 7 of the 15 materials tested received a score lower than or equal to 1, the sensory acceptability limit. Sensory characteristics that developed in most water samples included a yellowish discoloration or opaque appearance and off-odor and off-taste described as plastic at slight to moderate intensities. Netting materials (PE or PETE) containing rubber and rubber plus cotton resulted in higher sensory impacting characteristics compared to
40
Susan E. Duncan and Janet B. Webster
materials consisting of PE or PETE only. Only one sample of PETE thread (no rubber or cotton) produced objectionable plastic taste (score ¼ 2) and slight odor as well as a moderate difference in color (yellowish, dispersion development) compared to the water control sample. There was no direct relationship between sensory characteristics found in the water migration behavior of molecules in the various netting materials tested (Kontominas et al., 2006). Some samples that exhibited acceptable levels of migration had a moderate or greater intensity of offflavor, off-odor, or color impact. Not all materials with high migration levels demonstrated sensory impacts in the water. There were samples that had values below the upper limit for migration as well as no real change (less than or equal to 1) in sensory characteristics compared to the control water sample. Oxidation products of plastics materials may be responsible for off-odor development in heated plastic materials. Relevant sensory compounds are not the alkanes and alkenes, which have high sensory thresholds, but the less concentrated oxygenated compounds, which have low sensory thresholds (Piringer and Ruter, 2000). One-heptene-3-one and 2-nonenal were identified as important tainting compounds from PE-containing packaging materials (Piringer and Ruter, 2000). Kontominas et al. (2006) indicated 2-ethyl hexanal, heptanal, octanal, and 2,6-di-tert butyl quinone might have been responsible for tainting from the PE and PETE net and threads. PE odor is characterized by a number of sensory descriptors such as candle-like, stale, musty, stuffy, rancid, and soapy (Piringer and Ruter, 2000). Many of these sensory descriptors are used in describing oxidation of foods components. Heat during polymer processing (polycondensation and melt processing) and package formation can cause formation of low molecular weight acetaldehyde (Robertson, 2006). Dabrowska et al. (2003) and Ewender et al. (2003) reported that the concentration of acetaldehyde, due to the degradation of the PETE polymer during bottle formation, reaches 4.5–5.5 ppm when the process is continuous and stable. However, concentrations of more than 50 ppm have been observed even when there was only a short standstill in the bottle production line (Dabrowska et al., 2003). The production of acetaldehyde appears to be initiated when PETE bottles are exposed to ozonated water used for the disinfection and washing of bottles during their manufacture (Dabrowska et al., 2003). Several authors have reported that aldehydes are major oxidation products during ozone disinfection (Weinberg et al., 1993; Richardson, 1998) and Mehta and Bassette (1978) reported high amounts of acetaldehyde production after milk cartons were exposed to ethylene oxide sterilization. However, Song et al. (2003) reported that no new compounds were formed in PETE upon exposure to ozonated water. Acetaldehyde migration from PETE can alter flavor profiles of foods and beverages.
Sensory Impacts of Food–Packaging Interactions
41
Acetaldehyde imparts a fruity, green apple flavor and is found in many food products, including fruits, beverages, and yogurt. The sensory impact of the acetaldehyde taint from packaging is highly dependent on the food system. van Aardt et al. (2001a) determined human threshold levels for acetaldehyde in spring water (167 ppb), milk, and chocolate milk using a three-sample alternate forced choice test series with a panel of 25 people. Threshold values for acetaldehyde in milk (whole milk, 4040 ppb; low-fat milk, 4020 ppb; nonfat milk, 3939 ppb) were not affected by fat content. Chocolate milk had a threshold value of 10,048 ppb, which compared well with the results of Bills et al. (1972) who looked at the threshold level of acetaldehyde in strawberry milk (11,700 ppb). The higher threshold for chocolate milk, as compared to whole milk, is likely due to the masking effect that chocolate flavoring agents had on acetaldehyde flavor. Several authors have reported an increased acetaldehyde concentration due to exposure to light (Cadwallader and Howard, 1998; Cladman et al., 1998; Jenq et al., 1988; van Aardt et al., 2001b). The combination of these two sources—migration from PETE packaging and exposure to light—could potentially raise the level of acetaldehyde in a food product above threshold levels. van Aardt et al. (2001b) studied the sensory impact of acetaldehyde migration from PETE bottles into milk stored under light. Acetaldehyde concentration in light-exposed milk (3.25% milkfat, 18 days at 4 C) packaged in clear PETE, clear PETE with UV blocker, amber PETE and high density PE (HDPE) was about the same as milk packaged in glass stored at the same conditions (range for all packaging: 1265–2930 mg/kg). A significant difference in the concentration of acetaldehyde in lightexposed milk as compared to light-protected milk was found. A trained sensory panel (n ¼ 8) used a 9-point verbal category scale (1 ¼ not detectable; 9 ¼ very strong) and rated the acetaldehyde intensity around 2 (‘‘trace, not sure’’) for both light-exposed and light-protected milk from all packaging. No significant difference in sensory perception of acetaldehyde off-flavor due to either light exposure or bottle type was observed. This lack of a difference, even when significant differences in acetaldehyde concentration were found in light-exposed milk as compared to lightprotected milk, is likely due to acetaldehyde concentrations below sensory thresholds for milk. Setting maximum specifications for acetaldehyde concentrations in PETE would protect against sensory impacts in bottled water, which has a very low sensory threshold for this compound. Studying the sensory impact of materials in contact with water is valuable in understanding the impact of materials use in water distribution and food packaging. The trend for replacing copper piping with polymer-based plumbing materials has created a host of taste and odor problems associated with potable water (AwwaRF, 2002; Dietrich et al., 2005; Rigal 1992, 1995; Rigal and Danjou, 1999; Tomboulian et al., 2004).
42
Susan E. Duncan and Janet B. Webster
Cross-linked polyethylene (PEX) used as home or retail plumbing materials can cause perceptible changes in the odor of tap water (Durand and Dietrich, 2007). A water industry standard flavor profile analysis (Standard Methods 2170B) was used by 10 trained panelists to determine the odor profile of synthetic tap water stored in PEX pipe, manufactured by a silane cross linking procedure. Water, with and without disinfectants (chlorine, chloramine), was stored in PEX pipes for 3–4 days and for three consecutive periods. Odor descriptions generally were summarized as ‘‘burning-solvent, plastic’’ odor at the weak to moderate level. A variety of descriptors were used with ‘‘alcohol, plastic, and sweet chemical’’ terms common in the first flush period. The addition of disinfectants provided odor characteristics described as ‘‘glue.’’ Subsequent flushes changed the odor profile with terms such as fruity, spicy, bitter, almond, rotten swampy, and burning plastic suggesting that the odor was becoming more distinct and perhaps even more objectionable. The chemical 2-ethyoxy-2-methylpropane (ETBE) was identified as a contributor to the odor of PEX pipe. Marchsan and Morran (2002) found that flavor descriptions varied between chlorinated and nonchlorinated water in contact with PE and PP with stronger tastes frequently found in chlorinated samples. ‘‘Plastic/rubber’’ terms were used for chlorinated and nonchlorinated waters stored in PP and PE as well as in nonchlorinated waters from acrylonitrile/butadience/styrene (ABS). ‘‘Plastic/chemical’’ descriptors were used for chlorinated and nonchlorinated waters in PP and PE and polyurea materials, and in ABS materials for chlorinated waters only. Polyurethane materials contributed chemical tastes to chlorinated waters and medicinal flavors to nonchlorinated water. The ‘‘chemical’’ term also was applied to chlorinated water stored in PP, PE, and ABS and nonchlorinated water stored in ABS. ‘‘Medicinal’’ also was used to describe both nonchlorinated and chlorinated waters stored in PP. Off-flavors in water and milk packaged in PE-coated paperboard cartons have been described by consumers as ‘‘unpleasant plastic’’ (Berg, 1980). Leong et al. (1992) examined the sensory impact of milk (nonfat (0.05%), lowfat (2%), and whole (3.25%)) packaged in gable-top PE-coated paperboard cartons (half-pint (236 ml), quart (946 ml), and half-gallon (1890 ml). A 10-member panel was selected based on discriminating ability for milk off-flavors and a paired comparison test or a pairwise ranking test was used to evaluate packaging flavor in the samples on days 1, 3, and 6 of storage at 2.2 C. Control samples initially were packaged in HDPE at the manufacturer but transferred to glass containers within a few hours after transport to the laboratory. Half-pint milk samples (all fat levels) packaged in PE-coated paperboard were clearly distinguishable from milk packaged in glass. The ability to discriminate between milk packaged in PE-coated paperboard and glass increased
Sensory Impacts of Food–Packaging Interactions
43
with decreasing fat content. Packaging flavor seemed to develop most within the first three days of storage. More packaging flavor was found in half-pint packaged milk than in larger containers, probably because of the higher contact surface:volume ratio. Leong et al. (1992) documented that heat sealing of the cartons was not the source of the taint. Packaging flavor was found in water samples on day 1 similar to the results for low fat milk. Packaged flavor in milk stored in PE-coated paperboard develops within 24 h and is more easily detected with decreasing fat content, possibly because milk fat appears to mask or dilute this flavor defect (Leong et al., 1992). This conflicts with most evidence that as fat content in a food product increases, so does that of the migration of volatile compounds from the packaging material to the food product and so does the presence of off-odors and flavors. However, nonfat milk has even less flavor character than lowfat and whole milk, which may be one reason that the package flavor is more evident. The most common packaging material for milk is now HDPE. Many dairy processors blow mold HDPE containers for milk, juice, and water in the processing plant. Without proper specifications for sensory quality of the granules, taints may be readily noticeable in these products. HDPE pellets were reported to have low concentrations of odor-producing compounds but the sensory impact of these compounds, as determined by GCO was strong (Villberg et al., 1997). Leachate water from high quality HDPE was described as having a sweet, metallic, stony, and pungent taste and sweet, chemical, stale, dirty, and foul odor. Some HDPE pellets contributed negative taste (dusty, stale, plastic, foul, stink bug, and candle grease descriptors) and odors (stale, dirty, foul). The majority of odor compounds were carbonyl compounds. 2-octenal, which gives a mushroom odor, and butylacrylate, which gives a gluelike odor, were the strongest, while moderate odors were imparted by 2-propanal (glue-like odor) and methyl hexanal (green, pungent) and were found only in the poor quality pellets. Other compounds found to leach into water from poor quality HDPE were 2,4-heptadienal, nonanal, and undecadienal. Ethyl propanate leached out in extremely small amounts but gave a very strong odor, which smelled like glue (Villberg et al., 1997). HDPE and epoxy, frequently used in home plumbing, were implicated in having the greatest impact on odors in tap water (Dietrich, 2007). A trained sensory panel, using the water industry standard flavor profile analysis method, evaluated six different plumbing materials for odor intensity. In order of increased odors in tap water (simulated) as a function of material indicated chlorinated polyvinyl chloride (cPVC) as having the least increase in odors, with cross-linked PE (PEX)-a, copper, and PEX-b having increasing levels of odors. Increased odors can cause sensory annoyance. Water stored in HDPE was characterized as having
44
Susan E. Duncan and Janet B. Webster
a ‘‘waxy, plastic, citrus’’ odor at moderate levels (Heim and Dietrich, 2007b). Odor intensity increased in the presence of chlorine and ‘‘chemical, plastic’’ descriptors were used; chloramine disinfectants also caused an increase in odor intensity described as ‘‘waxy-crayon, plastic.’’ Piagentini et al. (2002) studied the effects of citric acid, ascorbic acid, and type of packaging film on the sensory characteristics, chlorophyll retention, and weight loss of fresh cut spinach in refrigerated storage. Spinach was packaged in either mono-oriented PP bags or in LDPE bags and stored at refrigeration temperatures for 14 days. A trained sensory panel evaluated off-odor, appearance, wilting, and color using a 10 mm unstructured line scale. Storage time significantly (< 0.001) affected sensory attributes, while the type of packaging film only influenced off-odor development (p < 0.001). Off-odor development was greater for oriented PP than for LDPE. The type of packaging film had no effect (p > 0.05) on visual sensory characteristics. The intensity of off-odor packaged in LDPE reached an average of 7.3 after 14 days (9 ¼ none, 0 ¼ severe). The oriented PP had an average sensory value of 5 after 14 days. Di Pentima et al. (1995) found similar results with broccoli, whole spinach, and asparagus packaged in different plastic films stored at 4 C.
4. Polyvinyl chloride (PVC) and chlorinated PVC (cPVC) PVC and cPVC can have a metallic odor and taste due to accidental contamination by antimony (Tamboulian et al., 2002). Phenolic and acetone odors are attributed to m-chlorophenol and cyclohexanone, respectively. Threshold levels for these compounds in water are 0.005 ppm for m-chlorophenol and 0.12 ppm for cyclohexanone. PVC pipes and polymer coatings have been found to have an organosulfur odor that is attributed to ethyl-2-mercaptoacetate. This compound arises from the interaction between low molecular weight alcohols with synthetic organic compounds added to PVC as a heat stabilizing agent (Sides et al., 2001). Heim and Dietrich (2007b) did not find a significant odor in synthetic tap water stored in cPVC pipes compared to water stored in glass control pipes.
5. Epoxy Epoxy is used as a lining for water reservoirs, water mains, and home plumbing systems (Heim and Dietrich, 2007a). These applications can impact sensory quality of tap water in food manufacturing, food service operations, and residential homes. This effect may be most noticeable in water but residual aroma and flavor compounds may cause a taint in foods prepared with these water sources. An odor assessment, using a water industry standard flavor profile analysis method, identified a strong relationship between water (simulated tap water, pH 7.7–7.9) stored in epoxy-lined copper pipes for 3–4 days and an odor described
Sensory Impacts of Food–Packaging Interactions
45
as ‘‘plastic, adhesive, putty.’’ In addition, a noticeable decrease in chlorine and chloramine disinfectant odors were identified (Heim and Dietrich, 2007a).
B. Taints from additives or noncontacting materials The primary chemicals associated with taints are solvents or inks, aliphatic aldehydes and ketones, phenols or halogenated phenols, and anisoles (Lord, 2003).
1. Printing inks, varnishes Color, graphics and labels on primary and secondary packages are used to inform, advertise, attract attention, and promote the product within. Printing (or packaging) inks and varnishes contain colorants, binders, solvents, and additives. Most of the off-odor related consumer complaints linked to packaging at Nestle were attributed to residual solvents from inadequate printing or converting processes or the use of low-quality promotional items (Huber et al., 2002). These systems may be characterized as solvent-based, water-based, oleo-resinous, or UV- or electron beam (energy) curing (Aurela and Soderhjelm, 2007). Although these materials are applied to the external surface of the packaging material, low molecular weight compounds will easily migrate through the packaging material, with the exception of glass and aluminum foils, into the food. Some of these compounds have a noticeable smell and can contribute to taint of the food product. Oxidized aromas in the foods may be partially related to oxidation of vegetable oils used in offset printing or alkyd resins, used as binders in inks. Aldehydes and ketones resulting from the oxidation process can unexpectedly modify the flavor and odor of food within the package, creating a negative, even repulsive, sensory response (Soderhjelm and Eskelinen, 1985). Mineral oils may contain aromatic compounds, which can diffuse readily through fibrous or plastic packaging materials. Toluene and xylene, which are aromatic compounds, should be avoided in printing of food packages (Aurela and Soderhjelm, 2007). Hydrocarbon compounds in lithographic inks have been sources of taints (Kilcast, 1996). Low odor inks and varnishes should be chosen. At the minimum, adequate time for airing for solvent or UVcured systems, which may produce taints from trace residues of acrylate monomers and from benzophene photoinitiators, before packaging should be allowed (Aurela and Soderhjelm, 2007; Kilcast, 1996). Printed and varnished paperboards containing residual solvents may be one of the main sources of taints in foods (Soderhjelm and Eskelinen, 1985). Placing printed premiums (coupons) within a food package is common but these materials also may be sources of taints. Premiums intended for packaged dry mix beverages were tested for their contribution of
46
Susan E. Duncan and Janet B. Webster
odors prior to inclusion (Apostolopoulos, 1998). Overwrapped (PE) and unwrapped premiums were placed in Mason jars, sealed, and heated at 49 C for 1 h and cooled. An odor evaluation panel (n ¼ 4), familiar with solvent odors associated with the packaging industry, rated odor intensity on a scale of 0–10 (0 ¼ no odor; 8–10 ¼ excessive odor), described the odor, and also indicated if the odor was objectionable or not. The solvent odor was attributed to the PE resin or the paint used with the premiums. Cyclohexane concentration, as determined by GC/MS was 16 times higher in unwrapped premiums than attributed to the overwrapped premiums; toluene, 2-methyl heptane, and 3-methyl hexane also were detected. The sensory panel identified no odor associated with the overwrapped premiums but identified a very strong (rating of 9), solvent-like, objectionable odor for the unwrapped premiums. The PE overwrap contributed no distinct odor and effectively contained the premium odor. The paint used in the manufacturing and printing of the premiums was implicated. However, unsealed or punctured overwrap would not provide appropriate protection, potentially leading to taints in the beverage powders.
2. Coloring agents Coloring agents are often used in materials to provide protection from visible and ultraviolet light, and should be considered as a potential source of taints. Heinio and Ahvenainen (2002) studied the odor of packaging materials as a function of different coloring agents. However, there was no direct indication, based on odor, that the coloring agent was the source of taint in the packaged food. They recommended that odor testing should only be regarded as an indicator.
3. Antioxidants
Tomboulian et al. (2002) has reported that butylated hydroxytoluene (BHT) can impart a ‘‘burnt plastic’’ odor and is an additive in HDPE pipes. Quinone may be derived from BHT due to interactions with residual chlorine in pipes (Anselme et al., 1985). Yam et al. (1996) reported that antioxidants, such as vitamin E, Irganox 1010, and BHT, contributed to off-flavors in water. Vitamin E yielded less off-flavor, possibly due to lower aldehyde and ketone concentrations. Extrusion temperatures over 280 C and exposure time for melt contributed to more oxidation of LDPE films and higher intensities of off-flavors in water in contact with LDPE with different antioxidants (Andersson et al., 2005).
C. Taints from recycled materials Recycled materials may contain absorbed odorous or flavorful molecules from earlier use that, when introduced into a new packaging material, may cause taint (Franz and Welle, 2003; Kilcast, 1996). Analytical detection
Sensory Impacts of Food–Packaging Interactions
47
limits of instrumentation may be higher than sensory thresholds from some flavor compounds so, while the recycled material may appear to have low or nondetectable concentrations of volatile contaminants, there may be sufficient levels for sensory detection (Franz and Welle, 2003). Limonene, for example, is readily absorbed from citrus juices into packaging materials. Analysis of recycled PETE after exposure to model compounds showed average and maximum values of 18.6 and 86 ppb for acetaldehyde and 2.8 and 20 ppb for limonene. Analysis of contaminants such as solvents in recycled plastics showed extremely low levels ranging from 1.4 to 2.7 ppb and resulted from only 0.03% to 0.04% of the recollected PETE bottles (Franz et al., 2004). Recycled HDPE, PP, PS, and PETE polymers demonstrated sensory properties characteristic of the virgin polymers but also additional odor notes (Huber and Franz, 1997). A sensory panel readily identified the recycled polymer from virgin polymers based on these additional odor notes. Recycled HDPE was most different from the virgin material whereas recycled PETE had the lowest odor deviation. Recycled PS and PP had more odor notes than the matched virgin material but did not have as great a difference as was found for HDPE. Use of recycled materials should be considered on a case by case basis, using appropriate sensory testing to verify sensory intertness of recycled PETE or any other materials (Franz and Welle, 2003). If the contained product is bland in odor and flavor, the impact of these molecules may be even more evident (Kilcast, 1996). Recycled materials would not be appropriate for water or milk packaging. With the continuing desire to recycle and reuse plastic packaging materials to reduce their environmental impact, the absorption of compounds into the packaging material will become increasingly important. It has been reported that the recycling process used for plastics are not completely efficient in their ability to eliminate absorbed compounds. These compounds can then desorb into the new product upon reuse of the plastic (Safa and Bourelle, 1999). Deep cleaning, or supercleaning, technologies for recycled polymers are safe and produce bottles with very low levels of contamination, positively influencing sensory properties of the recycled materials (Franz and Welle, 2003; Franz et al., 2004).
VII. SCALPING/SORPTION Sorption of flavor compounds, or more colloquially ‘‘scalping,’’ is considered a major factor in the degradation of food quality (Arora et al., 1991; Ayhan et al., 2001; Charara et al., 1992; Fukamachi et al., 1996). The term sorption encompasses the properties of absorption, adsorption, and cluster formation and describes the penetration and movement of a chemical compound into a polymer (Robertson, 2006). Aroma and flavor
48
Susan E. Duncan and Janet B. Webster
perception often involves the interplay of many compounds in a specific proportion. Therefore, any disturbance of this balance due to sorption can change the sensory characteristics of the product and reduce its acceptability (Arora et al., 1991; Sajilata et al., 2007). All plastic materials have some sorption capacity for flavor molecules (Gremli, 1996), which can result in a sensory impact. Higher storage temperatures will accelerate the sorption of volatiles. Volatiles associated with flavor of a given food product may be decreased by as much as 20% by sorption (Gremli, 1996). Absorption of flavor molecules into the package may be affected by a number of parameters associated with the material and the food (van Willige, 2002). Crystallinity, morphology, and polarity of polymers can influence the rate of absorption. Size, concentration, copermeants, and polarity of flavor molecules within the food system also affect absorption. Storage temperature and time exposed to the food matrix affect polymer and food matrices, creating additional challenges in determining effects of materials in contacts with foods. Not only can absorption alter the aroma and flavor of a product, it can also change the mechanical properties of the polymer. Swelling and gas permeability are factors that effect the physical properties of a polymer (Robertson, 2006; Sadler and Braddock, 1991; Safa and Bourelle, 1999). Swelling occurs when compounds are absorbed into the polymer and distort the shape of the package. As absorption increases there is also a subsequent increase in gas permeability. This increase in gas permeability can affect the shelf-life and sensory quality of a food by, for example, increasing oxidation. In very severe cases, absorption can affect package integrity. Color and appearance of a food are important quality aspects on which consumers base many initial purchase and consumption decisions. Nylon-6 polyamides may scalp dye materials from foods, altering the food color intensity. Liquids and moist foods (high water content) in direct contact with the polymer have the greatest potential for changes or loss of color. Migration of food colorants into packaging material can alter the package and product appearance or cause staining of food contact surfaces on household items. Such problems increase consumer complaints and lead to decreased sales (Oehrl et al., 1991). However, there is very little published research that directly considers color and appearance changes as a function of interaction of materials and colorants. There are a number of studies that look at the absorption of flavor compounds into different polymer packaging materials (Arora et al., 1991; Ayhan et al., 2001; Charara et al., 1992; Fukamachi et al., 1996; HernandezMunoz et al., 2001; Imai et al., 1990; Konczal et al., 1992; Letinski and Halek, 1992; Moshonas and Shaw, 1989; Nielsen et al., 1992; Sadler and Braddock, 1991; Safa and Bourelle, 1999; van Willige et al., 2000, 2002, 2003). There are fewer studies looking at the effect that this absorption has on the sensory quality of the food. The few sensory studies that have been
Sensory Impacts of Food–Packaging Interactions
49
done to date are contradictory and more research into this area is imperative (Ayhan et al., 2001; Durr et al., 1981; Kwapong and Hotchkiss, 1987; Mannheim et al., 1987; Moshonas and Shaw, 1989; Pieper et al., 1992; Sadler et al, 1995; Sharma et al., 1990). Several studies found that scalping of flavor volatiles by polymers did not affect the sensory quality of juices. Sharma et al. (1990) found that fruit squashes and tropical fruit beverages stored with PP and PE had no differences detected using triangle testing. Pieper et al. (1992) reported that a 50% decrease in limonene concentration, along with a small decrease in alcohol and aldehyde concentration, did not affect the sensory quality of orange juice. Sadler et al. (1995) tested the effect of volatile compound absorption from orange juice into LDPE, PETE, and EVOH stored at 4.5 C for 3 weeks on the sensory characteristics of the juice and observed no change. van Willage (2002) reported that a sensory panel (n ¼ 27) could not find a significant difference in flavor of reconstituted orange juice packaged in LDPE, PET, or polycarbonate (PC) although analytical flavor chemistry documented a large decrease in flavor constituents. Other investigators, however, found that the absorption of flavor compounds by polymer packaging material did affect the perception of odor and flavor (Ayhan et al., 2001; Kwapong and Hotchkiss, 1987; Mannheim et al., 1987; Moshonas and Shaw, 1989). A sensory panel found no significant difference in color in orange juice processed by pulsed electric field stored in glass, PETE, HDPE, and LDPE, even though there were significant analytical differences (Ayhan et al., 2001). A significant difference in flavor was found in juice stored in LDPE after 56 days compared to the other packaging materials but no significant differences were found in flavor after 112 days in glass, PETE, and HDPE. Overall, the retention of all flavor compounds was significantly higher in glass and PETE than HDPE and LDPE. An increase in storage temperature had adverse effects on flavor and color. There was more loss of aldehydes and esters in all packages after 2 weeks than hydrocarbons, and flavor loss was more advanced in HDPE and LDPE than in PETE and glass. These results can be explained by both the absorption of flavor compounds and by the acceleration of the production of degradation products due to oxygen permeability and increased storage temperature (Ayhan et al., 2001).
VIII. PROTECTION OF SENSORY QUALITY BY FOOD PACKAGING Packaging materials can significantly increase the shelf-life of a food by reducing or slowing the degradation of the food. Package characteristics such as decreased oxygen and light permeability, for example, are
50
Susan E. Duncan and Janet B. Webster
responsible for the increased shelf-life. The proper choice of packaging, while at times difficult to do, is extremely important for protecting and maintaining sensory quality. Saint-Eve et al. (2008) identified that packaging choice affected the sensory quality, specifically aroma, of flavored stirred yogurts with 0% or 4% fat content. A trained sensory panel (n ¼ 8–15) evaluated yogurt packaged in glass, PS, or PP over a 28-day refrigerated storage period. Evaluations, using an unstructured line scale with ‘‘weak’’ and ‘‘very intense’’ as the anchors, reflected 10 odor, 15 aroma, 2 taste, and 3 texture-in-mouth characteristics. While aroma intensity and profile changed for yogurts packaged in all materials over time, glass provided the best protection for aroma and flavor intensity as it had the best barrier properties. A time effect was evident with relation to sensory perception and scalping of aroma compounds. Loss of some aroma compounds was greater in yogurts stored in PP than in PS but the flavor chemistry stabilized before the 28th day of storage for yogurt packaged in PP. The kinetics of aroma compound sorption were slower in PS than in PP, perhaps because of the differences in crystallinity between the two materials at 4 C. Nonfat yogurts packaged in glass and PS developed similar sensory odor and aroma changes over the 28-day storage. Fruity notes were better retained in PS-packaged products compared to PP-packaging but more acids were also noted. The authors also suggested that higher fat content (4%) product may lose less volatiles into packaging by absorption because of the lipophilic nature of many aroma compounds, thereby reducing the interaction with packaging materials. Fewer or less intense flavor and odor defect characteristics were identified in 4% yogurts packaged in PS at the end of shelf-life than for yogurt packaged in glass or PP.
A. Protection against light One major reason for nutrient loss and off-flavor development today is due to extended exposure to fluorescent light in food retail display cases. Many foods and beverages are susceptible to light-induced reactions, especially those with photo-sensitizers. Natural pigments found in foods that commonly act as photochemical initiators are flavonoids, riboflavin (vitamin B2), chlorophyll, heme, and vitamin K. The chemical effects of photo-oxidation on food components results in off-flavor development or changes in pigmentation or appearance. Dairy products, which are very susceptible to photo-oxidation, develop a distinct, unpleasant flavor described as ‘‘cardboard’’ or ‘‘burnt feathers’’ (Bodyfelt et al., 1988). Sunstruck flavor of beer is also a noteworthy flavor defect caused by light. Fruits, vegetables, pigmented beverages, candies, and other colored food systems may demonstrate color change from photooxidation of pigments. Ingredients, such as powdered milk, flavorings, and
Sensory Impacts of Food–Packaging Interactions
51
colorants, will develop off-flavors, off-odors, or color changes as a function of light exposure. The use of packaging materials to protect food systems from the effects of light is common. However, the most effective solutions for protection of product and ingredient sensory quality is to provide a complete light block, which is not always the most effective method for marketing a product to consumers. The primary plastic packaging materials used for refrigerated milk products are HDPE and PETE (Anonymous, 2002). Clear glass, olefincoated fiberboard, blow molded PE, and high-density PC or PE also are used. Glass allows 91% light transmission, HDPE allows 57%, while fiberboard allows 4% light transmission. Fiberboard containers were found to protect milk from light oxidation for up to 48 h whereas milk in plastic or glass containers developed light oxidized flavor within 12 h of exposure to 100–200 ft-c fluorescent light. Milk in clear PE pouches showed off-flavor development in 6 h after exposure to 100 ft-c and 3 h at 200 ft-c. Many investigators have found that packaging materials that protect milk and dairy products from photo-oxidation are important to sensory quality (Christy et al., 1981; Cladman et al., 1998; Dimick, 1973; Gorgern, 2003; Papachristou et al., 2006a,b; Rysstad et al., 1998; Simon and Hansen, 2001; van Aardt et al., 2001b; Zygoura et al., 2004). Zygoura et al. (2004) found that both clear and pigmented (2% TiO2) PETE had significantly higher lipid oxidation than paperboard and 3-layer pigmented coextruded HDPE and monolayer pigmented HDPE between days 3 and 7 (end of test). Oxygen permeability of the packaging material did not affect oxidation, but this could have been due to the large headspace of the packages used. Sensory evaluation, using a trained panel and a flavor and intensity rating scale (0 ¼ unfit for consumption, 5 ¼ very good), showed that milk packaged in clear bottles had a much lower acceptability score than milk packaged in pigmented bottles. Simon and Hansen (2001), using an untrained panel and a ‘‘difference from control’’ test, found that milk packaged in oxygen barrier board (EVOH and foil) deteriorated much more slowly than milk packaged in standard or juice boards. The foil-lined board had the added benefit of a light block. Inhibition of oxygen permeation into a package does not solely protect against degradation. Milk packaged in HDPE with a carbon black layer (light barrier) and no oxygen barrier was shown to have better protection against light oxidation than HDPE with an EVOH oxygen barrier but no light barrier (Gorgern 2003; Moyssiadi et al., 2004). PETE has an advantage over high-density poly(ethylene) (HDPE), the polymer currently used for the larger sizes of milk packaging, since the oxygen transmission rate at 4 C, 50% relative humidity, and 21% oxygen of a commercial one-pint PETE bottle is 19 ml/day compared to 390– 460 ml/day for a commercial one-pint HDPE bottle (van Aardt et al., 2001b). However, translucent HDPE has an advantage over clear PETE
52
Susan E. Duncan and Janet B. Webster
in that it blocks approximately 40% light between 300 and 700 nm, whereas clear PETE only blocks 20% light in the same range (van Aardt et al., 2001b). The efficacy of film over-wraps, made from single and multilayers of iridescent film, to reduce the production of light oxidation in milk and for effectiveness in controlling light oxidized flavor in milk was tested (Webster, 2006; Webster et al., 2007). A balanced incomplete block multisample difference test using a ranking system and a trained panel was used for the evaluation of light oxidation flavor intensity. Packaging overwraps limited the production of light oxidation flavor in milk over time but not to the same degree as the complete light block. Blocking all visible riboflavin excitation wavelengths was better at reducing light oxidation flavor than blocking only a single visible excitation wavelength. However, blocking transmission of all riboflavin excitation wavelengths at the levels suggested by the International Dairy Federation (IDF) was not sufficient to completely protect against the production of light oxidation flavor, suggesting the presence of a photosensitizer other than riboflavin in the milk. Bray et al. (1977) found that 73% of 2000 consumers preferred nonexposed milk to light exposed milk.
B. Preventing moisture loss Protection of color and texture, as well as odor and flavor, of food products also is an important consideration when choosing the appropriate food packaging material. Nylon-LDPE plastics helped protect vacuum-packaged burnt coconut meat and coconut water from color and texture changes over a 28-day storage compared to a PVC film ( Jangchud et al., 2007). Burnt aromatic coconut water and meat were stored at 5 C and 80–90% relative humidity in two package treatments with an unwrapped treatment as the control and evaluated over a 28-day shelf-life. Packaging treatments included: (1) PVC film (thickness: 11 mm; water vapor transmission rate (WVTR): 210 g/(m2d); OGTR: 8133 cc/ (m2d bar)); and (2) under conditions of vacuum in a Nylon: linear LDPE plastic bag (15 mm, 120 mm thickness each, respectively; WVTR: 5.1 g/(m2d); OGTR: 73 cc/(m2d bar)). Sensory evaluation of color (water: yellow, transparency; meat: white), texture (hardness), as well as several odor and flavor characteristics were assessed on a 15-cm unstructured line scale (0 ¼ ‘‘weak’’; 15 ¼ ‘‘strong’’) by 12 experienced/trained panelists. Vacuum packaging effectively ESL of the burnt coconut by limiting color alteration in the coconut water and meat and maintaining hardness of the coconut meat. Panelists rated the coconut packaged in Nylon: linear LDPE with high acceptability after 28 days whereas the PVC-film wrapped treatment had declined in quality, partially as a result of
Sensory Impacts of Food–Packaging Interactions
53
increased microbial growth, by 14–18 days. The barrier properties of the Nylon: linear LDPE film maintained the desired vacuum storage of the product, which contributed to the improved product quality.
IX. USING PACKAGING TO IMPROVE SENSORY QUALITY Smart packaging, referring to value-added packaging features that enhance the functionality of a product, encompasses active packaging as well as other developments that improve product safety and efficiency (Robertson, 2006). Examples of active packaging systems include oxygen and ethylene scavengers, carbon dioxide scavengers and emitters, preservative releasers, ethanol emitters, moisture absorbers, and flavor/odor adsorbers. Innovative technologies for these systems may have intended impact on sensory characteristics of foods through control of microbial growth, oxidation reactions, moisture migration, and absorbance of undesirable odor-impacting molecules. While there is a large body of literature in the area of smart (active, intelligent, other) packaging materials and systems, the relationship to sensory impact on humans is secondary to the chemical or microbiological impacts, including analytical flavor chemistry, that are primarily studied. We have illustrated the potential for sensory impacts of some packaging innovations with a few examples.
A. Sensory impact of novel antimicrobial ingredients in packaging systems Controlling microbial degradation of food systems is paramount in maintaining sensory quality. Inclusion of antimicrobial agents from naturally derived sources within the polymer reduces the risk from a consumer perspective (Suppakul et al., 2008). Extracts of clove, grapefruit seeds, huanglian, rhubarb, and basil have been embedded in LDPE to test for antimicrobial effects (Suppakul et al., 2008). Linalool and methylchavicol, antimicrobial compounds from basil were embedded in LDPE and demonstrated inhibitory effects on microbial growth in naturally contaminated cheddar cheese samples. These materials (control LDPE film, linalool-LDPE, and methylchavicol-LDPE) also were tested for sensory impacts on wrapped, cubed cheddar cheese over 6 weeks of storage at 4 C. The materials and the cheese were both sterilized by ultraviolet light prior to contact. Using a triangle test method, a panel of 10 members evaluated the cheese for basil taint. Linalool-embedded LDPE did not impact flavor of the cheddar cheese over the 6-week storage as panelists could not detect the difference between cheese stored in LDPE or linaloolembedded LDPE. However, methylchavicol-embedded film caused a taint in the cheese as early as 7 days. This sensory impact was detectable
54
Susan E. Duncan and Janet B. Webster
again only in the fourth week of the study. Methylchavicol was reported to have a more distinct flavor than linalool and was more evident in the cheese. While the microbial shelf-life of cheese was improved with use of methylchavicol-embedded LDPE as compared to linalool-LDPE and LDPE control materials, sensory effects were noticeable with use of this compound, illustrating the potential for reducing the commercial success of the antimicrobial material (Suppakul et al., 2008). Selection of naturally occurring antimicrobial agents or other active additives should be based on both intended efficacy as well as potential for protecting the sensory integrity of the product.
B. Flavor and odor absorbers for improved flavor Some food processing methods impose negative sensory quality parameters on food systems. UHT processing of fluid milk and citrus juices, for example, increases the perception of cooked or stale flavor and odor notes because of thermal degradation of macro- and micronutrients within the system. An increase in low molecular weight aldehydes and ketones has been identified as primary contributors to these negative sensory characteristics in heat processed beverages (Suloff, 2001). Milk, soymilk, and water are bland, having low flavor and odor profiles, and low concentrations of migrating molecules from PETE or other materials can influence sensory quality (Norton, 2003; van Aardt et al., 2001a,b). Odor-adsorbing materials were synthesized from cyclodextrin, d-sorbitol, and nylon MXD6 blended with PETE for removal of aroma compounds associated with lipid oxidation (Suloff et al., 2003) and identified that selective adsorption of low molecular weight aldehydes, compared to larger aldehydes, occurred with partition coefficients of three to six times higher magnitude. Cooperatively, Norton (2003) demonstrated the human sensory impact by evaluating the sensory threshold for targeted compounds and the efficacy of the odor-absorbing compounds in spring water, milk, and UHT procesed ESL soymilk. Norton (2003) initially established the sensory impact of selected molecules to establish if absorption by the scavenging compounds would be at efficacious levels. Low molecular weight aldehydes and ketones (hexanal, 2-heptenal, 2-pentanone, and 2,4-nonadienal) in spring water, pasteurized milk (2% milkfat), and ESL soymilk were selected as representative compounds of oxidation and high temperature processing notes. Human thresholds, based on 12 panelists, were determined, in ascending order of concentration using a series of 3-sample alternate forced choice tests, by both logistic regression and the geometric mean approach. Testing was done under controlled conditions in individual sensory booths. Untrained panelists initially were provided a reference sample of the selected compound at supratheshold concentration in the
Sensory Impacts of Food–Packaging Interactions
55
testing medium. Thresholds calculated using logistic regression were consistently higher than those calculated using the geometric mean approach, but both methods found thresholds of hexanal, 2-heptenal, 2-pentanone, and 2,4-nonadienal to vary significantly when comparing the three media. Hexanal and 2,4-nonadienal had lower thresholds than 2-heptenal and 2-pentanone. Odor detection thresholds of 2-heptenal, 2-pentanone, and 2,4-nonadienal were lowest in water, followed by milk, then soymilk at product temperatures of 4 C. Triangle tests (n ¼ 36 panelists, a 0.05) verified that five combinations of absorber addition in volatile-spiked spring water, milk and soymilk altered the aroma of the product. Addition of b-cyclodextrin in both hexanalspiked spring water and milk significantly influenced odor. Panelists also found a significant difference (p < 0.05) in 2-pentanone spiked milk with the addition of both b-cyclodextrin and d-sorbitol and in 2-heptenal spiked soymilk with the addition of d-sorbitol. Because of its ability to adsorb odors caused by lipid oxidation, b-cyclodextrin may be a good scavenger in packaging systems for milk and soymilk. However, since b-cyclodextrin is very reactive with low molecular weight compounds, there is a possibility that desirable aromas could also be scavenged. d-sorbitol also was somewhat effective as a scavenger for aroma compounds, particularly in milk. The authors did not report on efficacy of these active components in absorbing volatiles when imbedded into the material.
C. Controlling oxidation through timed release of antioxidants Active packages can be designed that contain compounds, such as antioxidants, which can migrate into the food and improve sensory quality by reducing oxidation, increasing shelf-life, and increasing nutritive quality. Additives are commonly used in polymer processing to inhibit oxidative reactions (Robertson, 2006). Excess addition of antioxidants, such as butylated hydroxyanisole (BHA) or BHT, has been effective in controlling oxidized flavor in dry breakfast cereals and crackers and other dry products (Hoojjat et al., 1987; Jadhav et al., 1996; Miltz et al., 1995). van Aardt et al. (2005a,b, 2007) studied the effect of antioxidant addition into milk and poly(lactide-co-glycolide) (PLGA) films to determine if timed release of antioxidant addition from packaging might be effective in limiting photo-oxidation of fluid milk and milk powders. Similarity testing, using triangle test methods, identified that levels of added antioxidants (0.05% a-tocopherol or 0.025% a-tocopherol or 0.025% ascorbic acid) to reduced fat milk (2% milkfat) were not significantly perceptible by panelists (n ¼ 30). Subsequently, milk was spiked with and without antioxidants and exposed to light (1100–1300 lx) for 10 h at 4 C. A sensory panel (n ¼ 24) compared the samples for difference, using the triangle test method. Milk with tocopherol/ascorbic acid addition was different from
56
Susan E. Duncan and Janet B. Webster
the light-exposed control and had a fresher milk flavor, indicating that antioxidant addition had controlled photo-oxidation. However, tocopherol alone did not provide any benefit. A timed addition of antioxidants from packaging was studied by the direct addition of antioxidants to milk over a 6-week period under lighted refrigeration conditions (van Aardt et al., 2005b). The timed effect did reduce concentrations of some odoractive compounds associated with light-induced oxidation, as determined by GCO evaluations by a trained panel (n ¼ 3). A combination of BHA and BHT significantly reduced the concentrations of heptanal and 1-octene-3ol, which are common light oxidation off-flavor compounds in lightexposed milk (van Aardt et al., 2005b). The single, initial addition of a combination of a-tocopherol and ascorbyl palmitate significantly reduced hexanal production for the first 4 weeks of the study but not thereafter. In a separate study, antioxidants were incorporated into PLGA films; films were stored in milk powders and water and oil food simulants in the presence and absence of light (van Aardt et al., 2007). Milkfat was stabilized to a degree, based on GC analysis of volatiles, against photooxidation of milk powders in the presence of antioxidant-loaded PLGA but sensory analysis was not used as confirmation. Although the sensory studies completed (van Aardt et al., 2005a,b) document the potential benefit for timed-release of antioxidants, there is no direct evidence that the change in flavor chemistry from the PLGA films resulted in a sensory impact of the milk powders or food simulants. The potential for smart packaging for improvement of food quality and safety is very high. Although sensory evaluation is essential in the early stages of development, there is an important role for sensory evaluation simultaneously with analytical assessments of new materials in contact with food systems. New materials or novel applications of materials may deliver value in improved sensory quality or create unexpected sensory impact that may not be interpreted from analytical chemical or physical methods of assessment.
X. CONCLUSIONS Sensory impacts from food–packaging interactions are probably more prevalent than acknowledged. Studies that have included controlled sensory analysis concurrently with analytical methods for studying the effects of materials on foods have demonstrated that human assessment is needed. All sectors of the food packaging supply line are important in controlling the risk of taints and avoiding quality degradation from scalping of valuable volatiles and pigments. Commitment to understanding sensory impacts from interaction of food and packaging materials can lead to innovations for improved quality and shelf-life of food systems.
Sensory Impacts of Food–Packaging Interactions
57
ACKNOWLEDGMENT This research was supported in part by the NSF-IGERT grant under Agreement No. DGE-0333378 Macromolecular Interfaces with Life Sciences (MILES).
REFERENCES Ahvenainen, R. (ed.) (2000). ‘‘Novel Food Packaging Techniques.’’ Woodhead, CRC Press, Boca Raton, FL. 590pp. Andersson, T., Holmgren, M. H., Nielsen, T., and Wesslen, B. (2005). Degradation of low density polyethylene during extrusion. IV. Off-flavor compounds in extruded films of stabilized LDPE. J. Appl. Polym. Sci. 95, 583–595. Anonymous. (1988). Migration of toxicants, flavors, and odor-active substances from flexible packaging materials to foods. Food Technol. 42, 95–102. Anonymous. (2002). Packaging opportunities for fluid milk. Dairy Foods 5, A–H. Anonymous. (2007). Free from off odour and off flavour. Colorants and additives in food and beverage packaging. Verpackungs-Rundschau 58(12), 49–50. Anselme, C., Nguyen, K., Bruchet, A., and Mallevialle, J. (1985). Can polyethylene pipes impart odors in drinking-water. Environ. Technol. Lett. 6, 477–488. Apostolopoulos, D. (1998). Identification of the source of an off-odor in premiums intended for use with dry mix beverages. In ‘‘Food Flavors: Formation, Analysis and Packaging Influences’’ (E. T. Contis, C. T. Ho, C. J. Mussinan, T. H. Parliament, F. Shahidi, and A. M. Spanier, eds.), pp. 753–757. Elsevier Science BV, Amsterdam, The Netherlands (In Sajilata, M. G., Savitha, K., Singhal, R. S., and Kanetkar, V. R. (2007). Scalping of flavors in packaged foods. Comp. Rev. Food Sci. and Food Safety 6, 17–35). Arora, D. K., Hansen, A. P., and Armagost, M. S. (1991). Sorption of flavor compounds by low density polyethylene film. J. Food Sci. 56, 1421–1423. Aurela, B., and Soderhjelm, L. (2007). Food packaging inks and varnishes and chemical migration into food. Ch. 13. In ‘‘Chemical Migration and Food Contact Materials’’ (K. A. Barnes, C. R. Sinclair, and D. H. Watson, eds.), pp. 302–319. Woodhead, Cambridge, England. Australian Food and Grocery Council (AFGC). (2007). ‘‘Organohalogen Taints in Foods.’’ AFGC, Canberra, Australia. 16pp. (http://www.afgc.org.au). Ayhan, Z., Yeom, H. W., Zhang, Q. H., and Min, D. B. (2001). Flavor, color, and vitamin C retention of pulsed electric field processed orange juice in different packaging materials. J. Agric. Food Chem. 49, 669–674. Baigrie, B. (ed.) (2003). ‘‘Taints and Off-Flavours in Food.’’ Woodhead, CRC Press, Boca Raton, FL. 203pp. Baner, A. L. (2000). Case study: Styrene monomer migration into dairy products in single serve portion packs. In ‘‘Plastic Packaging Materials for Food: Barrier Function, Mass Transport, Quality Assurance, and Legislation’’ (O.-G. Piringer, and A. L. Baner, eds.), pp. 427–443. Weinheim (Feberal Republic of Germany), Wiley-VCH. Barnes, K. A., Sinclair, C. R., and Watson, D. H. (eds.) (2007). ‘‘Chemical Migration and Food Contact Materials,’’ 464pp. Woodhead, CRC Press, Boca Raton, FL. Berg, N. (1980). Sensoric and instrumental analysis of off-flavour giving compounds from polyethylene. Presented at 3rd Int. Symp. Migration, Hamburg, Germany, October 22–24. Unilever Forschungsgesellschaft mbH, Hamburg, Germany. In Leong, C. M. O., Harte, B. R., Partridge, J. A., Ott, D. B., and Downes, T. W. (1992). Off-flavor development in milk packaged in polyethylene-coated paperboard cartons. J. Dairy Sci. 75, 2105–2111.
58
Susan E. Duncan and Janet B. Webster
Berg, H. W., Filipello, F., Hinreiner, E., and Webb, A. D. (1955). Evaluation of threshold and minimum difference concentrations for various constituents in wines. Food Technol. 9, 23–26. Bills, D. D., Yang, C. S., Morgan, M. E., and Bodyfelt, F. W. (1972). Effect of sucrose on the production of acetaldehyde and acids by yogurt culture bacteria. J. Dairy Sci. 55, 1570–1573. Bodyfelt, F. W., Tobias, J., and Trout, G. M. (1988). ‘‘The Sensory Evaluation of Dairy Products.’’ Van Nostrand Reinhold, New York, NY. 598pp. Bray, S. L., Duthie, A. H., and Rogers, R. P. (1977). Consumers can detect light-induced flavor in milk. J. Food Prot. 40, 586–587. Brody, A. L. (2002). Flavor scalping: Quality loss due to packaging. Food Technol. 56(6), 124–125. Buettner, A., Strube, A., and Welle, F. (2007). Can sensory analysis of packaging materials be standardized? Verpackungs-Rundshau 58(6), 97–99. Cadwallader, K. R., and Howard, C. L. (1998). Analysis of aroma-active components of light activated milk. In ‘‘Flavor Analysis: Developments in Isolation and Characterization’’ (C. J. Mussinan, and M. J. Morello, eds.), pp. 343–358. ACS, Washington, DC. Caul, J. F. (1961). The American approach to odour in the packaging of food. World’s Paper Trade Rev. 2325–2337. Charara, Z. N., Williams, J. W., Schmidt, R. H., and Marshall, M. R. (1992). Orange flavor absorption into various polymeric packaging materials. J. Food Sci. 57, 963–972. Christy, G. E., Amantea, G. F., and Irwin, R. E. T. (1981). Evaluation of effectiveness of polyethylene over-wraps in preventing light-induced oxidation of milk in pouches. Can. Inst. Food Sci. Technol. J. 14, 135–138. Cladman, W., Scheffer, S., Goodrich, N., and Griffiths, M. W. (1998). Shelf-life of milk packaged in plastic containers with and without treatment to reduce light transmission. Int. Dairy J. 8, 629–636. Council AFaG. (1997). Organohalogen taints in foods. Food Australia 59(3), S1–S16. Dabrowska, A., Borcz, A., and Nawrocki, J. (2003). Adehyde contamination of mineral water stored in PET bottles. Food Addit. Contam. 20, 1170–1177. Dietrich, A. M. (2007). Sensory impacts from polymer pipes interacting with drinking water oxidants. Abstract POLY 684. Am. Chem. Soc. Annual Meeting, Boston, MA. Dietrich, A., Cuppett, J., Edwards, M., Powers, K., Duncan, S., Bosch, D., and Kleczyk, E. (2005). Corrosion of copper plumbing and its effects on consumer health and economics. Proceedings of the 2005 NSF DMII Grantees Conference, Phoenix, Arizona, p. 14. Dimick, P. S. (1973). Effect of fluorescent light on the flavor and select nutrients of homogenized milk held in conventional containers. J. Milk Food Technol. 36, 383–387. Di Pentima, J., Rios, J., Clemente, A., and Olias, J. (1995). Biogenesis of off-odor in broccoli storage under low-oxygen atmosphere. J. Agric. Food Chem. 43, 1310–1313. Duncan, S. E. (2007). Flavor and aroma of food and package interaction: Perception and communication. Abstract POLY 686. Am. Chem. Soc. Annual Meeting, Boston, MA. Durand, M. L., and Dietrich, A. M. (2007). Contributions of silane cross-linked PEX pipe to chemical/solvent odours in drinking water. Water Sci. Technol. 55(5), 153–160. Durr, P., Schobinger, U., and Waldvogel, W. (1981). Aroma quality of orange juice after filling and storage in soft packages and glass bottles. Alimenta 20, 91–93. (In Linssen, J. P. H., van Willige, R. W. G., and Dekker, M. (2003). Packaging-flavour interactions. Ch. 8. In ‘‘Novel Food Packaging Techniques’’ (R. Ahvenainen, ed.), pp. 144–171. Woodhead, CRC Press, Boca Raton, FL). Ewender, J., Lindner-Steinert, A., Ruter, M., and Piringer, O. (1995). Sensory problems caused by food and packaging interactions: Overview and treatment of recent case studies. In ‘‘Foods and Packaging Materials—Chemical Interactions’’ (P. Ackermann, M. Jagerstad, and T. Ohlsson, eds.), 12pp. Royal Society of Chemistry, Cambridge, England. Ewender, J., Franz, R., Mauer, A., and Welle, R. (2003). Determination of the migration of acetaldehyde from PET bottles into non-carbonated and carbonated mineral water. Deutsche Lebensmittel-Rundschau 6, 215–221.
Sensory Impacts of Food–Packaging Interactions
59
Feigenbaum, A., Lebosse, R., and Ducruet, V. (1998). Polypropylene as active packaging material for aroma sorption from model organge juice. In ‘‘Food Flavors: Formation, Analysis and Packaging Influences’’ (E. T. Contis, C.-T. Ho, C. J. Mussinan, T. H. Parliment, F. Shahidi, and A. M. Spanier, eds.), pp. 743–751. Elsevier Science BV, Amsterdam. Frank, M., Ulmer, H., Ruiz, J., Visani, P., and Weimar, U. (2001). Complementary analytical measurements based upon gas chromatography–mass spectrometry, sensory system and human sensory panel: A case study dealing with packaging materials. Anal. Chim. Acta 431, 11–29. Franz, R., Mauer, A., and Welle, F. (2004). European survey on post-consumer poly(ethylene terephthalate) (PET) materials to determine contamination levels and maximum consumer exposure from food packages made from recycled PET. Food Addit. Contam. 21, 265–286. Franz, R. and Welle, F. (2003). Recycling packaging materials. Ch. 23. In ‘‘Novel Food Packaging Techniques’’ (R. Ahvenainen, ed.), pp. 497–518. CRC Press, Boca Raton, FL. Fukamachi, M., Matsui, T., Shimoda, M., and Osajima, Y. (1996). Cross-linking treatment of EVA film to improve flavor preservation. J. Agric. Food Chem. 42, 1989–1992. Goldenberg, N., and Matheson, H. R. (1975). ‘‘Off-flavours’’ in foods, a summary of experience: 1948–1974. Chem. Ind. 551–557. Gorgern, M. (2003). Influence of bottle wall structure on the preservation of the organoleptic properties of milk. In ‘‘Proceedings of the 13th International Conference on Polyester Containers for Food and Beverages,’’ p. 357. Schotland Business Research, Skillman, NJ. Granzer, R., Koszinowski, J., Robinson-Mand, L., and Piringer, O. (1986). Verpack-Rundsch. 37, techn.-wis-sensch. Beilage, 53–58. (In Piringer, O. and Ruter, M. (2000). Sensory problems caused by food and packaging interactions. Ch. 13. In ‘‘Plastic Packaging Materials for Food: Barrier Function, Mass Transport, Quality Assurance, and Legislation’’ (O.-G. Piringer and A. L. Baner, eds.), pp. 407–426. Wiley-VCH, Weinheim (Federal Republic of Germany)). Gremli, H. (1996). Flavor changes in plastic containers. A literature review. Perfum. Flav. 21, 1–8. Heim, T. H., and Dietrich, A. M. (2007a). Sensory aspects of drinking water in contact with epoxy-lined copper pipe. Water Sci. Technol. 55(5), 161–168. Heim, T. H., and Dietrich, A. M. (2007b). Sensory aspects and water quality impacts of chlorinated and chloraminated drinking water in contact with HDPE and cPVC pipe. Water Res. 41, 757–764. Heinio, R. L. and Ahvenainen, R. (2002). Monitoring of taints related to printed solid boards with an electronic nose. Food Addit. Contam. 19, 209–220. Hernandez-Munoz, P., Catala, R., and Gavara, R. (2001). Food aroma partition between packaging materials and fatty food simulants. Food Addit. Contam. 18, 673–682. Heydanek, M. G. (1977). Tracing the origin of off-flavors in a breakfast cereal. Ann. Chem. 49, 901A. Heydanek, M. G. (1978). How to spot a stinker—Predicting flavor effects of packaging materials. Food Prod. Dev. 12(5), 17–18. Hoojjat, P., Harte, B. R., Hernandez, R. J., Giacin, J. R., and Miltz, J. (1987). Mass transfer of BHT from high density polyethylene film and its influence on product stability. J. Packaging Technol. 22, 78–81. Huber, M., and Franz, R. (1997). Studies on contamination of post-consumer plastics from controlled resources for recycling into food packaging applications. Deutsche lebensmittelRundschan 93(10), 328–331. (In Franz, R., and Welle, F. (2003). Recycling packaging materials. Ch. 23. In ‘‘Novel Food Packaging Techniques’’ (R. Ahvenainen, ed.), pp. 497–518. Woodhead, CRC Press, Boca Raton, FL). Huber, M., Ruiz, J., and Chastellain, F. (2002). Off-flavour release from packaging materials and its prevention: A foods company’s approach. Food Addit. Contam. 19(Suppl.), 221–228.
60
Susan E. Duncan and Janet B. Webster
Imai, T., Harte, B. R., and Giaicin, J. R. (1990). Partition distribution of aroma volatiles from orange juice into selected polymeric sealant films. J. Food Sci. 55, 158–161. Jadhav, S. J., Nimbalkar, S. S., Kulkarni, A. D., and Madhavi, D. L. (1996). Lipid oxidation in biological food systems. In ‘‘Food Antioxidants: Technological, Toxicological, and Health Perspectives’’ (D. L. Madhavi, S. S. Deshpande, and D. K. Salunkhe, eds.), pp. 5–64. Marcel Dekker, New York, NY. Jangchud, K., Puchakawimol, P., and Jangchud, A. (2007). Quality changes of burnt aromatic coconut during 28-day storage in different packages. LWT 40, 1232–1239. Jenq, W. R., Bassette, R., and Crang, R. E. (1988). Effects of light and copper ions on volatile aldehydes of milk and milk fractions. J. Dairy Sci. 71, 2366–2372. Karatapanis, A. E., Badeka, A. V., Riganakos, K. A., Savvaidis, I. N., and Kontominas, M. G. (2006). Changes in flavour volatiles of whole pasteurized milk as affected by packaging material and storage time. Int. Dairy J. 16, 750–761. Katan, L. L. (ed.) (1996). ‘‘Migration from Food Contact Materials.’’ Blackie Academic and Professional, New York, NY. 303pp. Khiari, D., Barrett, S., Chinn, R., Bruchet, A., Piriou, P., Matia, L., Ventura, F., (Mel) Suffet, I., Gittelman, T., Leutweiler, P., and AwwaRF. (2002). ‘‘Distribution Generated Taste-andOdor Phenomena.’’ AwwaRF, Denver, CO. Kilcast, D. (1996). Organoleptic assessment. Ch. 4. In ‘‘Migration from Food Contact Materials’’ (L. L. Katan, ed.), pp. 51–76. Blackie Academic and Professional, New York, NY. Kilcast, D. (2003). Sensory analytical methods in detecting taints and off-flavours in food. Ch. 2. In ‘‘Taints and Off-Flavours in Foods’’ (B. Baigrie, ed.), pp. 5–30. CRC Press, New York, NY. Kim-Kang, H. (1990). Volatiles in packaging materials. J. Food Sci. Nutr. 29(4), 255–271. Konczal, J. B., Harte, B. R., Hoojjat, P., Tanaka, Y., and Osajima, Y. (1992). Apple juice flavor compound sorption by sealant films. J. Food Sci. 57(4), 967–969. Kontominas, M. G., Goulas, A. E., Badeka, A. V., and Nerantzaki, A. (2006). Migration and sensory properties of plastics-based nets used as food-contacting materials under ambient and high temperature heating conditions. Food Addit. Contam. 23(6), 634–641. Kwapong, O. Y., and Hotchkiss, J. H. (1987). Comparative sorption of aroma compounds by polyethylene and ionomer food-contact plastics. J. Food Sci. 52, 761–763. (see also p. 785). Land, D. G. (1989). Taints—Causes and prevention. Ch. 2. In ‘‘Distilled Beverage Flavour— Recent Developments’’ (J. R. Piggott and A. Paterson, eds.), Ellis Horwood, Chichester. (In (2007). ‘‘Organohalogen Taints in Foods.’’ AFGC, Canberra, Australia. 16pp. http: //www.afgc.org.au). Larsen, H., Lea, P., and Rodbotten, M. (2005). Sensory changes in extruded oat stored under different packaging, light and temperature conditions. Food Qual. Pref. 16, 573–584. Lawless, H. T., and Heymann, H. (1998). ‘‘Sensory Evaluation of Food.’’ Chapman and Hall, New York, NY. 819pp. Leland, J. V. (1997). Flavor interactions: The greater whole. Food Technol. 51, 75–76. (see also pp. 78–80). Leong, C. M. O., Harte, B. R., Partridge, J. A., Ott, D. B., and Downes, T. W. (1992). Off-flavor development in milk packaged in polyethylene-coated paperboard cartons. J. Dairy Sci. 75, 2105–2111. Letinski, J., and Halek, G. W. (1992). Interaction of citrus flavor compounds with polypropylene films of varying crystallinities. J. Food Sci. 57, 481–484. Linssen, J. P. H., Janssens, J. L. G. M., Reitsma, J. C. E., and Roozen, J. P. (1991). Sensory analysis of polystyrene packaging material taint in cocoa powder for drinks and chocolate flakes. Food Addit. Contam. 8(1), 1–7. Linssen, J. P. H., and Roozen, J. P. (1994). Food flavour and packaging interactions. In ‘‘Food Packaging and Preservation’’ (M. Mathlouthi, ed.), pp. 48–61. Blackie Academic and Professional, Glasgow.
Sensory Impacts of Food–Packaging Interactions
61
Lord, T. (2003). Packaging materials as a source of taints. In ‘‘Taints and Off-Flavours in Food’’ (B. Baigrie, ed.), pp. 64–111. Woodhead, CRC Press, Boca Raton, FL. Lozano, P. R., Miracle, E. R., Krause, A. J., Drake, M., and Cadwallader, K. R. (2007). Effect of cold storage and packaging material on the major aroma components of sweet cream butter. J. Agric. Food Chem. 55, 7840–7846. Mannheim, C. H., Miltz, J., and Letzter, A. (1987). Interaction between polyethylene laminated cartons and aseptically packed citrus juices. J. Food Sci. 52, 737–740. Marchsan, M. and Morran, J. (2002). Tastes associated with products in contact with drinking water. Abstract from the Sixth IWA Symposium on Off-Flavours in the Aquatic Environment. Barcelona, Spain, October 7–10. Mehta, R. S. and Bassette, R. (1978). Organoleptic, chemical and microbiolgical changes in ultra-high-temperature sterilized milk stored at room temperature. J. Food Prot. 41, 806–810. Meilgaard, M. C., Civille, G. V., and Carr, B. T. (2007). ‘‘Sensory Evaluation Techniques.’’ CRC Press, Boca Raton, FL. 448pp. Miltz, J., Passy, N., and Mannheim, C. H. (1995). Trends and applications of active packaging systems. In ‘‘Foods and Packaging Materials—Chemical Interactions’’ (P. Ackermann, M. Jagerstad, and T. Ohlsson, eds.), pp. 201–210. The Royal Society of Chemistry, Cambridge, England. Moshonas, M. G. and Shaw, P. E. (1989). Flavor evaluation and volatile flavor constituents of stored aseptically packaged orange juice. J. Food Sci. 54, 82–85. Moyssiadi, T., Badeka, A., Kondyli, E., Vakirtzi, T., Savvaidis, I., and Kontominas, M. G. (2004). Effect of light transmittance and oxygen permeability of various packaging materials on keeping quality of low fat pastuerized milk: Chemical and sensorial aspects. Int. Dairy J. 14, 429–436. Nielsen, T. J., Jagerstad, I. M., and Oste, R. E. (1992). Study of factors affecting the absorption of aroma compounds into low-density polyethylene. J. Sci. Food Agric. 60, 377–381. Norton, J. (2003). Efficacy of odor scavengers in reducing odor compounds in water, milk and soymilk. M.S. Thesis, Virginia Tech, Blacksburg. Oehrl, L. L., Malone, C. P., and Keown, R. W. (1991). Interactions of food, drug and cosmetic dyes with nylon and other polyamides. Ch. 4. In ‘‘Food and Packaging Interactions II’’ (S. J. Risch, and J. H. Hotchkiss, eds.), pp. 37–52. ACS Symposium Series 473, ACS, Washington, DC. Papachristou, C., Badeka, A., Chouliara, I., Kondyli, E., Kourtis, L., and Kontominas, M. G. (2006a). Evaluation of polyethylene terephthalate as a packaging material for premium quality whole pasteurized milk in Greece. Part II. Storage under fluorescent light. Eur. Food Res. Technol. 224, 237–247. Papachristou, C., Badeka, A., Chouliara, I., Kondyli, E., Athanasoulas, A., and Kontominas, M. G. (2006b). Evaluation of polyethylene terephthalate as a packaging material for premium quality whole pasteurized milk in Greece. Part I. Storage in the dark. Eur. Food Res. Technol. 223, 711–718. Passy, N. (1983). Off-flavors from packaging materials in food products—Some case studies. In ‘‘Instrumental Analysis of Foods: Recent Progress,’’ p. 413. Academic Press, New York. Piagentini, A. M., Guemes, D. R., and Pirovani, M. E. (2002). Sensory characteristics of freshcut spinach preserved by combinated factors methodology. J. Food Sci. 67, 1544–1549. Pieper, G., Borgudd, L., Ackermann, P., and Fellers, P. (1992). Absorption of aroma volatiles of orange juice into laminated carton packages did not affect sensory quality. J. Food Sci. 57, 1408–1411. Piringer, O.-G. and Ruter, M. (2000). Sensory problems caused by food and packaging interactions. Ch. 13. In ‘‘Plastic Packaging Materials for Food: Barrier Function, Mass Transport, Quality Assurance, and Legislation’’ (O.-G. Piringer, and A. L. Baner, eds.), pp. 407–426. Wiley-VCH, Weinheim (Federal Republic of Germany).
62
Susan E. Duncan and Janet B. Webster
Reineccius, G. (2006). ‘‘Flavor Chemistry and Technology.’’ 2nd ed., Taylor and Francis Group, CRC, New York, NY. 489pp. Richardson, S. D. (1998). Drinking water disinfetion by-products. In ‘‘Encyclopedia of Environmental Analysis and Remediation’’ (R. A. Mayers, ed.), pp. 1398–1421. Wiley, New York. Rigal, S. (1992). The use of organoleptic investigations to evaluate the quality of materials in contact with drinking water. Water Sci. Technol. 4(25), 41–48. Rigal, S. (1995). Odour and flavour in waters: Quantitative method for a new European standard. Water Sci. Technol. 31(11), 237–242. Rigal, S. and Danjou, J. (1999). Tastes and odors in drinking water distribution systems related to the use of synthetic materials. Water Sci. Technol. 40, 203–208. Risch, S. J. and Ho, C.-T. (2000). ‘‘Flavor Chemistry: Industrial and Academic Research.’’ ACS Symposium Series 756. American Chemical Society, Washington, DC. 180pp. Robertson, G. L. (2006). ‘‘Food Packaging Principles and Practice.’’ 2nd Ed., CRC Press, Boca Raton, FL. 550pp. Rysstad, G., Ebbesen, A., and Eggestad, J. (1998). Sensory and chemical quality of UHT-milk stored in paperboard cartons with different oxygen and light barriers. Food Addit. Contam. 15, 112–122. Sadler, G. D. and Braddock, R. J. (1991). Absorption of citrus flavor volatiles by low density polyethylene. J. Food Sci. 56, 35–38. Sadler, G., Parish, M., Davis, J., and van Clief, D. (1995). Flavor–package interaction. In ‘‘Fruit Flavors’’ (R. L. Fourseff, and M. M. Leahy, eds.), pp. 202–210. American Chemical Society, Washington, DC. (In Linssen, J. P. H., van Willige, R. W. G., and Dekker, M. (2003). Packaging-flavour interactions. Ch. 8. In ‘‘Novel Food Packaging Techniques’’ (R. Ahvenainen, ed.), Woodhead Publishing Ltd., CRC Press, Boca Raton, pp. 144–171). Safa, H. L. and Bourelle, F. (1999). Sorption-desorption of aromas on multi-use PET bottles. A test procedure. Packaging Technol. Sci. 12, 37–44. Saint-Eve, A., Levy, C., le Moigne, M., Ducruet, V., and Souchon, I. (2008). Quality changes in yogurt during storage in different packaging materials. Food Chem. 110, 285–293. Sajilata, M. G., Savitha, K., Singhal, R. S., and Kanetkar, V. R. (2007). Scalping of flavors in packaged foods. Comp. Rev. Food Sci. Food Saf. 6, 17–35. Sander, R. A., Zyzak, D. V., Morsch, T. R., Zimmerman, S. P., Searles, P. M., Strothers, M. A., Eberhart, B. L., and Woo, A. K. (2005). Identification of 8-nonenal as an important contributor to ‘‘plastic’’ off-odor in polyethylene packaging. J. Agric. Food Chem. 53, 1713–1716. Sharma, G. K., Madhura, C. V., and Arya, S. S. (1990). Interaction of plastic films with foods. I. Effect of polypropylene and polyethylene films on fruit squash quality. J. Food Sci. Technol. 27, 127–132. (In Linssen, J. P. H., van Willige, R. W. G., and Dekker, M.(2003). Packaging-flavour interactions. Ch. 8. In ‘‘Novel Food Packaging Techniques’’ (R. Ahvenainen, ed.), Woodhead Publishing Ltd., CRC Press, Boca Raton, pp. 144–171). Sides, S. L., Polowy, K. B., Thornquest, A. D., Jr., and Burinsky, D. J. (2001). Identification of a pharmaceutical packaging off-odor using solid phase microextraction gas chromatography/mass spectrometry. J. Pharm. Biomed. Anal. 25, 379–386. Simon, M., and Hansen, A. P. (2001). Effect of various dairy packaging materials on the shelf life and flavor of ultrapasteurized milk. J. Dairy Sci. 84, 784–791. Soderhjelm, L. and Eskelinen, S. (1985). Characterization of packaging materials with respect to taint and odour. Appita 38(3), 205–209. Song, Y. S., Al-Taher, F., and Sadler, G. (2003). Migration of volatile degradation products into ozonated water from plastic packaging materials. Food Addit. Contam. 20, 985–994. Suloff, E. C. (2001). Modeling the adsorption of aldehydes and ketones to novel poly(ethylene terephthalate) blends by inverse gas chromatography to develop an improved polymer for food packaging. Ph.D. Dissertation, Virginia Tech, Blacksburg.
Sensory Impacts of Food–Packaging Interactions
63
Suloff, E. C., Marcy, J. E., Blakistone, B. A., Duncan, S. E., Long, T. E., and O’Keefe, S. F. (2003). Sorption behavior of selected aldehyde-scavenging agents in poly(ethylene terephthalate) blends. J. Food Sci. 68(6), 2024–2033. Suppakul, P., Sonneveld, K., Bigger, S. W., and Miltz, J. (2008). Efficacy of polyethylenebased antimicrobial films containing principal constituents of basil. Food Sci. Techol. 41(5), 779–788. Tomboulian, P., Schweitzer, L., Mullin, K., Wilson, J., and Khiari, D. (2002). Materials used in distribution systems—Contributions to taste-and-odor. Water Sci. Technol. 49(Pt. 2), 219–226. Tomboulian, P., Schweitzer, L., Mullin, K., Wilson, J., and Khiari, D. (2004). Materials used in drinking water distribution systems: Contribution to taste and odor. Water Sci. Technol. 49, 219–226. Torri, L., Piergoiovanni, L., and Caldiroli, E. (2008). Odour investigation of granular polyolefins for flexible food packaging using a sensory panel and an electronic nose. Food Addit. Contam. 25(4), 490–502. van Aardt, M., Duncan, S. E., Bourne, D., Marcy, J. E., Long, T. E., Hackney, C. R., and Heisey, C. (2001a). Flavor threshold for acetaldehyde in milk, chocolate milk, and spring water using solid phase micro-extraction gas chromatography (SPME-GC) for quantification. J. Agric. Food Chem. 49, 1377–1381. van Aardt, M., Duncan, S. E., Marcy, J. E., Long, T. E., and Hackney, C. R. (2001b). Effectiveness of poly(ethylene terephthalate) and high density poly(ethylene) in protection of milk flavor. J. Dairy Sci. 84(6), 1341–1347. van Aardt, M., Duncan, S. E., Marcy, J. E., Long, T. E., O’Keefe, S. F., and Nielsen-Sims, S. R. (2005a). Aroma analysis of light-exposed milk stored with and without natural and synthetic antioxidants. J. Dairy Sci. 88, 881–890. van Aardt, M., Duncan, S. E., Marcy, J. E., Long, T. E., O’Keefe, S. F., and Nielsen-Sims, S. R. (2005b). Effect of antioxidant (a-tocopherol and ascorbic acid) fortification on lightinduced flavor of milk. J. Dairy Sci. 88, 872–880. van Aardt, M., Duncan, S. E., Marcy, J. E., Long, T. E., O’Keefe, S. F., and Sims, S. R. (2007). Release of antioxidants from poly(lactide-co-glycolide) films into dry milk products and food simulating liquids. Int. J. Food Sci. Technol. 42, 1327–1337. van Willige, R. W. G. (2002). Effects of flavour absorption of foods and their packaging materials. Ph.D. thesis. Wageningen University, The Netherlands. (In Linssen, J.P.H., van Willige, R. W. G., and Dekker, M. Packaging-flavour interactions. Ch. 8. In ‘‘Novel Food Packaging Techniques’’ (R. Ahvenainen, ed.). Woodhead Publishing Ltd., CRC Press, Boca Raton, FL. pp. 144–171). van Willige, R., Linssen, J., and Voragen, A. (2000). Influence of food matrix on absorption of flavour compounds by low-density polyethylene: Oil and real food products. J. Sci. Food Agric. 80, 1790–1797. van Willige, R. W. G., Linssen, J. P. H., Legger-Huysman, A., and Voragen, A. G. J. (2003). Influence of flavour absorption by food-packaging materials (low-density polyethylene, polycarbonate and polyethylene terephthalate) on taste perception of a model solution and orange juice. Food Addit. Contam. 20(1), 84–91. Villberg, K., Veijanen, A., Gustafsson, I., and Wickstrom, K. (1997). Analysis of odour and taste problems in high-density polyethene. J. Chromatogr. 791, 213–219. Webster, J. B. (2006). Changes in aromatic chemistry and sensory quality of milk due to light wavelength. Ph.D. Dissertation. Virginia Tech, Blacksburg, VA. Webster, J. B., Duncan, S. E., Marcy, J. E., O’Keefe, S. F., and Sims, S. N. (2007). Packaging solutions for sensory degradation in foods and beverages due to photooxidation. Abstract POLY 685. Am. Chem. Soc. Annual Meeting, Boston, MA. Weinberg, H. S., Glaze, W. H., Krasner, S. W., and Sclimenti, M. J. (1993). Formation and removal of aldehydes in plants that use ozonation. J. AWWA 85, 72–85.
64
Susan E. Duncan and Janet B. Webster
Whitfield, F. B., Tindale, C. R., Shaw, K. J., and Stanley, G. (1984). Contamination of cocoa powder by chlorophenols and chloroanisoles adsorbed form packaging materials. Chem. Ind. 772–774. Yam, K. L., Yo, Y. C., Young, S. S., and Zambetti, P. F. (1996). Effect of resin types and antioxidants on release of off-flavor from HDPE bottles. Polym. Plastics Technol. Eng. 35(5), 727–755. Zygoura, P., Moyssiadi, T., Badeka, A., Kondyli, E., Savvaidis, I., and Kontominas, M. G. (2004). Shelf life of whole pasteurized milk in Greece: Effect of packaging material. Food Chem. 87, 1–9.
CHAPTER
3 Developmental Trajectories in Food Allergy: A Review A. DunnGalvin and J’ O. B. Hourihane
Contents
I. Background A. Transition points: neurocognitive development B. Research with children II. Prevalence, Mechanisms, and Clinical Manifestations of Food Allergy III. The Impact of Food Allergy on HRQL A. Research using generic HRQL measures B. HRQL Research using disease-specific measures IV. The Psychological Burden of Food Allergy V. The Influence of Parents on Child Adjustment VI. Social Support VII. The Impact of Stress on Biopsychosocial Development VIII. The Impact of Sex and Gender in Food Allergy A. The influence of sex B. The influence of gender IX. Risk Behavior in Food Allergy X. Developmental Pathways in Food Allergy A. Transition points in development for food allergic children B. Parental views and management of food allergy XI. Discussion and Implications for Future Research References
66 67 68 68 73 73 74 76 78 79 80 81 82 83 84 86 90 91 92 95
Department of Paediatrics and Child Health, Cork University Hospital, Ireland Advances in Food and Nutrition Research, Volume 56 ISSN 1043-4526, DOI: 10.1016/S1043-4526(08)00603-7
#
2009 Elsevier Inc. All rights reserved.
65
66
Abstract
A. DunnGalvin and J’ O. B. Hourihane
Increasing recognition of the importance of the relationships between perceptions, emotions, behaviors and health has changed the way health and disease are portrayed and researched. A chronic condition may affect and/or interact with already existing normative demands and changes in socialization. Although the prevalence of food allergy and anaphylaxis have been reportedly increasing, the emotional and social impact of growing up with food allergy has received little emphasis. In this paper, we present current findings on the biopsychosocial impact of food allergy on children in order to gain insight into the food allergy experience, from the perspective of the child, teen, and parent living with food allergy, with particular attention to developmental aspects. Due to the scarcity of publications on the psychosocial dimensions of food allergy, we also draw on selected literature on children’s and parent’s experience of, and coping with chronic disease that may inform research into food allergy. To this end, we review some general developmental mechanisms that may underpin and explain normative age-graded shifts in patterns of coping across childhood and adolescence. We also highlight gaps in the literature and assess implications of current research in food allergy and other chronic diseases for intervention and prevention of negative short and long term outcomes.
I. BACKGROUND The growing prevalence of allergic diseases present an increasing challenge for populations and health care systems around the world and food allergies constitute a notable part of this increase (Sicherer, 2002). The emotional and social aspects of food allergy have not received much attention to date although an association between allergy and anxiety disorders in children and adolescents have been found to persist into adulthood (Katon et al., 2004a,b), and evidence linking psychological stress to the expression of asthma and atopy continues to grow (Wright, 2005). In child/adolescent populations with asthma, up to one third may meet criteria for comorbid anxiety disorders (Bender-Berz et al., 2005; Ortega et al., 2002). Thoughts, feelings, and behaviors affect our health and well-being. Recognition of the importance of these influences on health and disease is consistent with evolving conceptions of mind and body and represents a significant change in medicine and the life sciences. Recent developments include the idea that emotional processes, such as stress, moderate activity in nearly all systems of the body and can directly influence the pathophysiology of disease. Discovery of these and other relationships
Developmental Trajectories in Food Allergy
67
between behavior and health has changed the way health and disease are portrayed. This interest is also reflected in the rapid development of health psychology and its more multidisciplinary cousin, behavioral medicine. These fields grew rapidly in the 1980s and now constitute major endeavors in most university and medical center settings (Barlow et al., 2002a). The movement coincides with the growing recognition of the importance of children’s views of their experiences that has begun to permeate many areas of research (Hill, 2006). Methods for research with children are, however, relatively underdeveloped, emphasizing the need to develop and document methods for research with children (Hennessy and Heary, 2004). Within the basic and clinical scientific community, there is increasing recognition that developmental trajectory frameworks offer a conceptual model for health development and a more powerful approach to understanding diseases. A developmental trajectory or pathway may be understood as a lifelong process of developmental integration that involves complex interactions between biological and environmental factors that influence the phenotypic expression of physiology, psychology, and behavior (Halfon and Hochstein, 2002). The health of children is a product of complex, dynamic processes produced by the interaction of external influences, such as children’s family, social, and physical environments, and their genes, biology, and behaviors. Because children are rapidly changing and developing in response to these interactions, the developmental process plays an important role in shaping and determining their health.
A. Transition points: neurocognitive development Developmental pathway models can take account of the cumulative and interactive contribution of physiological and environmental variables. They may also delineate sensitive or transition points in development when physiological or environmental variables associated with a chronic condition may have a relatively greater impact and/or interact with already existing normative demands and changes in socialization (Halfon and Hochstein, 2002). One key transition point occurs at between 4 and 7 years in most cultures and involves entry into formal education. Children’s social networks start to change from networks in which children primarily interact with adults to networks in which children primarily interact with other children, with consequent exposure to social comparison and competition in school classrooms and peer groups (Dixon and Stein, 2006). The clinical literature suggests that children’s patterns of social functioning in middle childhood is predictive of relationship patterns at later points in development (Gifford-Smith
68
A. DunnGalvin and J’ O. B. Hourihane
and Browning, 2003). Furthermore, there is evidence of shifts in cognitive development in which enhanced memory, new reasoning abilities, and new strategies for recall emerge (Dixon and Stein, 2006). Adolescence marks a similar transition point. Growth of primarily physiologically based processes, such as attention, perception, and information processing, provide the foundation for important social and emotional changes that occur during these years that contribute to children’s growing sense of identity and self-esteem (Harter, 1997). Research in chronic disease in childhood may be particularly apposite, as children who grow up with a chronic illness not only have to meet their age-related developmental tasks, but they also have to manage their disease, which leads to a heightened risk of maladaptation (Hill, 2006). A chronic condition may affect and/or interact with already existing normative demands and changes in socialization (Schmidt, 2003). Thus, although most children follow normative developmental pathways and encounter predictable transition points, disease-specific pathways may be embedded within these trajectories and influence the phenotypic expression of physiology, psychology, and behavior.
B. Research with children Children are increasingly acknowledged to have rights in the determination of medical decisions that affect them. This has encouraged research to be undertaken with children themselves to understand their own views on the impact of a disease on their experiences and relationships. It has become increasingly important for researchers and healthcare professionals to understand how the perceptions, experience, and impact of a chronic disease might influence a patient’s interpretation and response to it, so that we, in turn can respond more appropriately. Related to this, the role of psycho-educational interventions in facilitating adaptation to chronic disease has received growing recognition and is in keeping with policy developments advocating greater involvement of patients in their own care (Barlow et al., 2002a)
II. PREVALENCE, MECHANISMS, AND CLINICAL MANIFESTATIONS OF FOOD ALLERGY Atopy may be defined as a genetically and environmentally determined predisposition to clinically expressed disorders, including allergic rhinitis, atopic dermatitis or eczema, food allergy, and allergic asthma, regulated through immune phenomena in which many cells (i.e., mast cells, eosinophils, and T lymphoctytes) and associated cytokines, chemokines, and neuropeptides play a role.
Developmental Trajectories in Food Allergy
69
Atopy identifies allergic diseases such as atopic dermatitis (eczema), bronchial asthma, and hayfever, which tend to cluster in families and are associated with the production of specific IgE antibodies to common environmental allergens (Sicherer, 2002). The ‘‘atopic march’ refers to the natural progression of allergic conditions, characterized by a typical pattern of sensitization and manifestation of symptoms appearing at certain ages, persisting over years or decades, but often spontaneously resolving with age (Sampson, 2003). Epidemiological studies have attempted to disentangle the various phenotypes, focusing on single manifestations at certain age windows, since different specific phenotypes may be induced or modulated by different genetic, environmental, or lifestyle factors. Most investigators seem to agree that a complex interaction between genetic and environmental factors regulates development of different atopic features, however much of the natural history of atopic diseases and its determinants are still not well understood ( Johannson et al., 2004). Food allergy is defined as an adverse immune response to food allergens. The growing prevalence of allergic diseases present an increasing challenge for populations and health care systems around the world and food allergies constitute a notable part of this increase (Sicherer 2002; Sicherer et al., 2001; von Berg et al., 2003). The most common food allergens are peanuts, tree nuts, seafood, eggs, and milk, however, the list is constantly growing (Sampson, 2003). Several studies have confirmed the increase in prevalence of food allergies, especially peanut allergy and as a disease burden throughout the world; however, they seem to increasingly affect countries with a formerly low prevalence and is becoming a growing public health problem (Hourihane, Smith and Strobez, 2002). These findings were substantiated by Al-Muhsen and colleagues (2003) suggesting that peanut allergy now accounts for the majority of severe foodrelated allergic reactions. There appears to be a correlation between the increased consumption of a novel food and the risk of allergic reactions. Examples of foods introduced into the North American diet which consequently began to provoke allergic reactions as consumption increased include kiwi, mango, avocado, and other exotic fruits (Burks, 2006; Sampson, 2003). Food allergy affects 6–8% of young children and 3–4% of young adults in the UK, US, and Europe (Eggesbo et al., 2001; Sampson, 2005; Sicherer and Sampson, 2006). In contrast, food intolerance describes an abnormal physiological response to an agent which is nonimmunemediated ( Johansson et al., 2004). The prevalence of food allergies vary between countries. The only comprehensive permanent nationwide reporting system and register for severe allergic reactions to food was instituted in Norway in 2000 (Lovik et al., 2003), and so more quantitative estimates are difficult to provide. In Ireland, Hourihane (1998) estimates
70
A. DunnGalvin and J’ O. B. Hourihane
that there are 16,000 children with peanut allergy; however, there has been no formal prevalence study undertaken to date. In addition, current data on severe and fatal reactions may be misleading because anaphylactic reactions are often mislabeled as asthma deaths, because of a lack of antecedent history or information. In effect, inadequate diagnosis of food allergy may, in part, reflect the lack of adequate provision of relevant health services. The European Academy of Allergy and Clinical Immunology has proposed a revised nomenclature for allergic and related reactions ( Johansson et al., 2004). According to this proposal adverse reactions to food should be termed ‘‘food hypersensitivity.’’ The term food allergy should be used when immunological mechanisms have been demonstrated, and includes both IgE- and non-IgE-mediated reactions. All other reactions, which have sometimes been referred to as ‘‘food intolerance,’’ should be termed nonallergic food hypersensitivity (Fig. 3.1). Adverse reactions to foods were first described over 2000 years ago by Hippocrates, who is credited with the observation that cow’s milk could cause urticaria and gastric upset, and, 500 years later Galen also described a case of allergy to cow’s milk (Kimber and Dearman, 2002). Adverse reactions to foods were published intermittently during the twentieth century, but it is only during the last 20–30 years that an increasing awareness of food allergy has emerged in western industrialized societies. Food allergy occurs when the body’s immune system mounts an exaggerated response against the offending food, which acts as an allergen. It is a type of hypersensitivity reaction. It can be either: A type I, IgE-mediated reaction: This is the usual cause of food allergy.
After initial sensitization, the release of mediators such as histamine are triggered each time a person is exposed to the food. It is these mediators that cause the symptoms.
Food hypersensitivity
Food allergy
Non-allergic hypersensitivity
IgE-mediated food allergy Non-IgE-mediated food allergy
FIGURE 3.1
Continued
Developmental Trajectories in Food Allergy
Typical developmental pathway
Food allergy developmental pathway
Infancy and early childhood
Entry to formal schooling
71
Parental protection
Transition point
Precipitating events (stressful events in the children’s lives caused by developmental & food allergy related factors)
Cognitive appraisal and emotional effects
Biology • •
Living with uncertainty
Sex FAMechanisms
Search for normality
Environment Distal factors (socio-econom ic, education, culture) Integration of food allergic identity
Rejection of food allergic identity
Proximal factors • • • • • • •
Gender Age Context Prior experience Parent attitude Social support (parent, teacher, peers, friends) General awareness •
Functions (Interpers onal & Intrapersonal) • • • • • • •
•
Risk & safety Eating & food Control Self-esteem Identity Interaction with peers Emotional regulation Transition point
Principal strategies Maximisation/avoidance Supporting strategies
• •
Principal strategies Minimisation/risk Supporting strategies
Precipitating events (stressful events in the teens lives caused by developmental & food allergy related factors)
Adolescence Functions (Interpers onal & Intrapersonal) Habitual cognitive/emotional appraisal mechanisms • • • • • • • • • • •
Risk &safety Eating & food Control Self-esteem Reinforce identity New places New people Autonomy Assert independence Validation of beliefs Emotional regulation
Habitual response patterns (physiological, emotional, behavioural)
Longterm psychological impact
FIGURE 3.1 Nomenclature proposed by the European Academy of Allergology and Clinical Immunology (Johansson et al., 2004). The developmental pathway in food allergy. Our developmental model outlines the content of stressors (e.g., social events, restaurants, food, allergic reactions, concern for risk and safety, concern for identity); modifying variables on appraisal (e.g., attitudes of others, low general awareness, poor labeling, age, sex, context; parental attitude); specific resources (e.g., types of social support; restaurants; labelling); emotional impact (uncertainty, difference); psychological impact (e.g., generalized anxiety, low self-efficacy); functions (e.g., reducing uncertainty; ‘‘fitting-in’’) and consequences for behavior and participation (e.g., principal, supporting).
A delayed, type IV-mediated reaction: These reactions are mediated
mainly by T cells. They typically affect the gastrointestinal tract or skin, for example, exacerbation of eczema in children after milk ingestion.
72
A. DunnGalvin and J’ O. B. Hourihane
In an IgE-mediated reaction, symptoms involving the oropharynx and gastrointestinal tract may occur within minutes of ingesting a food allergen. Itching and swelling of the lips, tongue, and soft palate as well as nausea, abdominal pain, vomiting, and diarrhea have all been demonstrated secondary to food allergy (Sampson, 2003). Anaphylaxis refers to a sudden, severe, potentially fatal, systemic allergic reaction that can involve skin, respiratory tract, gastrointestinal tract, and cardiovascular system (Sampson, 2003). The term anaphylaxis is derived from the Greek words meaning ‘‘without’’ (ana) and ‘‘protection’’ (phylaxis). The most dangerous symptoms include breathing difficulties and a drop in blood pressure, or shock, which are potentially fatal. Symptoms of anaphylaxis may develop within seconds or a few hours after ingestion of a food allergen, with the vast majority of reactions developing in the first hour. Symptoms can include swelling (especially lips, tongue, or throat), difficulty breathing, abdominal cramps, vomiting, diarrhea, circulatory collapse, coma, and death. Typical allergy medications such as antihistamines work too slowly and cannot reverse the effects of chemical mediators. Adrenaline or epinephrine, therefore is the treatment of choice and must be administered by injection promptly. Food allergy, particularly to peanuts, is the most common cause of anaphylaxis outside hospital (Bock et al., 2001), yet there are other common food causes such as shellfish, fish, milk, soy, wheat, and eggs (Asero et al., 2007). These foods may not only cause fatal or near-fatal reactions, and they also tend to induce ‘‘persistent sensitivity’’ in most patients, in contrast to other foods such as milk, eggs, and soybeans, which are frequently associated with milder reactions and are usually ‘‘outgrown.’’ The life-threatening nature of anaphylaxis ‘‘makes prevention the cornerstone of therapy,’’ (Sampson, 2003). Avoidance of the responsible food allergen and emergency management in the form of injectable epinephrine (Epipen, Anapen, or Twinjet) in case food allergen is accidentally ingested is the only reliable therapy offered to those living with food allergy. Anticipatory guidance measures form the cornerstone of advice, including; reading food ingredient labels, concern for crosscontamination, vigilance in a variety of social activities, and immediate access to the Epipen (Munoz-Furlong, 2003). However, avoidance is complicated by the fact that peanuts, nuts, soy, can be found in many foods (e.g., breads, muffins, pastries, biscuits, cereals, soups, icecreams, seasoning, sauces) and in different forms as an emulsifier or thickening agent. A growing number of families must live and cope with food allergy on a day to day basis, socio-emotional impact of food allergy on children, and adolescents has been little researched to date (DunnGalvin et al., 2007). Although researchers in the field of food allergy have stated that the scarcity in psychological and social literature on the experience of
Developmental Trajectories in Food Allergy
73
living with food allergy demands attention, the majority of research in food allergy has been biomedical in orientation, focusing on issues such as the molecular structure of allergens, or methods of diagnosis (see DunnGalvin et al., 2007). More recently, there has a growing interest in the development of questionnaires to measure the impact of food allergy on health-related quality of life (De Blok et al., 2007). Health related quality of life (HRQL) is a multidimensional construct, consisting of physical, psychological, and social components (Eiser and Morse, 2001).
III. THE IMPACT OF FOOD ALLERGY ON HRQL Most studies to date used generic HRQL questionnaires to investigate the impact of food allergy on quality of life. However, a disease-specific questionnaire is necessary because generic measures although useful for comparison across diseases, do not incorporate issues directly related to the patients condition and therefore lack the sensitivity necessary to detect changes as a result of treatment interventions (Fayers and Machin, 2000). Further generic measures lack the detail that is required to assess how different subgroups are impacted by food allergy (De Blok et al., 2007).
A. Research using generic HRQL measures Primeau and colleagues (2001) studied a sample of 301 patients and evaluated the quality of life and family relations of children and adults with peanut allergy, compared to that of children and adults with rheumatological disease. Their study was the first on the subject and they compared HRQL in rheumatology and food allergy patients. It was shown that the parents of allergic children believed that their children had difficulties in many areas. Remarkably, their children had more impairment, especially in daily activities and in their familial social interactions, compared to children with significant rheumatological disease. The authors found that families with peanut allergic children experience significantly more disruption in their familial and social interactions/ activities than families with a child with CRD and suggested that this may be due to the constant risk of sudden death in the peanut allergy group leading to greater parent restriction of activities. The comparison was criticized by Avery et al. (2003) because children with rheumatological diseases have other symptoms to include severe pain and chronic physical disabilities, which exceeds experiences of children with peanut allergy. Primeau et al. (2000), at the time of their study, recognized and acknowledged this comparison as a potential
74
A. DunnGalvin and J’ O. B. Hourihane
compromise. They admitted that their choice was influenced by the accessibility to available data. Sicherer et al. (2001) measured the parental perception of physical and psychological functioning. The authors randomly selected 400 members of the food and anaphylaxis network, with families of children aged 5–18 years old and had 253 responses. Results indicated that peanut allergy impacted significantly on general health, parental distress, and family activities. Those with two or more food allergies scored significantly lower, depending on how many foods they were avoiding. There was also evidence to suggest that the educational and emotional support needs of these families are not being met. Avery et al. (2003) assessed the effect of peanut allergy on the quality of life in children aged 7–12 years old and contrasted this with experiences of children with insulin-dependant diabetes mellitus (IDDM). They recruited 20 children with peanut allergy and 20 children with IDDM (ages ranging from 7 to 12 years). Their results indicated that children with IDDM have similar problems as children with peanut allergy. This includes, food choices, social restriction, issues relating to school, the carriage and use of a syringe and the chronic nature of the condition. The results also suggested that children with peanut allergy are more anxious and parents feel that their needs are not taken into account. Results showed that children with peanut allergy had a poorer quality of life and are more anxious concerning accidental ingestion of peanut than children with diabetes are of having a hypoglycemic reaction. Another generic questionnaire-based study, among 1451 adolescents, indicated that 19% of the participants reported that they had a perceived adverse reaction to food, however their condition was not physiciandiagnosed (Marklund et al., 2004). When compared to adolescents without such conditions, this kind of allergy-like condition, regardless of the underlying mechanisms, was associated with lower HRQL.
B. HRQL Research using disease-specific measures The first validated HRQL food allergy specific measure; the Food Allergy Quality of Life–Parental Burden (FAQL-PB) questionnaire (Cohen et al., 2004) measures the parental burden associated with having a child with food allergy. Scores in the food-allergic cohort were significantly lower for general health perception, parental distress and worry, and interruptions and limitations in usual family activities, than in healthy controls. Scales were also lower in subjects with multiple food-allergies. More recently, several measures have been developed to assess quality of life in parents, children and teens, propelled by Europrevall. Europrevall has initiated/propelled a leap in research.
Developmental Trajectories in Food Allergy
75
Europrevall is Europrevall is an EU project which aims to improve quality of life for parents, children, teenagers and adults with food allergy (De Blok et al., 2007; Dunn Galvin et al., 2007). Europrevall multidisciplinary integrated project (IP) involving 17 European member-states, Switzerland, Iceland, and Ghana. Of the 63 partners, there are 15 clinical organisations and six small-medium sized enterprises (SMEs) as well as the leading allergy research organisations in Europe. Since the project began in 2005, new partners have also joined from New Zealand, Australia, Russia, India, and China. By integrating information and developing tools for the use of European food allergy scientists, health professionals, food and biotech industries, and consumers it is Europrevall’s hope that causes of food allergy can become better understood, diagnosis of food allergy can become swifter and the quality of life of food allergy sufferers improved. Two food allergy specific questionnaires have been developed and published under the auspices of Europrevall. The first measures HRQL in children aged 0–12 years and is parent administered; ‘‘Food Allergy Quality of Life’’ questionnaire (FAQLQ-PF; DunnGalvin et al., 2008) and the second measures quality of life in teens (FAQLQ-TF; Flokstra-DeBlok et al., 2008). The FAQLQ-PF and -TF were developed and validated in four stages: (1) item generation using focus groups with both children and parents, expert opinion, and literature review; (2) item reduction, using clinical impact and factor analysis; (3) internal and test-retest reliability and construct validity were evaluated; and (4) cross-cultural and content validity was examined by administering the questionnaire in a US sample (FAQLQ-PF, only). Both studies found a severe impact of food allergy on HRQL in relation to psychosocial aspects of children’s and teens everyday lives. For example, in the initial focus groups put in place to generate items for the FAQLQ-PF, parents suggested that the anxiety associated with the risk of a potential reaction has more profound effects on emotional and social aspects of a child’s everyday life, than clinical reactivity induced by food intake. The importance of a subscale assessing this aspect of anxiety was subsequently confirmed using clinical impact and factor analytic methodologies. In addition, multivariate analysis showed an interaction between sex and age group for impact of general emotional impact on HRQL scale, in effect, parents of boys reported higher mean scores up to the age of 6 years; parents of girls reported higher mean scores in the 6–12 years age group. Studies using these measures have thrown light on the everyday burden of living with food allergy.
76
A. DunnGalvin and J’ O. B. Hourihane
It has therefore been well established that food allergy can have a profoundly negative impact on quality of life, extending well beyond the immediate clinical effects of the individuals condition. However, very little is known about the attitudes, beliefs, and coping strategies of food allergic children about food allergy generally and their condition specifically and little attention has been paid to the most effective methods of communication with this group (Miles et al., 2006).
IV. THE PSYCHOLOGICAL BURDEN OF FOOD ALLERGY Although it appears that food allergy leads to intrapersonal (e.g., anxiety) and interpersonal (e.g., social restrictions) problems in adaptation, there has been little research into the socio-emotional impact of food allergy on psychological and social functioning. Indeed, in most studies on chronic diseases, usually psychological maladjustment factors such as behavior problems or depression are studied, but social maladjustment factors, such as social anxiety or social skills, have rarely been included (Meijer et al., 2002). The literature also lacks well-designed studies describing the psychological burden of food allergy. Studies are scarce and are often carried out in a mixture of diagnoses, reflecting the difficult field of adverse reactions to food. Many of the studies that have been undertaken are limited because they rely only on questionnaires, on the particular questionnaire selected, and on prestudy or preexisting hypotheses which limits novel findings. Questionnaire studies produce findings on ‘‘how often’’ or ‘‘how many’’ but are very limited in answering ‘‘why.’’ For example, patients attending allergy clinics were found to have higher levels of depression when compared with the general population, (e.g., Kova’cs et al., 2003). In a community sample, Knibb et al. (1999) found that women with a perceived food intolerance or allergy have significantly higher scores for somatic symptoms, anxiety and insomnia, and severe depression than women with no reported food allergy or intolerance. Vatn (1997), found that patients identifying themselves as sensitive to food or chemicals had high scores for depression, anxiety, shyness, and defensiveness. Coping has been defined as ‘‘action regulation under stress’’ and refers to how individuals ‘‘mobilize, guide, manage, energize, direct behavior, emotion, and attention and how they fail to do to’’ (Skinner and Wellborn, 1994, p. 113) under stressful conditions. However, it appears that literature on adjustment or coping in chronic illness focuses mainly on coping in illness-specific situations. However, general coping styles may be more predictive for the psychosocial development of chronically ill children and teens than illness-related coping
Developmental Trajectories in Food Allergy
77
because these styles reflect how children cope with developmental tasks (Meijer et al., 2002). General coping styles (e.g., in social situations) may be particularly relevant in food allergy because it is the anxiety associated with the risk of a potential reaction that has more profound effects on emotional and social aspects of a child’s everyday life, than clinical reactivity induced by food intake (DunnGalvin et al., 2008). In addition, food allergy, once diagnosed and after restrictions are put in place, may be primarily asymptomatic. Living with peanut allergy places increased stress on the child and the child’s parents and siblings (King et al., 2008). It also causes differing levels of anxiety throughout the family. The authors found that mothers of children with peanut allergy feel that they have significantly poorer HRQL and suffer more anxiety and stress than the father. However, clinicians assume that ‘‘appropriate’’ levels of anxiety is adaptive in children and parents living with food allergy. Primeau et al., 2000 and Avery et al., 2003 suggest that the high level of stress in families with a peanut allergic child may have beneficial effects on coping strategies. Only one study, using a study-specific questionnaire, suggested that deprivation due to restrictions in lifestyle may lead to social anxiety in food allergy (Bollinger et al., 2006). Adaptational processes of children and adolescents with chronic conditions are of utmost importance because of the long-term consequences of childhood conditions. For example, in child/adolescent populations with asthma, up to one third may meet criteria for comorbid anxiety disorders (Bender-Berz et al., 2005). In adult populations with asthma, the estimated rate of panic disorder ranges from 6.5% to 24% (Katon et al., 2004a,b). Children with any chronic condition have twice the risk of developing mental health disorders of healthy children without an accompanying physical disability (Schmidt, 2003). A recent paper (Knibb et al., 2008) may provide another clue to the impact of living with food allergy on long term adjustment. The authors used the Illness Perception Questionnaire (Moss-Morris et al., 2002) to measure the extent to which illness perceptions and coping strategies are associated with levels of psychological distress among 156 adults with food allergy. Results showed that a strong illness identity and emotional representations of food allergy in adults were associated with higher levels of psychological distress; as were less adaptive coping strategies such as focusing on and venting of emotions. Strong personal control beliefs were associated with the lower levels of distress, as were adaptive coping strategies such as positive reinterpretation and growth. Coping skills partially mediated the link between the illness perceptions and the outcome; however, illness identity, emotional representations, and personal control retained an independent significant association with psychological distress.
78
A. DunnGalvin and J’ O. B. Hourihane
V. THE INFLUENCE OF PARENTS ON CHILD ADJUSTMENT Parental perceptions can have a profound impact on the way that children themselves perceive their own health and illness and on how they interpret risk associated the disease (Noojin and Wallander, 1997). Although the impact of any condition on a child’s life varies according to the specific characteristics of the disease, the chronically ill child is often more dependent on parents with adolescents complaining more often a delay in independent life-styles (Wallander and Varni, 1998). An important developmental issue relates to the fact that children and adolescents are dependants, largely reliant on adults for significant aspects of illness behavior. This behavior includes the way in which symptoms are responded to, including the extent to which medical consultation and lifestyle alterations are undertaken. Parents and carers are powerful in responding to (or ignoring) children’s physical complaints, attributing significance (or reassurance) to these complaints, facilitating (or otherwise) the children’s use of health care facilities and their involvement in (or withdrawal from) normal life activities (Noojin and Wallander, 1997). Thus, parental distress, response to diagnosis and consequent coping strategies will have implications for how children themselves cope and manage food allergy along developmental trajectories. We could only find one published study that looked at the impact of parent influence in food allergy. Recent findings by Bollinger et al. (2006, p. 419) suggests that ‘‘children with food allergy may be at an increased risk of social-emotional developmental difficulties.’’ Many parents admitted to overprotecting their children through an understandable desire to ensure their children’s safety. However, such restrictions can stunt children’s social and emotional development and increase children’s perception of ‘‘illness intrusiveness.’’ Keating and Miller (2005) discuss research findings on infant distress which suggests two socioemotional roots of competence, with possible interactions between them: age-specific emotion regulation style; and parental response and facilitation. They use the term ‘‘habits of mind’’ to describe the increasing coordination and integration of the competence and regulatory system trajectories over the course of development. Our research group have carried out a study (submitted) on parent stress triggers using a specially designed Implicit Association Test (IAT). The IAT measures associative links between environmental triggers and psychological states. Explicit or ‘‘self-report’’ questionnaires have given rise to concerns related to the influence of conscious self-presentational biases which often result in inaccurate answers (Greenwald et al., 1998; Nosek, 2007). In addition, the very act of completing the self-report measure may change emotions attitudes and beliefs about the construct
Developmental Trajectories in Food Allergy
79
under investigation (Greenwald et al., 1998). Therefore, the IAT was used to measure the differential association of a target category (environmental stimuli) with an attribute dimension (anxious vs. relaxed) in parents of food allergic children (N ¼ 60) and matched controls (N ¼ 30). A study specific questionnaire (SSQ) was used to measure explicit attitudes. The IAT targets the specific concerns/attitudes/beliefs parents of food allergic children while minimizing confounders. In contrast, the perceived threat of social situations where food may or may not be present (a central concern of parents of food allergic children) is incomprehensible to parents of nonfood allergic children. Results showed that parents are significantly more anxious when they believe they are not in control of their child’s environment. In contrast, there were no significant results with regard to parents of nonfood allergic children. It is clear therefore that parents face a difficult balancing act between encouraging growing children’s independence and ensuring their safety that may have implications for children’s own perception of control in living and coping with food allergy.
VI. SOCIAL SUPPORT Not only interpersonal functioning, but also interpersonal needs of children with chronic conditions differ from those of healthy children. Social support is understood as a ‘‘resistance’’ or ‘‘buffering’’ factor in chronic diseases. Studies with children have shown that higher stress increases risk for adjustment problems, such as greater anxiety and depressive symptoms, while higher social support reduces the risk for adjustment problems (e.g., Wallander and Varni, 1998). Although many studies aggregate social support variables, the sources and types of available to children may differentially predict psychosocial adjustment (Schreurs and de Ridder, 1997). In addition, classmate support (contrasted with parent, friend, and teacher support) was found to be the strongest predictor of adjustment in children with limb deficiencies (Varni et al., 1991), cancer (Varni et al., 1994a,b,c), and pediatric rheumatic disease (Von Reiss et al., 2002). However, other studies have found that family social support, but not peer support, was a significant predictor of adjustment in children juvenile rheumatoid arthritis (Varni et al., 1988). The diverse findings may occur because social support needs differ as a function of the particular disease. However, given the sparse literature on support in chronic disease in children in general, and food allergy in particular, further research is needed to examine the relative importance of different sources and types of support on the adjustment of children. For example, in a recent qualitative study into children’s, teens, and parents perceptions of living with food allergy (DunnGalvin et al., 2009),
80
A. DunnGalvin and J’ O. B. Hourihane
we found that social support, per se, does not necessarily protect against emotional distress, but rather how the individual perceives and interprets his or her social network. In children with generalized avoidance strategies, friends or parents may actually reinforce children’s beliefs of a generally unfriendly world. Our findings show that parental distress, threat perception, and coping strategies reflect how children themselves respond to and manage food allergy. Parental influence and beliefs may have greatest impact during sensitive (psychological) or critical (physiological) periods of development when children and adolescents are most vulnerable, for example, during middle childhood and/or adolescence. This type of specificity permits precise empirical guidance in the development of our educational interventions for children, teens, and parents with food allergy. In addition, we found that different dimensions of social support may reflect different dimensions of adjustment, for example, children who spoke about classmate support or teacher support appeared more in control and much less anxious about food allergy, compared to those who referred only to support from parents and close friends. Those who did not seek (or receive) instrumental or emotional support from friends or teachers appeared to engage in much riskier behavior. These differential relationships underline what Varni (1994b, p. 35) refers to as the ‘‘specificity of the relationship between resistance factors and outcomes.’’
VII. THE IMPACT OF STRESS ON BIOPSYCHOSOCIAL DEVELOPMENT From a physiological viewpoint on development, we find that environmental stress may be important in perinatal programming (Wright, 2005). Studies in rodents and primates have shown that environmental manipulations that increase maternal stress result in elevated cortisol levels and dysfunctional behaviors in offspring that are mediated, in part, through effects on gene expression (Meaney and Szyf, 2005). Wright and colleagues (2005) found that higher caregiver stress in the first 6 months after birth was associated with increases in the children’s allergen-specific proliferative response (a marker of the allergic immune response) higher total IgE levels, and increased production of TNF-a and reduced IFN-g in a birth cohort of children predisposed to atopy. Therefore, during T cell maturation when the atopic phenotype is being determined by exposure to allergens, stress may be an additional factor. During an immune response, the brain and the immune system communicate with each other in order to maintain homeostasis (Knonfol and Remick, 2000). Two major pathways, the HPA-axis and the CNS are involved in this bi-directional interaction (Elenkov et al., 2000). The effects
Developmental Trajectories in Food Allergy
81
of stress on neuroimmuregulation in turn may modulate the hypersensitivity response in developing children. Cytokines play a crucial role in the pathogenesis of allergic diseases. In addition to acting as chemical messengers between immune cells, cytokines can serve as mediators between the immune system and the brain (Dantzer, 2001). Catecholamines, glucocorticoids, and proinflammatory cytokines (TNF-a) are considered to be the principal messengers between the immune system and the nervous system in the stress response. For example, chronic stress enhances the production of TNF-a, in turn increased TNFa levels can activate the HPA axis. For example, increased numbers of regulatory T cells in peripheral blood were observed in both atopic and nonatopic students under exam stress, as well as skewed Th1/Th2 ratio and reduced NK cell numbers that were unique to atopic students (Kang and Fox, 2000). Although some studies (e.g., Wright, 2005) have examined the impact that the activity of the stress system may have on immune activation and symptoms, very few studies have considered whether immune activation and the experience of having an atopic disease, particularly during childhood, influences the long-term responsiveness of the HPA axis. For example, Rosencrantz et al., 2005 used fMRI during antigen challenge to examine regional brain activation in adults with mild allergic asthma and identified activity in the anterior cingulated cortex (ACC) in response to asthma associated emotion words (e.g., wheeze). After antigen challenge, increased levels of IL-1 and IL-6 have been noted (Marshall et al., 2002). Depression has also shown to be associated with excessive secretion of proinflammatory cytokines such as IL-1 and IL-6 (Maddox and Pariante, 2001).
VIII. THE IMPACT OF SEX AND GENDER IN FOOD ALLERGY The relationship of sex and gender to health and disease is complex, and varies across an individual’s lifespan, and between cultures and different social contexts. Attention to sex and gender in biomedical and health sciences research is being actively promoted by the European Union Commission under their research policy of ‘‘mainstreaming gender equality’’(Klinge and Bosch, 2005). Sex denotes the differences attributed to biological origins alone, while gender refers to the social and cultural influences that lead to differences between women and men (DunnGalvin et al., 2007). One consequence of variables related to both sex and gender is that potentially differing patterns of disease prevalence, different degrees of severity, and different patterns of mortality and morbidity may be identifiable between men and women (e.g., Wizeman and Pardue, 2001).
82
A. DunnGalvin and J’ O. B. Hourihane
A. The influence of sex In population based studies sex differences in atopy (assessed as skin test reactivity to one or more of a panel of allergens) have been reported throughout childhood and into early adulthood, such that rates in girls are lower than in boys up to at least 15 years of age, in most studies up to 25 years of age, but are not consistently observed thereafter (Forde et al., 2003). In contrast to the sex differences in atopy assessed as skin test positivity which vary and change direction across the human lifespan, sex differences in atopy assessed as total serum IgE levels are consistent across the lifespan, with levels in females being lower than those in males (Burney et al., 1997). With reference to asthma and food allergy, prevalence is higher in boys before puberty, while this sex ratio is reportedly reversed after puberty (Becklake and Kauffman, 1999). Physiological pathways for these sex differences have been discussed with reference to ‘‘immune dimorphism,’’ the term given to differences in immune responses and regulation between the sexes. The mechanistic involvement of sex hormones in immune reactions has increasingly been acknowledged in recent years (Osman, 2003). Testosterone and oestrogen affect diverse cellular processes including protein synthesis, cell division and migration, neuronal growth and axonal branching, and synaptic remodeling. Receptors for sex steroids have been identified on lymphocytes, monocytes, and mast cells (e.g., Balzano et al., 2001). Lymphocytes are known to express both testosterone and estrogen receptors (Osman, 2003) whereas androgens enhance CD8þ lymphocyte activity and are correlated with the activation of IFN-g-secreting cells in healthy adults (Balzano et al., 2001). In allergy, sex hormone receptors on lymphocytes and leukocytes may modulate the type of immune reaction and regulate inflammation. For example, estrogens have a receptormediated effect on the releasibility of mast cells influencing the threshold levels in the effector phase of allergy (Da Silva, 1999). Different patterns of cytokine responses between males and females may be implicated in gender specific effects. Mechanisms linking psychological stress, personality, and emotion to neuroimnoregulation as well as increased risk of atopy have been increasingly elucidated (Wright, 2005). We already know that components of stress and the stress response differ between men and women. The tend-and-befriend response, mediated by oxytocin and endogenous opioids, may be more applicable to women than the fight-or-flight response, which was based largely on studies of men (Wright, 2005). Even within the flight-or-flight response pattern there are sex-based differences. The HPA axis interacts with reproductive function, such as menstruation. For example, when challenged by psychosocial stressors, males have been found to show a significant increase in glucocorteroid sensitivity
Developmental Trajectories in Food Allergy
83
but decreased proinflammatory plasma cytokine production, whereas females show a significant decrease in glucocorticoid sensitivity but unchanged proinflammatory plasma cytokine production (Rohleder et al., 2001). Furthermore, these sex differences were found only under active mental stress, not under passive cold stress. This raises the possibility that sex differences in self-reporting (see DunnGalvin et al., 2006) and even prevalence in some diseases could at least in part be explained by sex differences in the nature of the physiological response to stress, and, further that the nature of stressors may also influence sex differences in immune reactivity to stress (Kang et al., 2004) involving a complex interaction between biology and environment. For example, there are gender differences in the types of stressors to which an individual is likely to be exposed. The complexity of these sex and gender based interactions may explain the more adverse effects of food allergy on female over male general emotional well-being. For example, females with food allergy were found to be at increased risk of negative socio-emotional outcomes (Bollinger et al., 2006). Patients attending allergy clinics reported higher levels of depression compared to the general population (Kova’cs et al., 2003). A birth cohort study in Finland (Timonen, 2003) revealed that, at epidemiological level, skin prick test positive females exhibited up to an 1.8-fold greater risk of developing lifetime depression when compared with skin prick test negative subjects. In addition, the corresponding risk increased up to 2.7-fold among females, who had a positive skin prick test together with selfreported allergic symptoms. Maternal atopy alone almost doubled the risk of lifetime depression in female probands when compared with families in which no maternal atopy existed. In contrast, parental atopy did not predict any type of depression in male probands. However, as previously discussed, social adjustment or social anxiety was not investigated; therefore, we only have results on more extreme or clinical psychological or behavioral disorders, such as those described in the DSM-IV.
B. The influence of gender There has been little psycho-social research on the influence of gender in the context of food allergy (see DunnGalvin et al., 2007). Marklund and colleagues (2004) investigated the extent to which females and male adolescents experience perceived allergy-like conditions and the impact of these on everyday life. They found that adolescent females reported allergy-like conditions more frequently than adolescent males. Although all adolescents with allergy-like conditions reported significantly lower HRQL in seven of eight health scales that measured bio-psycho-social functioning, however, females reported more severe HRQL-deterioration compared with males. This is consistent with research that shows an
84
A. DunnGalvin and J’ O. B. Hourihane
excess of psychological vulnerability in adolescent girls with chronic conditions when compared to boys suffering from the same conditions, including epilepsy and asthma (Austin et al., 2000), insulin dependent diabetes mellitus (La Greca et al., 1995a,b), and cerebral palsy (MaGill and Hurlbut, 1986). Of the allergy conditions reported by Marklund, more than 50% of the adolescents stated they had food hypersensitivity with positive allergy tests. However, a sex and/or gender breakdown for confirmed food hypersensitivity or method of diagnosis was not included. Thus, it is not possible to determine if there is a gender difference in perceived versus actual food allergy for these individuals. However, work by Knibb and colleagues (1999) has demonstrated a gender bias in reporting self-diagnosed food allergy and intolerance, with significantly more females self-reporting than males. Differences in self-reports of ill-health and psychological distress have also been observed in adolescents in the general population. For example, Sweeting and West (2003) found that self-reported general ill-health and physical symptoms, as well as psychological distress was significantly higher and increasing from age 11 for females compared to males. This increased with age and by age 15, there was a female excess in general illhealth, including psychological distress and ‘‘malaise,’’ limiting illness, poor self-rated health, headaches, stomach problems, and dizziness. This may also explain possible gender differences in self-assessed health in the context of perceived food allergy. This has been explained by the concept of ‘‘illness centrality’’ (Wiebe et al., 2002), or the level at which particular illness has been integrated into the self-concept. In a recent study (DunnGalvin et al., 2009), we found that as the child develops, the level of integration of food allergy into the self concept also develops in a mostly gender-specific manner and has consequences for the child’s understanding of food allergy and everyday management of the condition. We found that girls tend to incorporate food allergy into their self-concept, making it a defining part of who they are, whereas males ‘‘contain’’ the illness by minimizing its importance. Interestingly, although boys evinced lower anxiety levels because of this tendency, they were more prone to ‘‘risky’’ behavior (e.g., not bothering to read labels), whereas girls were more anxious, but also demonstrated more self-care behaviors.
IX. RISK BEHAVIOR IN FOOD ALLERGY Risk is considered as the probability of a negative event occurring and can be quantified. However, this form of risk can be understood as ‘‘danger.’’ The perception of risk is a socially constructed phenomenon, and is more difficult to measure. Psychological risk is based on perception rather than
Developmental Trajectories in Food Allergy
85
fact, and is therefore based on qualitative, not quantitative characteristics of the hazard being considered. For example, when one individual feels a sharp pain, he/she may interpret it as a possible ‘‘heart-attack’’ and phone for an ambulance, whereas another individual might just reach for an indigestion remedy. As already discussed, food allergy has an impact on quality of life, due in part to the constant vigilance required on the part of the allergic individual, or their caregiver, to ensure accidental ingestion of food allergens does not occur. However, limited research exists in risk perception in food allergy and there is no research to date on process variables and/or causal pathways involved in the initiation, treatment, and cessation of health risk behaviors in food allergy. Research looking at the incidence of severe allergic reaction has suggested that adolescents and young adults are more at risk. In a Norwegian study on severe allergic reactions to food, the main risk group was comprised of young adults aged 20–35 (Lovik et al., 2003). Teenagers represent a high-risk group for anaphylactic fatalities caused by food allergy, accounting for 53% of a group of UK fatalities (Pumphrey, 2000). A recent paper by Sampson and colleagues (2006) found that adolescents and young adults appear to be at an increased risk for fatal food allergic reactions, and suggested that they may adopt more risk-taking behaviors with regard to their food allergy; however, gender differences or possible causal mechanisms were not explored. The study population included persons with a high degree of severity of food-induced allergic disease, with numerous food allergies, and frequent and severe reactions, however, 37% with severe symptoms did not receive epinephrine, and 38% did not have it with them during severe reactions. The authors suggestions for intervention was to encourage clinicians to emphasize to patients that food is often a part of all group activities, and an accidental exposure could occur, making it necessary and safest to always have an epipen available. They also suggested teaching parents to remind teenagers about carrying epipens to social events. It is clear, therefore, that research is needed to establish underlying emotional risk factors associated with risk taking. Perception of risk in adults is usually described, or interpreted, with reference to health belief models (HBMs) (Ajzen and Fishbein, 1980; Janz and Becker, 1984). Individuals are more likely to engage in health behaviors if they perceive: vulnerability to health threats; that the consequences are severe; that treatment or preventive measures will be successful. Although there are variants to the framework, the different models share many of the same elements. In effect, theories assume that individuals rationally weigh benefits and costs and act according to the outcome of this analysis. Subsequent modifications to the models include the addition of perceived social or monetary barriers to the adaptive response. A cue to action which can be internal (e.g., symptoms) or external (e.g., health communication) is hypothesized to trigger these
86
A. DunnGalvin and J’ O. B. Hourihane
cognitive processes. Of course, demographic and sociopsychological variables may influence perceptions and indirectly affect the likelihood of the response (van der Pligt, 1994, 2002). Although the HBM have been found to predict children’s expectations to use medicines to treat illnesses (Bush and Iannotti, 1990), there is limited support for the model and for components of the model predicting children’s health behaviors. This is especially evident for perceived vulnerability to health threats. For example, early research revealed mostly negative relations between children’s health behaviors and perceived vulnerability to risk (e.g., Gochman and Saucier, 1982) and initial findings were later extended to adolescents (Greening and Stoppelbeim, 2000), thus challenging the hypothesis postulated by HBM that greater perceived risk is related to adaptive behaviors in children and adolescents. Such findings suggest that risk perception is a complex process that warrants a deeper understanding from both health educators and researchers. For example, in light of the incongruence between knowledge and action in terms of compliance and risk, there is an increasing recognition of the need to qualitatively explore people’s experiences, perceptions, and understandings of what it is like to live with a chronic condition, including its management, in order to better understand the decisions people make about managing their condition. Developmental factors are also important when considering risk perception in food allergy. For example, in middle childhood, as processes relating to the impact of social comparison develop, children may be tempted to reject safety rules in order to ‘‘fit-in.’’ Adolescents and young adults frequently eat away from home, they face growing peer pressure, and alcohol consumption may be high. The latter may both impair their ability to assess risk and augment the physiological effects of allergen encounter such as vasodilation (Sampson, 2004). For many adolescents, the social pressure for psychosexual autonomy directly clashes with the prolonged dependence on family, which may be particularly pronounced in chronic diseases such as food allergy. Researchers in the field of allergy suggest that teenagers should be a priority group for the development and evaluation of interventions to improve their adherence to management plans. However, interventions will not prove successful unless we know the disease-specific developmental trajectory by which some teenagers become high risk and others do not.
X. DEVELOPMENTAL PATHWAYS IN FOOD ALLERGY All these studies represent outcomes, but, there has been little, if any, research in allergic diseases in general and food allergy in particular, into the developmental pathways that lead to these observable consequences.
Developmental Trajectories in Food Allergy
87
This limits the ability of clinicians, researchers, and policy makers to predict and evaluate cognitive and emotional development in the food allergic child, with implications for prevention, treatment, intervention and health policy. For example, allergists assume, but have never confirmed, that high levels of vigilance in children performs an adaptive protective function and psycho-social outcomes have not been investigated to date. Furthermore, a developmental perspective has the potential to provide an explanatory framework for previous disparate findings of the impact of food allergy on children and adolescents. The way in which children and adolescents cope with chronic health conditions is considered as an increasingly important predictor of health in clinical and psychosocial research (Schmidt, 2003). Consideration of the developmental perspective is not only useful when studying children, but also helps to explain cognitive processes in response to stressors in adulthood. However, coping has been studied less frequently in children and adolescents than in adults. Researchers contend that this may be due to the difficulty in assessing developmental processes that are occurring simultaneously in children and teens (Schmidt, 2003). The interrelatedness of coping and development implies that coping is a process that is shaped by developmental organization and, likewise, development is shaped by coping processes (Schmidt, 2003; 214). However, early adopted strategies of coping with chronic disease may serve as a buffer against these disease-related consequences, even if a certain stability of coping strategies across situations and developmental stages cannot be assumed. Coping has not only been shown to be related to patient well-being, but mediates health behavior as well as health care utilization (Barlow et al., 2002b). For health care providers, there is socioeconomical interest to support the development of adaptive and active coping strategies in children with chronic conditions as early as possible. A recent study (DunnGalvin et al., 2009), under the aegis of Europrevall, represents a first attempt to provide an integrated developmental framework to explain the onset, development, and maintenance of food allergy related cognitions, emotions and behavior. 62 children/teenagers aged 6–15 years took part in 15 age appropriate focus groups, 52% of children were female. Parents were also interviewed. All children were physician diagnosed with IgE-mediated food allergy and had been issued with an anapen/epipen. Through qualitative enquiry, a framework for evaluating children with food allergy was developed. Developmentally appropriate techniques were designed to stimulate discussion, maintain interest, and minimize threat to the child’s self-esteem. Six main themes emerged from the analysis that encompassed precipitating events (stressful events in the children’s lives caused by food allergy related factors); psychological impact (cognitive appraisal and emotional effects); and behavioral
88
A. DunnGalvin and J’ O. B. Hourihane
consequences or coping strategies. Findings indicate that coping in food allergy is more than simply a strategy, it is a cumulative history of interactive processes (both age, gender, and disease specific) that are embedded in a child’s developmental organization. Our findings show that food allergy is a central ‘‘lens’’ in children’s lives through which they interpret experiences. When children and teens are confronted with a stressful event, such as a birthday party, a novel situation, an allergic reaction, or making new friends, the way in which they appraise the event, and its attendant emotional impact are viewed through this lens. How this lens is constructed and its psychological impact (uncertainty, anxiety, confusion, difference) on individual children is modified by age, gender, context, prior experience, attitudes of parents, attitude of peers, and level of general awareness. Age is an important factor in determining the type of event children are likely to encounter from structured events such as birthday parties in younger children to more unstructured and unplanned events as children become more independent. Children live within the context of their families, which have interaction patterns, rules, organizing principles, and general belief systems, as well as those specifically regarding health and disorder. Parental stress and perceptions of level of threat and consequent anxiety has a profound impact on the way that children themselves perceive risk and control. In focus groups with parents, many admitted to overprotecting their children through an understandable desire to ensure their children’s safety. Teens used very similar language and phrases when talking about aspects risk and control that younger children ascribed to their parents—and that parents themselves used in focus groups. In most children under the age of 8 years, there is a certainty of parental and adult knowledge and a consequent sense of control of events relating to food allergy. Children in the 6–8 years age group described the food they eat as ‘‘special’’ and also described themselves in this manner. In addition, younger children are more confident in social situations because of the protective presence of the parent. However, a transition point occurs around >8 years when children begin to describe themselves as different and the term special is ascribed to parents and takes on a negative connotation. At this time, children also begin to learn or feel that they cannot conclusively prevent an allergic reaction from occurring, giving rise to a state of uncertainty, and impacting on children’s perception of autonomy and self-belief in their ability to control events in their lives. Older children have to learn to live with constant uncertainty, often reinforced by parent’s own anxiety about safety. This sense of uncertainty continues to grow as children develop and when children were asked ‘‘when do you have a reaction,’’ many recounted stories of accidental
Developmental Trajectories in Food Allergy
89
ingestion of allergic foods when visiting relatives, at school, or at social events ‘‘even though I was careful.’’ This state of uncertainty is also reinforced as children and teens encounter a widespread lack of awareness that encompasses teachers, schoolfriends, classmates, shops, restaurants, coffeeshops, afterschool activities, social events, and other parents. Therefore, children become aware, both through parents or through direct experience that a clear dichotomy exists between safe places and people and risky places or people, perceived by participants as ‘‘those who understand and those who don’t.’’ This directly impacts on perception of control and illness intrusiveness. In addition, difficulties in adapting concrete rules often results in a perception of external control over a particular event, particularly in the transition period between parental and ‘‘self’’ control. A food allergic child’s identity appears to be closely tied to the dietary and social restrictions that come with their condition. In middle childhood as peer comparison begins to appear, children with food allergy begin to make causal connections between experiences in the world at large and inner feelings, a strong negative association can develop between appraisal in terms of an objective health threat and appraisal in terms of the emotional response to a health threat. This perception of ‘‘difference’’ is isolating and has consequences for how children perceive themselves and how they feel they are perceived by others. Indeed, teens speak about social events mostly in terms of their restrictions. Events are appraised in the context in which they occur and an awareness of expected behavior. For example, whereas an allergic reaction which takes place in the home may be regarded as a low-anxiety event, one which takes place in school can be appraised a highly stressful because it impacts on the child’s goal to ‘‘fit-in.’’ This goal confronts children with the difficult balancing act of protecting their ego and managing risk. He or she learns to appraise (and weight) threats to personal safety with threats to social identity. The stress appraisal process may result in children ‘‘just chancing it will be ok’’ and deliberately eating an allergic food when in the company of others, whereas others protect their self-esteem by avoiding novelty as much as possible. With regard to risky behavior, another important factor emerged as a motivation for deliberately eating an allergic food. It appears that some older children attempt to determine their own risk thresholds; ‘‘I would have a little bit. . . and see what happens. . . you might get a bit sick only.’’ This may be a way to gain some control over feelings of uncertainty that are integral to growing up with food allergy. Levels of anxiety appear to range from mild anxiety in a situation where the allergic food is present to more severe ‘‘trait-like’’ anxiety. About 20% of the participants mentioned being anxious ‘‘a lot of the time,’’ the majority of which were girls. Coping strategies were found to
90
A. DunnGalvin and J’ O. B. Hourihane
lie on an maximization/avoidance to minimization/risk continuum. The majority of girls were found to use ‘‘avoidant’’ strategies to cope with living with food allergy. Many clinicians assume that these strategies are necessary and adaptive, if they are proportionate. However, results showed high levels of anxiety, vigilance, and generalized avoidance of situations and people not directly related to food consumption are associated with maladaptive avoidant strategies. A surprising finding was that anxious children and teens are not necessarily those who experience the most or recent reactions. For example, many of the children who described themselves as anxious or worried about food allergy could not remember ever having had a serious reaction. ‘‘Minimizing’’ strategies, such as not reading labels, not telling others that you are food allergic in risky situations and deliberately eating an allergic food (mostly found in boys), are also maladaptive in that children who use them are more vulnerable to experiencing an allergic reaction. Opportunities to engage in risky behavior increase as children enter their teens. Teenagers represent a high-risk group for anaphylactic fatalities caused by food allergy, accounting for 53% of a group of UK fatalities (Pumphrey, 2000).
A. Transition points in development for food allergic children With reference to transition points in development, a negative representation of food allergy can result in psychological distress, which may accompany, or follow, transition (or sensitive) points in the developmental pathway, for example, when children learn that they (or their parents) cannot conclusively prevent an allergic reaction. In the middle childhood years, children must begin to gain autonomy and self-belief in their ability to control events in their lives. Appraisal processes outlined above may result in increasing attention being given to processing information relating to food allergy, accompanied by emotional arousal and consequent increased detection of threat, whether social (self-concept) or personal (safety). However, recent experience of a reaction is not a necessary precondition for this. Because food allergy places particular limitations on children’s lives and frequently leads to restrictions in a variety of activities, there is the potential for these perceived illness-induced limitations to generalize to disease- unrelated events as children progress along the development pathway. For example, we have evidence of an over-interpretation of ambiguous information in terms of processing of potential threat not directly related to food; ‘‘you’d be worried if you were somewhere new. . . that you weren’t before; ‘when other kids see an advert for a circus, they think fun, I think danger.’’ In adolescence we encounter another transition point, when teens encounter unstructured ‘‘novel situations’’ and peers and their automatic
Developmental Trajectories in Food Allergy
91
response patterns are challenged resulting, as before, in increased attention to threat. This appears to apply particularly to those with avoidant coping strategies at the middle to high points of the avoidance/risk continuum and has an impact on cognitions, emotions and ultimately behavior. Although many teens appear to accept an ongoing conflict between personal safety and social self-concept, others resolve this by intensifying previous behaviors, whether minimization/risk or maximization/avoidant, moving them up or down the continuum. Finally, increasing coordination and integration of regulatory systems (e.g., information processing, and appraisal processes) over the course of development means that by the time children reach their teens, self-perception, emotional reactions, and cognitive appraisal mechanisms have become relatively stable and consistent.
B. Parental views and management of food allergy Food allergy adds an extra element to the ordinary challenges of parenting. There was to considerable overlap between parent and children/teen focus group findings. Both parents and children spoke of feelings of anxiety, frustration, anger, difference, embarrassment, and uncertainty in response to situations stemming from food allergy. In addition, parents spoke about the difficulty of balancing the needs of their nonallergic children against those of the child with food allergy, and negotiating the precarious line between supporting children’s independence and keeping them safe. These challenges sometimes led to conflict between parents and children and between father and mother which, in turn, impacted on the quality of family life. Furthermore, the sense of huge responsibility, over and above that felt by the parent of a ‘‘normal’’ child, and guilt, means that levels of stress often continue for significant periods of time. In order to keep their children safe, and cope with their own anxiety, a substantial proportion of mothers, in particular, appear to micromanage their young children’s lives, which may lead to dependence and low self-efficacy in some teens and, in contrast, to rebellion and risky behavior in others. On this point, as in the child/teen transcripts, there appears to be gender differences, with parents reporting that boys tend to hide their food allergy from others and girls rely largely on parents, and longstanding friends as they enter adolescence, to manage social situations involving food. There are also gender differences between fathers and mothers, although this difference may pertain to language style and role expectations, in that mothers were more likely to talk about feelings of anxiety and helplessness. Parents had different concerns at different developmental points, for example, when children were young parents keenly felt the burden of responsibility in managing risk in their children’s lives, but were less
92
A. DunnGalvin and J’ O. B. Hourihane
anxious about children’s safety because of higher perceived control. Parents were particularly anxious in the transition period from primary to secondary school. In primary school, classes are generally smaller and parents are familiar with friends, however, in secondary school children encounter new social and peer pressures, and hence an increased risk. Parents worried that children would not carry their ‘‘pens,’’ would take risks with food in response to peer pressure, or would not inform others of their food allergy.
XI. DISCUSSION AND IMPLICATIONS FOR FUTURE RESEARCH Because food allergy is only sporadically symptomatic, the anxiety and uncertainty associated with a potential reaction has more profound effects on a child’s everyday life than physiological symptoms; therefore, children with food allergy may be at increased risk of negative short- and long-term socioemotional outcomes. Low general awareness encountered in the social world appears also to contribute to the later development of psychological distress and/or to risky behavior. Additionally, variables, such as personality traits and gender are likely to moderate individual impact. Findings also demonstrate the need for health professionals to work closely with the understandings individuals have about their food allergy, about ‘compliance’ and about risks associated with the condition. Studies to date show that many of the issues around food allergy reported by participants and their parents are operative at an emotional level. As such, they fall outside the responsibility or indeed the capacity of family doctors or allergists to address in a direct way, entailing the need for an educational intervention specific to food allergy. Many existing interventions in chronic diseases focus only on disease management and information and fail to address the wider psychosocial consequences of living with chronic disease (Barlow and Ellard, 2004; Gage et al., 2004). Although it is very important that children and their families understand the condition and its treatment, Gibson and colleagues (2004) maintain that asthma education limited to information transfer alone is ineffective. Further, research suggests that the goal of an intervention should not only to be change the mean level of a particular coping strategy but also to create a stable growth dynamic by including issues that are food allergy specific, as well as age and context specific. Research in food allergy suggests that greater support and clear information is important at time of diagnosis and at the different transition points along the development pathway. Greater emphasis is needed on the social and emotional aspects of food allergy, on knowing ‘‘what to expect,’’ and on enhancing the selfmanagement skills of children and their families. In addition, public
Developmental Trajectories in Food Allergy
93
information on food allergy is needed in general, and in health, education, catering, and retailing, in particular. Food allergic children appear to experience disease-specific stressors and use particular coping strategies that evolve in response to age-, gender-, and context-specific stressors. Particular issues revealed by the literature are summarized below: For young children particular issues include: how to feel part of social
occasions such as birthday parties while being and feeling safe; how to communicate with friends, restaurant staff, and other children and adults in novel situations; how to deal with difficult people and situation (e.g., teasing and bullying); how to create a positive self-image that includes food allergy. For older children and young teens particular issues include: how to balance peer pressure and positive self-perception while staying safe; how to communicate in novel situations such eating out and making new friends; how to manage feelings such as embarrassment, selfconsciousness, difference, managing the removal of the parental safety net, and development of effective self-care. For older teens particular issues include: how to deal with going out independently with friends, starting to drink alcohol, the first independent holiday, romantic relationships, and attending clubs or discos. For parents particular issues include: anxiety; teaching children to be independent and safe, what to expect as children grow, managing the removal of the parental safety net and helping children develop effective self-care. Gender specific issues include: how to deal with gender roles (e.g., boys do not seek social support; girls find it difficult to be assertive).
Over the last 20 years, new lifecourse frameworks have been developed which offer a conceptual model for health development and a more powerful approach to understanding diseases (Halfon and Hochstein, 2002). Such models are germane in the field of food allergy since biological hypersensitivity to environmental stimuli is a central feature of atopic disorders. Although it is recognized that allergen contact can elicit symptoms at higher and lower dose at different time points, rendering different thresholds in allergen provocation tests within the same individual, there is only a limited understanding of the mechanisms involved in the developmental course of food allergies. A developmental framework has the potential to link formerly disparate concepts such as health-related quality of life, the maturation of the immune system, and delineate mechanisms linking psychological stress, personality and emotion to neuroimmunoregulation as well as to increased risk of negative impact. Such a model may be used to explain both physiological and psychological phenomena, and their interaction,
94
A. DunnGalvin and J’ O. B. Hourihane
and consequently provide a shared language as a basis for multidisciplinary studies in food allergy. A growing appreciation of the interactions between behavioral, neural, endocrine, and immune processes have underlined the need for a multidisciplinary approach. For example, it is possible that psychological distress and atopy have overlapping biological contributors, increasing the likelihood of ‘‘trait’’ anxiety. Mechanisms linking psychological stress and emotion to neuroimnoregulation as well as increased risk of atopy have been increasingly elucidated (e.g.,Wright, 2005). Stress early in life can, for example, result in long-term alterations of the function of the HPA-axis (Helm et al., 2002). In addition, an association between TNF-a and the cognitive affective subscale of the Beck Depression Inventory, which measures depressed mood independent of physical symptoms, demonstrated a negative affect-specific activation of proinflammatory cytokines that may actually promote disease progression (Knonfol and Remick, 2000). A biopsychosocial framework may reveal new links between physiological and psychological systems that, in turn, may provide new insights to guide future explorations that result in novel clinical or therapeutic treatments that relieve the burden of food allergy. Such a framework entails the adoption of methodologies that illuminate pathways in development such as qualitative methods and structural equation modeling. One of our aims in ongoing research is to move beyond quantitative reports of HRQL impact to an examination of the underlying mechanisms. A developmental framework, such as that illustrated (Fig. 3.1), has the potential to link formerly disparate concepts such as health-related quality of life, the maturation of the immune system, cytokine secretion, to the influence of sex and gender variables, as well as to increased risk of negative physiological or psychological impact. Such a model may be used to explain both physiological and psychological phenomena, and their interaction, and consequently provide a shared language as a basis for multidisciplinary studies in food allergy. Longitudinal studies are necessary when the goal is to investigate cause and effect. At present Europrevall is conducting a birth cohort study in food allergy, that includes clinical and psychological measures. Such studies may also lead to novel treatment options in the future. For clinicians, the early recognition and incorporation of a developmental framework into a treatment plan is essential and sets the stage for an effective medical care and the eventual transition from paediatric to adult care. For health care providers, there is socio-economical interest to support the development of adaptive and active coping strategies in children with food allergy as early as possible, targeted at specific transition points, and with age and gender relevant content.
Developmental Trajectories in Food Allergy
95
REFERENCES Austin, J. K., Dunn, D. W., and Iluster, G. A. (2000). Childhood Epilepsy Asthma: Changes in Behavior Problems Related to Gender and Change in Condition Severity. Epilepiu 41(5), 615–623. Ajzen, I. and Fishbein, M. (1980). ‘‘Understanding Attitudes and Predicting Social Behavior.’’ Englewood Cliffs, NJ, Prentice-Hall. Al-Muhsen, S., et al. (2003). Peanut allergy: An overview. Can. Med. Assoc. J. 168(10), 1279–1285. Anderson, A. S. (2008). Increase in hospitalisations for anaphylaxis in the first two decades of life. J. Allergy Clin. Immunol. 121(2), Abstract 104. S27. Asero, R., Ballmer-Weber, B. K., Beyer, K., Conti, A., Dubakiene, R., Fernandez-Rivas, M., et al. (2007). IgE-Mediated food allergy diagnosis: Current status and new perspectives. Mol. Nutr. Food Res. 51, 135–147. Avery, N. J., King, R. M., Knight, S., and Hourihane, J. O. B. (2003). Assessment of quality of life in children with peanut allergy. Pediatr. Allergy Immunol. 14, 378–382. Balzano, G., Fuschillo, S., Melillo, G., and Bonini, S. (2001). Asthma and sex hormones. Allergy 56, 13–20. Barlow, J. H. and Ellard, D. R. (2004). Psycho-educational interventions for children. Child: Care Health Dev. 30(6), 637–645. Barlow, J. H., Wright, C., Sheasby, J., Turner, A., and Hainsworth, J. (2002a). Self-management approaches for people with chronic conditions: A review. Patient Educ. Couns. 48, 177–187. Barlow, J. H., Wright, C. C., Shaw, K. L., Luqmani, R., and Wyness, I. J. (2002b). Maternal stressors, maternal wellbeing and children’s wellbeing in the context of juvenile idiopathic arthritis. Early Child Dev. Care 172, 358–367. Becklake, M. R. and Kauffman, F. (1999). Gender differences in airway behaviour over the life-span. Thorax 54, 1119–1138. Bender-Berz, J., Klein Murdock, K., and Mitchell, D. (2005). Children’s asthma, internalising problems, and social functioning: An urban perspective. J. Child Adolescent Psych. 18, 181–197. Bock, S. A., Munoz-Furlong, A., and Sampson, H. A. (2001). Fatalities due to anaphylactic reactions to foods. J. Allergy Clin. Immunol. 107, 191–193. Bollinger, M. E., Dahlquist, L. M., Mudd, K., Sonntag, C., Dillinger, L., and McKenna, K. (2006). The impact of food allergy on the daily activities of children and their families. Ann. Allergy Asthma Immunol. 96, 415–421. Burks, A. W. (2006). Food allergies. In ‘‘ACP Medicine, section 6, chap. 16’’, (D. C. Dale and D. D. Federman, Eds.). WebMD, New York. Burney, P., Malberg, E., Chinn, S., Jarvis, D., et al. (1997). The distribution of total and specific IgE in the European Community Respiratory Survey. J. Allergy Clin. Immunol. 99(3), 314–322. Bush, P. J. and Iannotti, R. J. (1990). A children’s health belief model. Medical Care 28, 69–86. Bush, P. and Iannotti, R. (1993). Alcohol, cigarette, and marijuana use among fourth grade school children in 1988/89 and 1990/91. Am. J. Public Health 83, 111–114. Cohen, B. L., Noone, N. S., Munoz-Furlong, A., and Sicherer, S. L. (2004). Parental burden in food allergy. J. Allergy Clin. Immunol. 114(5), 1159–1163. Da Silva, J. A. P. (1999). Sex hormones and glucocorticoids: Interactions with the immune system. Ann. NY Acad. Sci. 876, 102–117. Dantzer, R. (2001). Cytokine-induced sickness behaviour: Where do we stand? Brain Behaviour and immunity 15, 7–24. De Blok, B. M. J., DunnGalvin, A., Dubois, A. E. J., and Hourihane, J. O. B. (2007). A framework for measuring the social impact of food allergy across Europe. Allergy 62(7), 733–737.
96
A. DunnGalvin and J’ O. B. Hourihane
Dixon, S. D. and Stein, M. T. (2006). ‘‘Encounters With Children: Pediatric Behavior and Development’’, 4th Ed. Mosby Elsevier, Philadelphia. DunnGalvin, A. (2007). Talking Life Sciences to Both Sexes: Research Communication and Gender. Available from: European Commission, Directorate General Research, Brussels, Belgium. DunnGalvin, A., Du Bois, B., De Blok, B. M. J., and Hourihane, J. O. B. (2007). Child vs maternal perception of HRQoL in food allergy: Developmental trajectories and evolution of risk behaviour. Allergy 62(Suppl. 83), 70–166. DunnGalvin, A., Hourihane, J. O. B., Frewer, L., Knibb, R. C., Oude Elberink, J. N. G., and Klinge, I. (2006). Incorporating a gender dimension in food allergy research: A review. Allergy 61(11), 1336–1343. DunnGalvin, A., de Blok, B. M. J., Dubois, A., and Hourihane, J. O. B. Development and Validation of the Food Allergy Quality of Life—Parent Administered Questionnaire(FAQLQ-PF) for food allergic children aged 0–12 years. Clin. Exp. Allergy 38, 977–986. DunnGalvin, A., Gaffney, A., and Hourihane, J. O. B. Developmental trajectories in food allergy: A new theoretical model. Allergy (2009). DunnGalvin, D., DunnGalvin, A., Hourihane, J. O’. B. The Implicit Association Test in Food Allergy: A New Direction in Assessment of Chronic Disease (unpublished data). Eggesbo, M., Botten, G., Halvorsen, R., and Magnus, P. (2001). The prevalence of allergy to egg: A population-based study in young children. Allergy 56, 403–411. Eiser, C. and Morse, M. (2001). Quality of Life in chronic diseases of childhood. Health Technology Assess 5(4), 1–157. Elenkov, I. J., Wilder, R. L., Chrousos, G. P., and Vizi, E. S. (2000). The sympathetic nerve – an integrative interface between two supersystems: The brain and the immune system. Pharmacol. Rew. 52, 595–638. Fayers, M. and Machin, D. (2000). Quality of life: Assessment, analysis, and interpretation Chichester. John Wiley, New York. Flokstra-De Blok, B. M. J., DunnGalvin, A., Dubois, A. E. J., and Hourihane, J. O’. B. (2008). The development and validation of the self-administered food allergy quality of life questionnaire for teens. JACI 14, 121–125. Forde, E., Mannino, D. M., Homa, D. M., Gwynn, C., Redd, S. C., Moriarty, D. G., et al. (2003). Self-reported asthma and health related quality of life. Chest 123, 119–127. Gage, H., Hampson, S., Skinner, T. C., Hart, J., Storey, L., Foxcroft, D., Kimber, A., and Cradock, S. (2004). Educational and psychosocial programmes for adolescents with diabetes: Approaches, outcomes and cost-effectiveness. Patient Educ. Couns. 53, 333–346. Gibson, P. G., Coughlan, J., Wilson, A. J., Hensley, M. J., Abramson, M., Bauman, A., and Walters, E. H. (2004). Limited (information only) patient education programs for adults with asthma (Cochrane Review). In: Barlow, J. H., Ellard, D. R. Psycho-educational interventions for children. Child Care Health Dev. 30(6), 637–645. Gifford-Smith, M. E. and Browning, C. A. (2003). Child peer relationships: Social acceptance, friendships and peer networks. J. School Psychol. 41(4), 235–284. Gochman, D. S. and Saucier, J. F. (1982). Perceived vulnerability in children and adolescents. Health Educ. Q. 9, 46–59. Greening, L. and Stopplebeim, L. (2000). Young drivers Health Attitudes and Intentions to Drink and Drive. J. Adolesc. Health 27, 94–101. Greenwald, A. G., McGhee, D. E., and Schwarz, J. L. K. (1998). Measuring individual differences in implicit cognition: The Implicit Association Test. J. Pers. Soc. Psychol. 74 (6), 1464–1480. Halfon, N. and Hochstein, M. (2002). Life course health development: An integrated framework for developing health, policy and research. In, Milbank Q. 80(3), 433–479.
Developmental Trajectories in Food Allergy
97
Hampson, S. E., Skinner, T. C., Hart, J., Storey, L., Gage, H., Foxcroft, D., Kimber, A., Shaw, K., and Walker, J. (2001). Effects of educational and psychosocial interventions for adolescents with diabetes mellitus: A systematic review. Health Technol. Assess. 5, 1–79. Harter, S. (1997). The Construction of the Self: A Developmental Perspective. The Guilford Press, New York. Helm, R. M., Furuta, G., Stanley, J. S., Yui, J., Cockrell, G., Connaughton, C., Bannon, G. A., and Burks, A. W. (2002). A neonatal swine model for peanut allergy. J. Allergy Clin. Immunol. 109, 135–142. Hennessy, E. and Heary, C. (2004). Exploring Children’s View through focus groups. In ‘‘Researching children’s experiences: Approaches and methods’’, (S. Green and D. Hogan, Eds.), p. 236. Sage, London. Hill, M. (2006). Children’s voices on ways of having a voice: Children’s and young people’s perspectives on methods used in research and consultation. Childhood: A Glob. J. Child Res. 13(1), 69–89. Hourihane, J. O. (1998). Prevalence and severity of food allergy–need for control. Allergy 53(46 Suppl), 84–88. Hourihane, J. O., Smith, P. K., and Strobel, S. (2002). Food allergy in children. Ind. J. Pediatr. 69(1), 61–67. Janz, N. K. and Becker, M. H. (1984). The Health Belief Model: A decade Later. Health Educ. Q. 11, 1–47. Johansson, S. G. O., Bieber, T., and Dahl, R. (2004). Revised nomenclature for allergy for global use: Report of the Nomenclature Review Committee of the World Allergy Organisation 2003. J. Allergy Clin. Immunol. 5, 832–836. Kang, D. H., Kim, C. J., and Suh, R. (2004). Sex Differences in Immune Responses and Immune Reactivity to Stress in Adolescents. Biological Research For Nursing Vol. 5(No. 4), 243–254. Kang, D.-H. and Fox, C. (2000). Neuroendocrine and leukocyte responses and pulmonary function to acute stressors. Ann. Behav. Med. 22, 276–285. Katon, W. J., Richardson, L., Lozano, P., and McCauley, E. (2004a). The relationship of asthma and anxiety disorders. Psychosomatic Medicine 66, 349–355. Katon, W. J., Richardson, L., Lozano, P., and McCauley, E. (2004b). Differences in Immune Responses and immune Reactivity to Stress in Adults. Biological Research For Nursing Vol. 5, (No. 4), 243–254. Keating, D. P. and Miller, F. K. (2005). Individual pathways in competence and coping: from regulatory systems to habits of mind. In Developmental Health and the Wealth of Nations: Social, Biological, and Educational Dynamics, (D. P. Keating and C. Hertzman, Eds). Guilford Press, New York, NY. Kimber, I. and Dearman, R. J. (2002). Factors affecting the development of food allergy. Proc. Nutr. Soc. 61(4), 435–439. King, R., Knibb, R. C., and Hourihane, J. O’B. (2008). Quality of life in children with peanut allergy and their immediate family. Allergy. (in press). King, R. M., Knibb, R. C., and Hourihane, J. O. (2005). A Study to Assess the Quality of Life in Children with Peanut Allergy, their Parents and Siblings Presented at the Annual Meeting of the AAAAI Control/Tracking No: 06-A-2407-AAAAI. Klinge, I. and Bosch, M. (2005). Transforming research methodologies in EU life sciences and biomedicine: Gender-sensitive ways of doing research. Eur. J. Women’s Stud. 12(3), 377–395. Knibb, R. A., Booth, D. A., Platts, R., Armstrong, A., and Booth, L. W. (1999). Psychological characteristics of people with perceived food intolerance in a community sample. J. Psychosom. Res. 47, 545–554. Knibb, R. C. and Horton, S. L. (2008). Can illness perceptions and coping predict psychological distress amongst allergy sufferers? Br. J. Health Psychol. 13, 103–111.
98
A. DunnGalvin and J’ O. B. Hourihane
Knonfol, Z. and Remick, D. G. (2000). Cytokines and the brain: Implications for clinical psychiatry. Am. J. Psychiatry 157, 683–694. Kova’cs, M., Stauder, A., and Szedma’k, S. (2003). Severity of allergic complaints and the importance of depressed mood. J. Psychosomatic Res. 54, 549–557. La Greca, A. M., Swales, T., Klemp, S., Madigan, S., and Skyler, J. (1995a). Adolescents with diabetes: Gender differences in psychosocial functioning and glycemic control. Child Health Care 24, 61–78. LaGreca, A., Auslander, W., Greco, P., Spetter, D., Fisher, E., and Santiago, J. (1995b). ‘I get by with a little help from my family and friends - adolescents support for diabetes care’. Journal of Pediatric Psychology 20, 449–476. Lovik, M., Wilker, H. G., Stensby, B. A., Kjelkevik, R., Sommer, A. K., Mangschou, B., Omholk-Jenson, B., and Mork, A. V. (2003). The Norwegian National Reporting System and Register of Severe Allergic Reactions to Food. In ‘‘Clinical Immunology and Allergy in Medicine,’’ (G. Marone, Ed.), pp. 461–466. JGC Publishers, Napoli. Maddock, C. and Pariante, C. M. (2001). How does stress affect you? An overview of stress, immunity, depression and disease. Epidemiol. Psychiatr. Soc. 10(3), 153–162. Magill, J. and Hurlbut, N. (1986). The self-esteem of adolescents with cerebral palsy. Am. J. Occup. Ther. 40, 402–407. Marklund, B., Ahlstead, B., and Nordstrom, G. (2004). Health-related quality of life among adolescents with allergy-like conditions—with emphasis on food hypersensitivity. Health Qual. Life Outcomes 2, 65–69. Marshall, P. S., O’Hara, C., and Steinberg, P. (2002). Effects of seasonal allergic rhinitis on fatigue levels and mood. Psychosomatic Medicine 64, 684–691. Meaney, M. J. and Szyf, M. (2005). Maternal care as a model for experience-dependent chromatin plasticity? Trends Neurosci. 28, 456–463. Meijer, S. A., Sinnema, G., Bijstra, J. O., Mellenbergh, G. J., and Wolters, W. H. G. (2002). Peer interaction in adolescents with a chronic illness. Pers. Individual Differences 29, 799–813. Miles, S., Fordham, R., Mills, C., Valovirta, E., and Mugford, M. (2005). A framework for measuring costs to society of IgE-mediated food allergy. Allergy 60, 996–1003. Miller, J. J. (1993). Psychosocial factors related to rheumatic disease in childhood. J. Rheumatol. 20 S38, 1–11. Moss-Morris, R., Weinman, J., Petrie, K. J., Horne, R., Cameron, L. D., and Buick, D. (2002). The Revised Illness Perception Questionnaire (IPQ-R). Psychol. Health 17, 1–16. Munoz-Furlong, a (2003). Daily Coping Strategies for Patients and Their Families. Paediatrics 111(6), 1654–1661. Noojin, A. B. and Wallander, J. L. (1997). Perceived problem-solving ability, stress, and coping in mothers of children with physical disabilities: Potential cognitive influences on adjustment. Int. J. Behav. Med. 4, 415–432. Nosek, B. A. (2007). Implicit-explicit relations. Curr. Directions Psychol. Sci. 16, 65–69. Ortega, A., Huertas, S., Coniro, G., Ramiraz, R., and Rubio-Stipec, M. (2002). J. Nerv. Ment. Dis. 190, 275–281. Osman, M. (2003). Therapeutic implications of sex differences in asthma and atopy. Arch. Dis. Child 88, 587–590. Primeau, M. N., Adkinson, N. F. Jr., and Hamilton, R. G. (2001). Natural rubber pharmaceutical vial closures release latex allergens that produce skin reactions. J. Allergy Clin. Immunol. Jun, 107. Primeau, M. N., Kagan, R., Joseph, L., Lim, H., Dufresne, C., Duffy, C., Prhcal, D., and Clarke, A. (2000). The psychological burden of peanut allergy as perceived by adults with peanut allergy and the parents of peanut-allergic children. Clin. Exp. Allergy 30, 1135–1143. Pumphrey, R. (2000). Anaphylaxis: Can we tell who is at risk of a fatal reaction? Curr Opin Allergy Clin Immunol 4, 285–290.
Developmental Trajectories in Food Allergy
99
Rohleder, N., Shommer, N. C., Hellhammer, D. H., Engel, R., and Kirschbaum, C. (2001). Sex differences in glucocorticoid sensitivity of proinflammatory cytokind production after psychosocial stress. Psychosomatic Med. 63, 966–972. Rosencranz, M. A., Busse, W. W., Johnstone, T., Swensson, C. A., Crisafi, G. M., et al. (2005). Neural circuitry underlying the interaction between emotion and asthma symptom exacerbation. Proceedings of the Natural Academy of Scientists 102(37), 13319–13324. Sampson, H. A. (2003). Food allergy. J. Allergy Clin. Immunol. 111, S540–S547. Sampson, H. A. (2004). Update on food allergy. J. Allergy Clin. Immunol. 113, 805–19. Sampson, H. A. (2005). Food allergies. In Sleisenger and Fordtran’s Gastrointestinal and Liver Disease, (M. Feldman, et al., Eds), 8th Ed., Vol. I, pp. 427–439. Saunders, Philadelphia. Sampson, M., Munoz-Furlong, A., and Sicherer, S. H. (2006). Risk-taking and coping strategies of adolescents and young adults with food allergy. J. Allergy Clin. Immunol. 117(6), 312–318. Schmidt, S. (2003). Coping with chronic disease from the perspective of children and adolescents—a conceptual framework and its implications for participation. Child: Care Health Dev. 29(1), 63–75, Blackwell Publishing Ltd. Schreurs, K. M. G. and de Ridder, D. T. D. (1997). Integration of coping and social support perspectives: Implications for the study of adaptation to chronic diseases. Clin. Psychol. Rev. 17, 89–112. Sicherer, S. H. and Sampson, H. A. (2006). Food allergy. J. Allergy Clin. Immunol. 117, S470–5. Sicherer, S. H. (2002). Food allergy. Lancet 360, 701–710. Sicherer, S. H., Noone, S. A., and Munoz-Furlong, A. (2001). The impact of childhood food allergy on quality of life. Ann. Allergy Asthma Immunol. 87, 461–464. Skinner, E. A. and Wellborn, J. G. (1994). Coping during childhood and adolescence: A motivational perspective. (D. Featherman, R. Lerner, and M. Perlmutter, Eds.), Vol. 12, pp. 91–133. Erlbaum, Hillsdale, NJ. Sweeting, H. and West, P. (2003). Sex differences in health at ages 11, 13 and 15. Soc. Sci. Med. 56, 31–39. Timonen, M. (2003). The association between atopic disorders and depression: The Northern Finland 1966 Birth Cohort Study. Acta Universitatis Ouluensis Medica. Van der Pligt, J. (1994). Risk appraisal and health behaviour. In Social Psychology and Health: European Perspectives, (D. R. Rutter and L. Quine, et al., Eds), pp. 131–152. Avebury, Aldershot, England. Van der Pligt, J. (2002). Cognition and affect in risk perception and risky decision-making. In ‘‘Psychology at the turn of the millennium, Social, developmental, and clinical perspectives,’’ (C. von Hofsten and L. Ba¨ckman, Eds.), Vol. 2, pp. 247–270. Psychology Press, Hove, England. Varni, J. W. and Wallander, J. L. (1998). Effects of pediatric chronic physical disorders on child and family adjustment. J. Child Psychol. Psychiatry 39, 29–46. Varni, J. W., Wilcox, K. T., and Hanson, V. (1988). Mediating effects of family social support on child psychological adjustment in juvenile rheumatoid arthritis. Health Psychology 7(5), 421–431. Varni, J. W., Setoguchi, Y., Rappaport, L. T., and Talbot, D. (1991). Effects of stress, social support, and self-esteem on depression in children with limb deficiencies. Arch. Phys. Med. Rehabilitation 72, 1053–1058. Varni, J. W., Katz, E. R., Colegrove, J. R., and Dolgin, M. (1994a). Perceived social support and adjustment of children with newly diagnosed cancer. J. Dev. Behav. Pediatr. 15, 20–26. Varni, J. W., Rapoff, M. A., and Waldron, S. A. (1994b). Juvenile rheumatoid arthritis: Psychological issues. In ‘‘The Sourcebook of Pediatric Psychology,’’ (R. A. Olson, L. L. Mullins, J. B. Gillman, and J. M. Chaney, Eds.), pp. 75–89. Allyn & Bacon, Boston.
100
A. DunnGalvin and J’ O. B. Hourihane
Varni, J. W., Katz, E. R., Colegrove, J. R., and Dolgin, M. (1994c). Perceived social support and adjustment of children with newly diagnosed cancer. J. Dev. Behav. Pediatr. 15, 20–26. Vatn, M. H. (1997). Food intolerance and psychosomatic experience. Scand. J. Work Environ. Health 23(Suppl. 3), 75–78. Von Berg, A., Koletzko, S., Grubl, A., et al. (2003). The effect of hydrolyzed cow’s milk formula for allergy prevention in the first year of life: The German Infant Nutritional Intervention Study, a randomized double-blind trial. J. Allergy Clin. Immunol. 111, 533–534. Von Weiss, R. T., Raposs, M. A., Varni, J. W., Lindsley, C. B., Olsen, N. Y., Madsen, K. L., et al. (2002). Daily Hassles and Social Support as Predictors of Adjustment in Children With Pediatric Rheumatic Disease. J. Pediatr. Psychol. 27(2), 155–165. Wallander, J. L. and Varni, J. W. (1998). Effects of pediatric chronic physical disorders on child and family adjustment. J. Child Psychol. Psychiatry 39, 29–46. Wiebe, D., Berg, D., Palmer, D., Korbel, C., and Beveridge, R. (2002). Illness and the Self: Examining Adjustment Among Adolescents with Diabetes. Washington DC: Paper presented at the Annual Meeting of the Society of Behavioural Medicine. Wizeman, T. M. and Pardue, M.-L. (Eds.), for the Committee on Understanding the Biology of Sex and Gender Diffrences (2001). Exploring the Biological Contributions to Human Health: Doex Sex Matter? Board on Health Sciences Policy, Institute of Medicine, National Academy Press, Washington, DC. Wright, R. (2005). Stress and atopic disorders. J. Allergy Clin. Immunol. 116(6), 1301–1306. Wright, R. J., Cohen, R. T., and Cohen, S. (2005). The impact of stress on the development and expression of atopy. Current Opinion in Allergy and Clinical Immunology 5(1), 23–29.
CHAPTER
4 Maple Syrup—Production, Composition, Chemistry, and Sensory Characteristics Timothy D. Perkins and Abby K. van den Berg
Contents
I. II. III. IV. V. VI. VII.
Introduction History Maple Sap Flow Sap Collection Sap Processing: Evaporation Annual Syrup Production Sap Chemistry A. Transformation during storage B. Transformations during reverse osmosis/nanofiltration C. Transformations during evaporation by heating VIII. Scale/sugar Sand Formation During Sap Processing IX. Syrup Standards X. Syrup Chemistry A. Density B. Carbohydrates C. pH D. Conductivity E. Color F. Rheology G. Inorganic composition H. Organic acids I. Flavor compounds J. Sensory evaluation of flavor
102 102 104 106 107 110 111 115 116 117 121 124 126 126 126 126 127 127 127 128 131 131 133
Proctor Maple Research Center, University of Vermont, Harvey Road, Underhill Center, Vermont, USA Advances in Food and Nutrition Research, Volume 56 ISSN 1043-4526, DOI: 10.1016/S1043-4526(08)00604-9
#
2009 Elsevier Inc. All rights reserved.
101
102
Timothy D. Perkins and Abby K. van den Berg
K. Off-flavors in maple syrup L. Nutritional aspects of maple syrup XI. Other Maple Products XII. Contamination XIII. Adulteration XIV. Summary References
Abstract
134 135 136 137 137 140 140
Maple syrup is made from sap exuded from stems of the genus Acer during the springtime. Sap is a dilute solution of primarily water and sucrose, with varying amounts of amino and organic acids and phenolic substances. When concentrated, usually by heating, a series of complex reactions produce a wide variety of flavor compounds that vary due to processing and other management factors, seasonal changes in sap chemistry, and microbial contamination. Color also forms during thermal evaporation. Flavor and color together are the primary factors determining maple syrup grade, and syrup can range from very light-colored and delicate-flavored to very dark-colored and strong-flavored.
I. INTRODUCTION Maple syrup is produced from the sap of several species of maple (Acer), chiefly through the concentration of sap via thermal evaporation. Although the chemistry of maple syrup is dominated by sucrose, a wide variety of sap collection and processing factors, microbiological interactions in sap, environmental influences, as well as the packing and storage of the finished product, combine to produce a range of chemistry and flavor profiles in maple syrup. Because of the large concentration factor (40 gal of sap are required to produce 1 gal of syrup) and often delicate flavor profiles involved, several off-flavors are commonly found. Finally, the large price differential between maple syrup and other sweeteners provides incentive for adulteration.
II. HISTORY Several different legends describe how Native Americans discovered that the sap of maple trees was sweet and could be boiled down to form maple sugar (Heiligmann et al., 2006). The most likely explanation is that they observed birds and animals cutting holes or gashes into the twigs of trees, or drops of sap falling after branch breakage by snow or wind. These small wounds ooze sap in the spring, forming small drops of sap that are
Maple Syrup—Production, Composition, Chemistry, and Sensory Characteristics
103
concentrated by the sun and wind, or form sweet icicles of sap. The Native Americans undoubtedly recognized this and collected sap by cutting slashes in the trunks of maple trees and used skins, hollowed wooden vessels or clay pots to collect the sap, which was then concentrated by boiling to form a very dark and strong-tasting sugar. At certain times of the year, maple sugar could comprise a significant portion of their total caloric intake. Early colonists throughout New England took up the practice of making maple sugar due to the high cost of imported sugars, and because the practice occurs at a time of year when other agricultural endeavors are not possible. Spouts made of hollowed out stem sections of elder or sumac twigs were inserted into holes cut in the trunks of maple trees with chisels. Later, metal spouts were produced and used in holes drilled with augers. Sap was first collected into hollowed-out tree trunks, then later wooden or metal buckets. In most cases, the final product was maple sugar (a solid), and only a relatively small amount of maple syrup (liquid) was produced. Over the next few hundred years, the practice of tapping trees and collecting sap changed considerably. While some maple producers continue to use metal spouts and metal buckets to collect sap, plastic spouts and tubing are now relatively common. Some experimentation with metal tubing began in the late-1800s and early-1900s, however, the first successful commercial tubing systems arose in the 1950s and early-1960s with the introduction of plastic (PVC) tubing and associated plastic (Nylon) spouts and fittings to collect and transport sap to a central location, greatly reducing the labor required to collect sap. Initially tubing systems were run across the ground and vented, but continued experimentation and use resulted in tubing lines being suspended, unvented, and a drop line introduced to reduce reabsorption of sap by trees further down the line in the collection system. Shortly after tubing came into use, some researchers and producers began attempting to augment sap yield by applying vacuum to the tubing systems. The early results were encouraging, but maple sap yield was greatly bolstered with the advent of a new generation of tubing composed of Polyethylene, along with associated changes in spouts and fittings, and increased use of vacuum pumps designed for maple applications in the mid-1990s. The end result is that current high-yield production methods can achieve sometimes double or more the standard yield from buckets or gravity (nonvacuum) tubing installations. The processing of sap into maple syrup has also changed greatly from colonial days. Early settlers used a batch method to boil sap in large kettles over open fires. This required a very long time and huge quantities of wood to produce a very dark and strong-tasting maple sugar with a moderate to substantial load of impurities. Modern maple evaporators
104
Timothy D. Perkins and Abby K. van den Berg
provide relatively continuous processing of sap, are very energy efficient, and generally produce a much lighter-colored and lighter-flavored maple syrup fairly quickly. In addition to changes in collection and processing, the product itself has also changed. Presently, most maple is made into and marketed as syrup (liquid), with much smaller sales of maple sugar, cream, or candy than was historically the case.
III. MAPLE SAP FLOW Under the appropriate conditions, a sweet sap can be collected from most maple (Acer) species. In general, however, only the sap of sugar, black, and red maple is commonly used to make maple syrup. Where maple trees are found in abundance and weather conditions are appropriate, commercial maple production can occur. This ranges from Nova Scotia to Minnesota from east to west, and from southern Ontario and Quebec in the north to areas of West Virginia in the south (Heiligmann et al., 2006). Boxelder is sometimes tapped in areas of Manitoba and the Pacific northwest. The physiological process responsible for sap flow in maple trees probably results from a combination of physical and osmotic forces (Cirelli et al., 2008; Milburn and O’Mally, 1984; Tyree, 1983). In the physical model, fluctuations in wood temperature that span the freezing point during the leafless period (fall or spring) create alternating negative and positive pressures within the trunk and branches. When wood temperature falls below freezing, the water vapor within the billions of airfilled lumen of fiber cells freezes, forming a frost-like layer on the inside of the cell wall. Since the vapor pressure is much lower over ice than liquid water, this, and to a much lesser degree the contraction of the air bubble, create a vapor pressure gradient, causing water to move apoplastically (along cell walls) into the lumen, where it continues to freeze. Due to strong cohesion, and the vapor pressure gradient, water is pulled up through vessel elements towards the lumen. Eventually, the entire wood (fibers and vessel elements) and sap freezes. Freezing occurs first in the fine branches in the crown of the tree, then progressively downward. The amount of water uptake is dictated by soil water availability and the rate of freezing. A slow freeze ensures maximum uptake, whereas in a rapid freeze vessel elements in the stem of the tree may freeze before water uptake is complete. During the warming phase, as the wood increases in temperature above the freezing point, the frost layer thaws, and the gas bubble expands. Sap pressure at the stem level increases very rapidly to a peak pressure, which is largely caused by gravitational potential, and
Maple Syrup—Production, Composition, Chemistry, and Sensory Characteristics
105
somewhat also by gas bubble expansion. This pressure may reach up to 40 psi (275 kPa). Over time, the pressure slowly recedes as sap is forced out of small wounds or into other areas of the tree, until the pressure within the tree equals the air pressure outside the tree, at which point flow ceases. The flow rate and total yield from tapholes is proportional to tree size and to the pressure gradient, thus sap flows faster earlier in a ‘‘run’’ than later. Root pressure is not a significant factor in maple sap exudation. Recent evidence (Cirelli et al., 2008) clearly demonstrates that there is also a considerable osmotic component to the development of sap pressure in maple due to anatomical barriers to sucrose between the vessel system and fibers. Further work is necessary to determine the precise contribution of physical and osmotic factors on sap exudation. Maple producers exploit the sap flow phenomenon during the time of year when temperatures are expected to fluctuate around the freezing point by drilling small holes into the stem, inserting spouts, and collecting the sap in some fashion. Only trees that have reached a certain diameter (10–12 in. at breast height) are generally tapped. This ensures that the tree will be able to withstand the stress of tapping and regrow sufficient wood during the growing season to compensate for the loss due to tapping and the accompanying zone of discoloration (walling off, a normal wound response in trees to limit microbial infection). Sap will typically only flow for 1–2 months before microbial contamination of the taphole, or the lack of proper flow conditions (freeze–thaw) cause the flow to cease. In general, each taphole will produce about 10–20 gal of sap during the season, depending upon the collection technology employed, the environmental conditions during the season, and the size and sap sugar content of the tree. Although sap will flow in both the fall and spring of the year, the vast majority of maple production occurs in the spring for several reasons. Sap in the spring is sweeter than in the fall, and decreasing temperatures as the season progresses from fall to winter can cause damage to equipment (split bags and buckets due to frozen sap) and frost-heaving of spouts out of tapholes. Trees should also not be tapped more than once per year. Sap sugar content is not high at all times of the year. There is a strong seasonal pattern of production, accumulation, and utilization of nonstructural carbohydrate forms in maple. Starch, the dominant form of reserve carbohydrate in sugar maple, tends to be quite low during the photosynthetic period, and accumulates in the stem and twig wood towards the end of the growing season. Soluble sugars tend to increase during the winter and early spring as a function of temperature (Cortes and Sinclair, 1965). Sucrose is clearly the dominant soluble sugar in the xylem, with only minor amounts of glucose and fructose. Still lesser amounts of stachyose, raffinose, and xylose are also present (Wong et al., 2003).
106
Timothy D. Perkins and Abby K. van den Berg
IV. SAP COLLECTION Sap may be collected into galvanized or aluminum buckets or plastic bags, all of which require periodic (up to several time daily during high flow periods) emptying. When buckets or bags are used, maple producers must collect the sap into a larger, generally mobile container and bring it to the site where it is processed. Sap may also be collected with a network of plastic tubing (Fig. 4.1). Collection with tubing generally does not require visiting each tree during the season after the intial tapping, as the sap flows through progressively larger tubing into a holding tank, often located at the site of sap processing into syrup. When using tubing, proper design, layout, and maintenance of the tubing system must be observed in order to maximize sap yield. The general rule for installing a tubing system is ‘‘tight, straight, and downhill.’’ Other factors that affect collection with tubing include the size of tubing, the number of taps on a lateral line, tubing layout, and several other considerations. Sap yields on gravity (buckets, bags, or tubing without vacuum) average about 8–10 gal of sap (0.20–0.25 gal of syrup equivalent) per tap over a season. Vacuum is used in many modern tubing installations, with a pump evacuating the tubing system to a level of 20–25 in. mercury (in Hg). By applying vacuum to the tubing network, the pressure gradient between the inside of the stem and the ambient air inside the tubing increases, resulting in a much higher sap flow rate and sap yield. This practice does
FIGURE 4.1 Maple tubing system. Spout is inserted into a hole drilled into the tree (on upper right), a dropline extends downward from the spout, and connects to the lateral 5/16 in. tubing line. The lateral line runs from tree to tree, and connects to the mainline (1 in. diameter pipeline in this case) which is suspended on tensioned steel wire.
Maple Syrup—Production, Composition, Chemistry, and Sensory Characteristics
107
not significantly impact sap sugar content, chemistry, or the amount of internal tree damage (Wilmot et al., 2007a,b). The primary purpose is to augment sap yield, and to encourage sap flow during time periods when flows might otherwise be marginal. Sap yields on a well-designed and operated vacuum tubing system can reach 25 gal per tap (0.6 gal of syrup equivalent) each season. Tapholes are drilled into tree trunks at a height of 3–7 in. above the soil within an area that is free of visible damage or older tapholes (which become progressively harder to see as the tree grows). Tapholes are normally 7/16 in. in diameter, but 5/16 in. diameter spouts introduced in the mid-1990s produce similar yields under vacuum with less tree damage (Wilmot et al., 2007a) are gaining widespread acceptance. Tapholes are drilled to 1.5–2.5 in. deep into the wood at a slight uphill angle to allow sap to naturally flow out of the taphole. A metal spout (if using buckets or bags) or a plastic spout is placed in the hole and lightly hammered in place. Spouts are generally removed at the end of the maple sugaring season to allow the tree to heal. Several attempts have been made to increase the length of time sap flows from the taphole. Paraformaldehyde (PFA) was used for a few decades as a microbicide to reduce taphole ‘‘drying.’’ PFA use was phased out when it was determined to be harmful to the tree, and the use of any microbicide is currently illegal in all maple production areas. More recently, a denatured ethanol has been promoted to ‘‘sterilize’’ tapholes in some areas, but the lack of any lasting impact on microbial populations has limited its effectiveness in producing higher sap yields; in fact, some research has shown the practice to significantly reduce sap flow from such ‘‘sterilized’’ tapholes. There is considerable ongoing research into how to maximize sap yield in maple production, particularly in vacuum tubing operations. Because sap is a perishable product, it is generally processed relatively soon after collection to minimize microbial contamination and the accompanying reduction in syrup quality. Sap is often filtered and UV-sterilized after collection to reduce microbial loads in order to maintain sap quality. Ozone treatment, although useful in water treatment, does not appear to be effective in maple applications, presumably due to the strong protective effect of sugars on microbial populations (Labbe et al., 2001).
V. SAP PROCESSING: EVAPORATION After collection, maple sap must be transformed through some means of concentration into maple syrup. The two major processes utilized are evaporation by heating and reverse osmosis followed by heating. The modern maple evaporator (Fig. 4.2) is typically composed of several
108
Timothy D. Perkins and Abby K. van den Berg
Sap inlet 2
1
Floatbox
3 4
5 6 Syrup drawoff
FIGURE 4.2 A contemporary maple syrup evaporator. Sap follows a winding path from the sap inlet, sequentially through the pan sections to the syrup drawoff.
parts: a heat source, an arch to contain and concentrate the heat, and pans, which contain the liquid and allow it to become concentrated. Most evaporators utilize wood or oil as fuel, and commercial evaporators (as opposed to hobby-sized units) are insulated and efficient. Although wood was a common fuel source historically, it is being largely supplanted by oil due to the high cost of labor and convenience. Oil is easy to move around (as opposed to wood) and is easy to control (nearly instant on and off, with high and low settings). Most commercial evaporators are oil-fired, although some very large operations use high pressure steam to achieve a very rapid processing rate and to avoid scorching of evaporator pans. The arch is made of cast iron or steel and sits under the pans to contain the heat. Oil-fired units have the burner located at the front, slanting upward, and with the fire pointing towards the rear (stack). The inside of an oil-fired arch is lined with ceramic blanket insulation. Wood-fired
109
Maple Syrup—Production, Composition, Chemistry, and Sensory Characteristics
units are usually lined with fire-brick and contain metal grates, sometimes with forced ventilation. A wood-fired evaporator has doors in the front through which wood is added. The hot gases flow back towards the rear of the unit and stack. Pans, normally composed of stainless steel, sit on top of the arch. Today, nearly all evaporator pans are Tungsten Inert Gas (TIG) or Metal Inert Gas (MIG) welded, and conform to widespread food industry construction practices. Formerly, most pans were soldered, and constructed of stainless steel, although sometimes older units were composed of English tin, or less frequently, copper. Prior to 1994, most solder contained lead; after 1994 lead-free silver solder (tin-silver) was used for pan fabrication. Considerable attention has been recently focused on developing and promulgating standards for maple equipment manufacturing (LMEA, 2001). Most evaporators have two distinct types of pans, the back pan (also called the flue pan or sap pan) and the front pan (or syrup pan) (Figs. 4.2 and 4.3). The back pan, where the majority of the water is evaporated from the sap, has deep flues that are designed to maximize heat exchange and evaporation. Back pans come in two configurations, drop-flue and raisedflue, based upon whether the flues in the pan extend above or are even with the position of the pan on the arch rails. Neither type is clearly dominant in the industry. Sap enters into the back pan; generally via gravity feed through a pipe connected to the sap storage tank and is regulated by a mechanical float or an electronic valve. The back pan is internally divided into two or more sections which result in a semichanneled flow from the sap inlet to an outlet near the front pan. Frontpan section
Backpan section 11%
47% 29% 19% 15%
Syrup drawoff
Exhaust flue
Oil burner
62%
Sap inlet 2%
66–67%
FIGURE 4.3 Schematic diagram of a maple evaporator. Numbers and shading density show the density ( Bx) of liquid within different partitions of the evaporator during evaporation. Sap enters the back pan section of the evaporator at the lower right and flows through a feed-pipe into the partition at the upper right. It then flows around the back, and then forward. Two front pans contain two partitions each. Sap flows into the front pan section through a pipe connecting the back pan to the front pan at the lower middle. Sap/syrup then flows through each front pan partition, increasing in density and developing color and flavor during the process. Syrup exits the evaporator at the lower left via an automatic drawoff, which opens when the density of syrup close to the drawoff area is correct, then closes again when the density drops below that of finished syrup.
110
Timothy D. Perkins and Abby K. van den Berg
Sap flows from the back pan into the front pan, sometimes using another float or valve to regulate the entry of sap. The bottom of the front pan is flat, and the front ‘‘pan’’ may actually be made up of one or more pans, with each separate pan itself divided into several partitions. Front pans are made in one of two configurations: those in which the sap runs parallel to the main axis of the evaporator, termed reverse-flow evaporators, and pans in which the sap flows from side-to-side, termed cross-flow evaporators. In both cases, syrup flows out the last partition of the front pan via a manual or automatic draw-off valve into a pail or directly into a pipeline feeding a filtration system. Modern maple evaporators are designed to have a relatively continuous or semicontinuous flow. In theory, an evaporator can be envisioned as a long continuous stainless steel ‘‘gutter,’’ with heat applied under the entire surface. Sap flows in one end of the gutter (the back pan), and syrup flows out of the other (draw-off of the front pan). Because it is impractical to have an evaporator that is tens of feet long, the gutter is bent into sections to reduce the overall size, and to allow a single heat source to be located under the evaporator pans. Commercial evaporators are commonly sized to fit the number of taps in the sugaring operation, and range in size from 3 ft. 10 ft. (width length) up to 6 ft. 18 ft., with the back pan section typically occupying about two-thirds of the overall evaporator surface area. The depth of sap in the pans is generally kept quite low, only 1.5–2.0 in., to maximize boiling rate. Ancillary equipment is often used in conjunction with evaporators. Hoods are employed to channel steam out of the processing facility (termed a sugar house, sugar shack, maple house, etc.). Several types of devices are manufactured which sit over the evaporator pans to take advantage of the steam energy to preheat or preconcentrate the sap before it enters the back pan. Often syrup is drawn off at a density slightly lower than of finished maple syrup, and is brought to final density in a separate finishing pan. Finished syrup is filtered to remove solids and generate a clear product. Filtration through a wool or synthetic cone filter is sometimes used in smaller operations; most modern commercial production scale operations employ a pressure filter utilizing diatomaceous earth as the filtering media. Bulk syrup is hot-packed in 30–50 gal drums (stainless or galvanized steel, epoxy lined steel, or plastic) for storage before being reheated and packaged into retail containers as needed.
VI. ANNUAL SYRUP PRODUCTION The yearly worldwide production of maple syrup is roughly 8–9 million gal (45,000 metric tons), of which 15% is produced in the United States, and 85% produced in Canada. The New England/New York region
Maple Syrup—Production, Composition, Chemistry, and Sensory Characteristics
111
cumulatively produce about 75% of the total domestic U.S. crop. Quebec is the largest Canadian producer, with about 90% of the total Canadian production originating there. The largest markets of maple are in the U.S., Canada, Europe, and, increasingly, Asia.
VII. SAP CHEMISTRY Maple sap is a dilute solution of mainly water and sugar, along with trace amounts of other substances, including organic acids, free amino acids, protein, minerals, and phenolic compounds. Although the proportions are somewhat variable, sap is normally composed of 95–99% water and 1–5% sugar. The primary sugar found in uncontaminated sap is sucrose, which ranges from 96–99% of the total sugar present (Table 4.1). As microorganisms contaminate sap, particularly later in the sap flow season, invert (reducing) sugar levels increase, reaching up to about 0.20–0.25% of the total sugar concentration in sap (Dumont, 1994; Heiligmann et al., 2006). At times, several other sugar forms, including mono-, di-, tri-, and higher oligosaccharides, may also be found in maple sap (Dumont, 1994; Haq and Adams, 1961; Stinson et al., 1967), although typically in very low concentrations. Sap pH ranges from 3.9 to 7.9, but in most cases is only slightly acidic, with a typical range of 6.5–7.0. There is a slight trend for sap to become more acidic as the sap flow season progresses (probably due to microbial action). Conductivity normally ranges from 320 to 520 mS/cm. The total concentration of sugars in sap varies due to several factors. Variation between individual trees may be quite large due to differences in genetics, growth rate, and crown density. However, the ranking of any one tree relative to its neighbors tends to remain relatively constant both within a season and from season to season (Taylor, 1956). TABLE 4.1 Sugar composition (% dry weight) of the solid fraction of maple sap (from Perkins et al., 2006, used with permission) Sucrose
96–99
Polysaccharides Oligosaccharides Glucose Fructose Quebrachitol Unidentified
nd— 0.5 nd—0.02 nd—0.17 nd—0.10 nd—0.15 nd—0.67
112
Timothy D. Perkins and Abby K. van den Berg
Sap sugar content within an individual tree may vary from one season to another to a relatively high degree, probably as a result of photosynthetic carbohydrate gain in the prior growing season. Sap sugar content also varies within a season, typically decreasing throughout the season. Although at one time, it was believed that the use of tubing systems with vacuum might result in a dilution of sap sugar content and increased internal wounding in trees, recent research has shown that this is not the case. Wilmot et al. (2007b) demonstrated that sap collected at high vacuum levels (>18 in. Hg) did not result in diluted sugar content, did not contain significantly different levels of minerals, and did not cause larger zones of discoloration in maple stems. The average sugar content of a forest stand utilized for maple production has a large impact on the economics of maple production, as a sugarbush containing a higher level of sugar in the sap will require considerably less energy to produce 1 gal of syrup compared to a sugarbush with a lower sugar content. Maple producers can increase sap sugar content by selecting individual trees with higher sap sugar content during thinning, and by encouraging good crown and stem growth through crop tree management techniques. Fertilization of trees, although not a common practice in maple operations for a number of reasons, can be used to correct nutritional deficiencies and stimulate growth (Perkins et al., 2004a), and may increase sugar yield from a site (Perkins et al., 2004b). The inorganic composition of maple sap is highly variable (Table 4.2). Potassium and calcium make up the bulk of the inorganic fraction of sap, with substantial amounts of magnesium and manganese, and only trace amounts of sulfur, phosphorus, zinc, and copper, followed by aluminum, sodium, boron, and iron. In contrast to what is observed in sugar concentration, most of these elements show a slight or strong tendency to increase in concentration during the sap flow season (Marvin and Greene, 1959). Lead is generally below detection limits unless contaminated by lead-containing equipment. The gases expressed from maple stems during sap exudation show higher carbon dioxide and correspondingly lower oxygen concentrations compared to ambient air, indicating substantial respiratory activity in maple wood during the leafless period of late-winter to early-spring (Marvin and Greene, 1959). A wide range of free amino acids are found in sterile maple sap (Heiligmann et al., 2006), including glycine, alanine, asparagines, threonine, leucine, isoleucine, valine, and methionine. Morselli and Whalen (1986) examined the change in the distribution of various amino acids over two maple sap seasons. Their results indicated that initially, only a small number (6–7) of amino acids were found in sap, all in relatively low concentration. As the season progressed, the diversity of amino acids increased to 12–15. In addition, the concentration of amino acids present
Maple Syrup—Production, Composition, Chemistry, and Sensory Characteristics
TABLE 4.2
Inorganic (minerals and metals) composition of maple sap
Element
Potassium Calcium Magnesium Manganese Sulfur Phosphorus Zinc Copper Aluminum Sodium Boron Iron Lead a b c
113
Rangea (ppm)
27–95 21–77 2.6–9.0 2.7–9.7 2.5–8.0
nd—4.0 nd—3.7
Range (ppm)
Meanb (ppm)
Meanc (ppm)
50–81 35–75 3.9–8.1 1.7–5.5 0.18–1.80 0.2–1.1 0.24–1.47 0.08–1.56 0.04–0.20 0.02–0.17 0.02–0.16 0.01–0.07 nd
65 50 5.6 3.5 0.77 0.65 0.55 0.50 0.10 0.08 0.08 0.04 nd
25 40 3
b
Marvin and Greene (1959). van den Berg and Perkins (unpublished). Dumont (1994).
in late season sap may rise to over 10 ppm (Dumont, 1994), probably reflecting the breaking of winter dormancy and an acceleration of metabolic activity in the trees. These changes could have a significant impact on the characteristics of syrup produced, as the nitrogen content of sap, primarily amino-N (Pollard and Sproston, 1954), has a large influence on the development of flavor and off-flavor compounds in maple syrup. Uncontaminated maple sap has the same appearance as water in that it is nearly colorless, with very high light transmission through the visible range (Fig. 4.4). Some absorption is found in sections of the UV and near infrared (NIR) ranges. Several organic acids are also found in maple sap (Table 4.3). In general, the total quantity of organic acids starts out low, and rises throughout the sap flow season. Malic acid (concentration 800–45,000 ppb) is by far the most common organic acid, ranging from just over 50–99% of the total acid present. Succinic acid and oxalic acid are also fairly dominant forms. Other acids occur sporadically in low concentration (Dumont, 1994; Mollica and Morselli, 1984). A wide range of phenolic compounds, varying in type and concentration, can also be found in maple sap (Dumont, 1994). Most of these appear to be derived from lignin (Kermasha et al., 1995). They range in concentration in sap up to 0.1 ppm. The most dominant phenols present tend to
114
Timothy D. Perkins and Abby K. van den Berg
100 90 Light transmittance (%)
80 70 60 50 40 30 20 10 1090
1040
990
940
890
840
790
740
690
640
590
540
490
440
390
340
290
240
190
0
Wavelength (nm)
FIGURE 4.4 Transmission profile of maple sap at 2 Bx from 190–1100 nm. Sap was collected on March 29, 2006, then stored frozen until analyzed. TABLE 4.3
Organic acids in maple sap (Dumont, 1994).
Organic Acid
Mean (ppb)
Malic Fumaric Succinic Oxalic Aconitic Citric Tartaric Total
14,940 1677 910 380 193 109 59 18,033
be: sinapic acid, coumaric acid, syringaldehyde, and coniferaldehyde, with lesser amounts of vanillic acid, syringic acid, homovanillic acid, and ferulic acid (Table 4.4). No clear temporal trends in total phenol levels are apparent (Dumont, 1994), although some work does suggest that variation in vanillin glycoside concentrations are correlated with seasonal changes during the sap flow period (Belford et al., 2006). A variety of flavonoids, including catechin, flavanols, and dihydroflavonols, are also found in maple sap (Deslauriers, 2000). Some of these exhibit strong seasonal tendencies.
Maple Syrup—Production, Composition, Chemistry, and Sensory Characteristics
115
TABLE 4.4 Phenolic compounds in maple sap (after Dumont, 1994)
a
Compound
Mean (ng/ml)a
Sinapic acid Coumaric acid Syringaldehyde Coniferaldehyde Vanillin Vanillic acid Homovanillic acid Syringic acid Ferulic acid Coniferol Total
31.8 15.0 14.9 13.4 9.7 9.5 8.4 3.9 2.4 1.6 110.7
Normalized to 1 Bx.
Many of these compounds are likely to be flavor precursors, although the influence of each on the final flavor profile of the resulting maple syrup is greatly affected by storage, as well as the type and length of processing. Because there are a number of transformations during processing, and many of these compounds are volatile when heated, the concentration is not necessarily increased in maple syrup. There is an increasing interest in the quantity and composition of phenols in maple sap and syrup, due to the antioxidant, antiradical, and antimutagenic activities of these compounds (The´riault et al., 2006).
A. Transformation during storage While in the xylem, sap generally is considered to be sterile. Immediately upon being exposed to taphole conditions, it is acted upon by a wide variety of microorganisms, including bacteria, fungi, and yeasts. The major impact of this contamination on the syrup making process is the conversion of a small quantity of sucrose by invertase. Because it is colder and all the equipment is clean, sap collected early in the maple sap flow season tends to be very low in microbial load, and thus low in invert sugar concentration. As the season progresses, daily temperature tends to increase. This results in the sap collecting system becoming colonized with microorganisms, increasing the invert level of the sap. Some producers use intermediate season tubing washes or rinses to reduce microbial contamination, although the efficacy of mid-season cleaning is most likely relatively minor and short-lived.
116
Timothy D. Perkins and Abby K. van den Berg
Proper filtering of sap, cool storage before processing, and rapid processing are commonly used to reduce microbial growth in sap. Growth of microorganisms may slightly or significantly alter the form of several elements in sap, especially by causing the forms of nitrogen and phosphorus to be changed, by increasing the protein component in sap, and by altering the turbidity of sap. Sap that has been extensively colonized by yeasts may undergo fermentation. Extremely high levels of microorganisms, especially at the end of the season, may cause syrup to become ‘‘ropey,’’ a sticky, stringy, gelatinous texture that is almost impossible to remove, thus rendering it essentially unmarketable.
B. Transformations during reverse osmosis/nanofiltration Due to the high cost of fuel, and to reduce the time required to process sap into maple syrup, an increasing number of maple producers are using reverse osmosis/nanofiltration (hereafter collectively termed RO) to increase the sugar concentration of sap prior to boiling. Sap processed via RO is called ‘‘concentrate,’’ and tends to be slightly yellow in color. Although early units were adapted from water desalination units, commercial RO machines specialized for the maple industry are now available from a variety of manufacturers. There is also a strong trend towards increasing levels of concentration of sap with RO. In the past, most maple producers were content to concentrate sap to 8 Bx. Currently, many producers are striving to concentrate to the highest degree possible (25 Bx). The major effect of preconcentration of sap via RO is the removal of water. Going from 2 to 8 Bx achieves a 75% reduction in the amount of evaporation necessary (from about 43 gal of sap to about 11 gal to produce a gallon of maple syrup). A further concentration of sap to 16 Bx would require only 5.5 gal of sap to produce 1 gal of maple syrup (Fig. 4.5). Concentrate is extremely perishable. Most maple operations utilize RO machines that are capable of providing only slightly more concentrate per hour than can be utilized by the evaporator. In practice, most producers will concentrate a small amount of sap, start the evaporator, and continue to run the RO during boiling so that the concentrate is not allowed to build up and spoil. The by-product of sap concentration by an RO, permeate, is also used in maple operations as a source of very clean water. Due to the low mineral concentration of permeate water, it is used for cleaning tubing and evaporator equipment, as well as the RO membrane itself, which should be run through a wash and rinse cycle after each use. The chemicals and dosage to be used are specified by the membrane and RO manufacturers, and should be carefully followed to avoid damaging the membrane or contaminating the sap concentrate.
117
Maple Syrup—Production, Composition, Chemistry, and Sensory Characteristics
60 Sap
55
Concentrate
Syrup
50 45
Sap (gallons)
40 35 30 25 20 15 10 5 0 0
5
10
15
20
25
30
35
40
45
50
55
60
65
70
Density (⬚brix)
FIGURE 4.5 The relationship between the density and the number of gallons of sap required to produce 1 gal of maple syrup. The normal range of raw sap, sap concentrated by reverse osmosis (concentrate), and finished syrup are shown.
Obviously, the use of RO has tremendous consequences to the amount of evaporation, and thus heating time necessary to make syrup. Because much of the color and flavor development occur during the heating phase of processing, RO use should strongly affect the resultant color and flavor of maple syrup produced, although there has been little research clearly demonstrating the effects. Ongoing efforts (van den Berg and Perkins, unpublished data) do appear to indicate some significant changes in syrup compositional attributes. Depending upon the membrane and RO used, the chemistry of the concentrated sap may be somewhat different than what might be expected by a simple concentration factor alone. Some membranes sacrifice a slight amount of sugar and mineral passage for high flow rates, while others provide very high sugar retention, with correspondingly low flow rates. Although it has not been extensively investigated, limited research has determined that phenols are concentrated, but aldehydes and alcohols in sap are reduced during RO use (Kermasha et al., 1995). Prefilters and membranes may also serve as a source of microbial contamination of sap.
C. Transformations during evaporation by heating Although the single greatest influence of the process of transforming sap into syrup is the removal of water, the effect on the chemistry and flavor is not simply due to concentration. Rather, a number of complex reactions
118
Timothy D. Perkins and Abby K. van den Berg
are involved which result in the chemistry and flavor profile of maple syrup. Given the huge number of permutations in storage, processing conditions and rates, and temporal variations in sap chemistry, only generalizations are possible. The first and most obvious transformation that occurs during evaporation is the change in sugar content (Fig. 4.6). Sap enters the process at an average of around 2 Bx, and the finished product, maple syrup is around 66–67 Bx (Figs. 4.3 and 4.5). The process inside the evaporator is theoretically one in which a semicontinuous gradient of sugar is formed. Initially, as an evaporator is first started with sap only, the inflow of sap into the back pan will gradually cause a gradient to form. This gradient eventually perpetuates throughout the entire evaporator, with a density near that of syrup near the draw-off, and near that of sap at the inlet of the back pan. In reality, the evaporator is more or less a series of interconnected pans, each with their own sugar content, but certainly influenced by the sugar concentration in the pan immediately before and after it. The semicontinuous flow of sap in, and regular draw-off of syrup, maintains the sugar gradient within the entire evaporator system. Sometimes, after a short period of boiling, maple producers will ‘‘sweeten’’ the partition nearest the draw-off with syrup to hasten the development of a proper gradient, although this is not necessary for the gradient to form. As the density gradient along the evaporator develops, a concomitant increase in boiling temperature also is found (Isselhardt et al., 2007). Raw sap boils at 212 F, and syrup at 66.5 Bx boils at 219.3 F (at standard atmospheric pressure). Most producers use a hydrometer, calibrated in Brix or Baume and corrected for temperature, to determine the finished syrup density. The second most obvious transformation during heating is a change in color (Fig. 4.6). As sap progresses through the evaporator, it darkens due to nonenzymatic browning, a complex suite of chemical changes arising from nonenzymatic activity acting on sugar solutions. The longer the time and intensity of heating, the greater the effect on color and flavor development (Willits et al., 1952). These same processes are also intimately involved in flavor formation. The rate of nonenzymatic browning reactions varies greatly depending upon sap chemistry (particularly the invert sugar and amino acid concentration) and on processing rate and conditions. In general, syrups that are low in invert sugar concentration (typical early season sap) produce light-colored and light-flavored maple syrup, whereas those that have high invert levels produce darker colored and stronger tasting syrups due to increased substrate availability for nonenzymatic browning reactions (Naghski and Willits, 1957). Maple producers attempt to control the color and flavor profiles of the syrup they produce primarily though rapid processing of sap.
119
Maple Syrup—Production, Composition, Chemistry, and Sensory Characteristics
B 80
220
70 Sugar content (⬚Bx)
222
218 216 214 212
50 40 30 20 10
40 ⬚F
8.0
7.0 6.5
p
4
ru Sy
tp on
Fr
Fr
on
tp
an
an
3
2
1 tp Fr
on
tp on
Fr
an
2 n
an
1 ck Ba
1400 1200 1000 800 600 400 200
6.0
p
4
ru Sy
3
an tp on
Fr
Fr
on
tp
an
2
1 tp on Fr
Fr
on
tp
an
an
2 n
1 Ba
ck
pa
ta
pa ck
Sa
on Fr
n
nk
p
4
ru Sy
3 tp
an
2 Fr
on
tp
an
1
an
Fr
on
tp on
Fr
tp
an
2 n
1 n
pa ck Ba
Ba
ck
Sa
p
pa
ta
nk
0
Ba
pH
pa
ta 1600 Conductivity (µS·cm)
1800
9.0
7.5
Light transmittance (%)
ck
D 9.5
8.5
E
pa
p Sa
on Fr
C
n
nk
p
4
ru Sy
3 tp
an
2
an tp on
Fr
Fr
on
tp
an
2
an
n
tp
Fr
on
ck Ba
Ba
ck
pa
n
nk
pa
ta p Sa
1
0
1
208
Ba
210
60
p
Temperature (⬚F)
A
100 90 80 70 60 50 40 30 20 10
4
Sy ru p
3
an
Fr on
tp
2
an
Fr on
tp
1
an tp
Fr on
an
2 Fr on
tp
n
1 n
pa Ba
ck
pa ck
Ba
Sa p
ta
nk
0
FIGURE 4.6 Changes in temperature (A), sugar content (B), pH (C), conductivity (D) and light transmittance (E) at different stages of thermal processing from maple sap to syrup.
Light-colored syrups are generally more difficult to produce, and are very useful in blending of maple syrup to achieve a good color and flavor balance. Lighter syrups therefore generally command a higher price than darker syrups. During heating, sap pH initially increases rapidly as the solution becomes more concentrated, generally transitioning from neutral or slightly acidic to slightly to moderately alkaline (Fig. 4.6). Akochi et al. (1997) demonstrated that this change is most likely the result of chemical reactions occurring during the heating process rather than the loss of
120
Timothy D. Perkins and Abby K. van den Berg
organic acids. By the time sap reaches a concentration of 8–12 Bx, sap pH begins to decline and continues to steadily drop until reaching the point when it is the density of syrup. During this time, hexoses undergo alkaline degradation to form triose sugars, which readily decompose during heating to produce color bodies. Alkaline degradation and color formation do not occur exactly simultaneously, as the conversion of hexoses to trioses appears to occur early in the boiling process, but most of the color formation occurs in the latter stages of heating, primarily in the last few partitions of the evaporator. Cyclotene, furaneol, isomaltol, and other thermal sugar degradation products are probably formed at this stage of processing (Potter and Fagerson, 1992). A large number of lignin-derived flavors have been identified in maple syrup (Filipic et al., 1969; Potter and Fagerson, 1992). During boiling, there are large increases in phenol-related flavor compounds such as furaldehydes, vanillin, and syringyl aldehyde (Kermasha et al., 1995). Furfural and hydroxymethylfurfural color precursors form as a result of carmelization and Maillard reactions between amino acids and reducing sugars, as well as via oxidative polymerization of phenolic compounds. These are further reduced to caramels and melanoidins, and eventually to colored polymer bodies. Due to the variable concentration of precursors in sap, the range of color and of flavor compounds in maple syrup is very broad. Like the case for sap, there is a tendency for syrup to decrease in pH during the production season. Because one of the dominant color formation pathways involves the reactions among amino acids and reducing sugars, controlling invert sugar levels has historically been the key to producing light maple syrup. RO usage and highly efficient evaporators that result in rapid processing and low sap residence time can also affect color and flavor development. The recent innovation of injecting air through small pipes into boiling sap has proven to also produce light-colored syrup. Recent work has shown that significant improvement in syrup light transmission can result from air injection (Fig. 4.7, van den Berg et al., 2009a), with few other significant alterations to bulk syrup chemistry. The way in which air injection achieves this is still somewhat unclear. Air injection does result in an overall lowering of temperature of the boiling sap (Isselhardt et al., 2007; UVM Proctor Maple Research Center, unpublished data), suggesting the lightening effect may be due simply to reduced thermal-induced browning. Research at Centre Acer in Quebec, Canada, has shown that air injection may increase the formation of oxidative species which affect color and flavor formation during evaporation. These projects are expected to elucidate in more detail the effects of air injection and other new maple processing technologies on the chemistry and flavor profiles of maple syrup.
Maple Syrup—Production, Composition, Chemistry, and Sensory Characteristics
121
Control 3/15
3/25
3/29
3/31
4/3
4/5
4/7
Air injection
FIGURE 4.7 Maple syrup produced in paired evaporators boiling sap from the same source on the same dates. The evaporator that produced the syrup on the top was the standard, control evaporator. The syrup on the bottom was produced in an identical evaporator equipped with air injection.
VIII. SCALE/SUGAR SAND FORMATION DURING SAP PROCESSING As the sap is boiled in an evaporator, the concentration of other substances also changes. Dissolved minerals and metals go through a saturation phase, and eventually precipitate as a scale-like substance on the surfaces of the evaporator. This scale, also termed ‘‘niter’’ or ‘‘sugar sand,’’ can take many forms, and is quite variable in composition (Table 4.5, Fig. 4.8). The type and amount of scale changes throughout the season, and is highly variable from one season to the next. Nearest the sap inlet of the back pan it is composed of a protein-rich, sticky substrate that is rich in calcium malate and is probably largely caused by the denaturing of organic substances and microbes. In general, sugar sand is a mixture of calcium malate and sugar. Generally, the higher the levels of calcium and malic acid present in the sap, the greater the amount of scale formation (Davis et al., 1963). Further into the evaporator the scale deposited is denser and adheres strongly to pan surfaces. In the front pans, as the density of the solution becomes quite high, scale forms very rapidly on evaporator surfaces, and small particles of scale become suspended in the syrup. This suspended material, also often referred to as ‘‘niter’’ or ‘‘sugar sand,’’ must be filtered from syrup prior to packing in drums or into retail containers, as it will impart a gritty texture to syrup, and can cause an undesirable off-flavor in the syrup.
122
Timothy D. Perkins and Abby K. van den Berg
TABLE 4.5 Composition of maple sugar sand (from Perkins et al., 2006, used with permission)
FIGURE 4.8
Sugar sand (in run)
0.05–1.42% dw
pH Calcium Potassium Magnesium Manganese Phosphorus Iron Copper Boron Molybdenum Free Acid Total malic acid Acids other than malic Undetermined material Calcium malate Sugars in dried samples Sugar sand in dried samples
6.30–7.20 0.61–10.91% 0.146–0.380% 0.011–0.190% 0.06–0.29% 0.03–1.18% 38–1,250 ppm 7–143 ppm 3.4–23 ppm 0.17–2.46 ppm 0.07–0.37% 0.76–38.87% 0.08–2.62% 6.94–34.16% 1.30–49.41% 33.90–85.74% 14.26–66.09%
Variation in maple scale (sugar sand/nitre) appearance and form.
Maple Syrup—Production, Composition, Chemistry, and Sensory Characteristics
123
Scale on evaporator surfaces, especially in the front pans, is a nuisance to the maple producer. If allowed to build up excessively, it reduces heat transfer to the liquid, can cause off-flavors, and may result in scorching of the pans. Maple sugar makers deal with scale accumulation in their evaporators in several ways. The most apparent is to shut down the process, drain the partially processed sap, and clean the pans. Often a food-approved acid solution is used to hasten the process (great care must be taken in handling and disposal of acids and in ensuring that all acid residues are adequately rinsed from the pans prior to using them again). An alternative process is used with reverse-flow pans in which the flow of sap in the front pans is changed by switching the location of the draw off. By alternating sides on which syrup is removed, the incoming partially processed sap will redissolve a portion of the scale from the pans, and thus delay the necessity to shut down and clean the pans. Producers using cross-flow pans will often remove the pan nearest the draw-off, move the second front pan forward, and insert a spare clean pan in that position, thereby reducing the amount of time in which the evaporator is shut down. When boiling sap concentrate (8–20 Bx), producers obviously experience much more rapid build-up of scale in evaporators than when using sap (2 Bx), and the scale can also begin to form further back in the evaporator system. Given the increasing quantity of syrup being produced using reverse osmosis, this problem (and the use of acid to clean pans) is increasing rapidly. Research is ongoing in several locations to better characterize the scale, and to find easier and more environmentally safe methods to deal with the problem. One recent interesting approach is the use of electrodialysis to demineralize sap prior to evaporation by heating (Bazinet et al., 2007). Although the process did reduce calcium and malic acid levels in concentrated sap, thus presumably reducing scale formation, it may be considered illegal under most existing maple purity laws. A similar issue in evaporator pans is foam. Foam develops in all portions of evaporator pans during boiling, and must be controlled to maximize heating and to prevent scorching of pans. Historically, milk, cream, or animal fats were used to cut the foam. Present practice is to use vegetable oils or other commercially available defoamers. The primary determinant of which defoamer to use is often based upon whether or not the maple producer is ‘‘organic’’ certified, as certification requires the use of certified organic vegetable oils. Although there has been relatively little scientific study of the composition of foam, an analysis of foam skimmings has shown foam to be a sink for lead (Stilwell and Musante, 1996), and occasionally foam has been suggested as a source of off-flavors in maple syrup.
124
Timothy D. Perkins and Abby K. van den Berg
IX. SYRUP STANDARDS Pure maple syrup generally must meet strict standards for density, clarity, color, and flavor. In general, there is agreement between the various grades of maple syrup produced in the U.S. and Canada; however, the specific names may vary somewhat (Table 4.6). Some U.S. states have their own grade and color designations. A complete description of maple syrup grade and color descriptors is available in the North American Maple Syrup Producers Manual (Heiligmann et al., 2006, p. 172). The Canadian standards are mandatory, whereas U.S.D.A. standards are voluntary. The minimum solids content to meet U.S.D.A and Canadian regulations is 66% total solids (at 20 C in Canada, unspecified in the U.S. Standards). Individual states in the U.S. are able to set their own standards. Vermont and New Hampshire set a minimum of 66.9 Bx (at 60 F). Some jurisdictions set an upper density limit (typically 68.9% solids), whereas others do not. Density is generally measured with a hydrometer, hydrotherm, or, increasingly with refractometers, although regulations often specify the ‘‘legal’’ method in each area. In all cases, syrup must be clear of suspended crystals or particulates that might cause cloudiness and impart a gritty texture to the product. Syrup color is determined by spectrophotometric light transmittance at 560 nm (Fig. 4.9). Syrup must meet (or exceed) a certain cut-off level in TABLE 4.6 Grades of maple syrup in Canada and the U.S.
Light transmittancea
75.0% 60.5–74.9% 44.0–60.4% 27.0–43.9%
<27.0% a
Canada
U.S.
66.0% Solids
66.0% Solids
Grade
Color
Grade
Color
Canada No. 1 Canada No. 1 Canada No. 1 Canada No. 2
Extra light (AA) Light (A)
U.S. Grade A U.S. Grade A
Medium (B)
U.S. Grade A
Light amber Medium amber Dark amber
Amber (B)
Canada No. 3
Dark (C)
U.S. Grade B for reprocessing Substandard
U.S. Standards are based upon the transmittance of permanent glass standards with photometric equivalents. Canadian Standards are based upon photometric light transmission at 560 nm. Some states may have different names for the grade classes or slightly different density requirements.
Maple Syrup—Production, Composition, Chemistry, and Sensory Characteristics
125
order to be placed within a grade. In practice, producers generally use a visual comparator based upon ‘‘permanent’’ glass or plastic filters, or more commonly, with a temporary standard composed of burnt sugar suspended in glycerol. Such temporary grading kits are useful as guides, but care should be taken in their use. At least one company has recently introduced a digital maple syrup transmittance meter, however, for several reasons such devices are not currently in widespread use within the industry. Syrup is also graded based upon its flavor. Although this is somewhat subjective, experienced producers are able to distinguish the general category and intensity of maple flavors, as well as detect gross off-flavors that cause syrup to be down-graded to a commercial or substandard level. Because of maple purity laws in both the U.S. and Canada, there are relatively few processing techniques which can be employed by maple producers. In general, processes may neither add, nor remove substances from the product, other than what is accomplished through evaporation. Certain exceptions are allowed, such as the use of diatomaceous earth for filtering, the use of reverse osmosis machines, and various defoaming agents. Addition of other sugars, decolorizing agents (activated carbon, anion exchange decolorizing resins, bone char, flavor modifying agents, pH modifiers), preservatives, flavorings, colorants, and other food additives are not allowed.
100% r
be
90%
m
a ht
Light transmittance
80% de ra
70%
A
lig
G
M
60%
ed
iu
m
Da
50%
am
rk
be
am G
r
be
d ra
r
e
l
cia
B
er
m
m
co
/ rd
da
40%
an
t bs
Su
30% 20% 10% 0% 190
290
390
490
690 590 Wavelength (nm)
790
890
990
1090
FIGURE 4.9 Transmission profiles of maple syrup. Profile for each grade was derived from 10–40 scans of individual syrups. Grades delineations are based upon U.S.D.A. standards. Color grade is determined by the transmittance of light at 560 nm.
126
Timothy D. Perkins and Abby K. van den Berg
X. SYRUP CHEMISTRY A. Density Maple syrup is, by definition, 66–68% total dissolved solids (at 20 C). Since the major solid constituent (98% or more) of syrup is sugar, % total solids and Bx are generally used interchangeably. Syrup that is less dense than 66 Bx is subject to fermentation if contaminated by microbes. Syrup denser than 67 Bx may develop crystals inside the container. Therefore, a fairly narrow range of acceptable density is required to ensure that quality issues are not found. Water makes up the bulk of the remaining solution (32–34%), but due to the high concentration of dissolved solids, the water activity of maple syrup is fairly low, averaging around 0.87–0.88, thereby restricting the development of pathogenic microorganisms.
B. Carbohydrates Most of the total carbohydrate fraction of syrup (88–99 þ %) is in the form of sucrose, with varying amounts of invert sugars present, primarily as a function of microbial contamination of sap during storage prior to processing (Edson, 1910; Edson and Jones, 1912). In some cases, conversion of sucrose to invert sugars may continue after packing into retail containers, especially if syrup is low in density and fermentation occurs. A slight amount of sucrose hydrolysis may also occur during thermal processing. Total sucrose composition in syrup may range from 47–74%, typically with a mean of 65–68%. Glucose and fructose may range from nondetectible up to over 9% (Stuckel and Low, 1996), but typically are both found in the 0.4–0.5% range in finished syrup. A recent study by van den Berg et al. (2006) revealed no consistent relationships among maple syrup grades and carbohydrate profiles (sucrose, glucose, fructose). However, lighter-colored syrups are generally thought to contain lower amounts of invert sugars than darker syrups, and are thus selected for use in making maple candy and cream, where crystallization properties are important.
C. pH The published range of pH of maple syrup varies from somewhat acidic (4.7) to slightly basic (8.7), but typically is near neutral, with reported mean values around pH 6.3–6.8 (Che´nard and Gagnon, 1997; Perkins et al., 2006). The pH of the syrup is related to several factors, including collection and storage of sap, the methods employed to process the sap into syrup, time of season, and microbial contamination.
Maple Syrup—Production, Composition, Chemistry, and Sensory Characteristics
127
D. Conductivity Due to dissolved minerals in maple syrup, the conductivity can range from about 90 to slightly above 230 mS/cm, with a trend towards increasing conductivity as light transmittance of syrup decreases. Conductivity of syrup is sometimes used as a rapid screening test for adulteration of syrup, especially in Canada.
E. Color Maple syrup ranges greatly in color and opacity. The light transmittance from the UV through the near IR regions is shown in Fig. 4.9. UV light is not well transmitted through maple syrup. Transmittance increases rapidly in the visible spectrum, reaching a gradually upward sloping plateau at around 650 nm before reaching a peak in the near IR range at around 920 nm. The color of maple syrup ranges from a very light yellow-amber to near black. Generally, most syrups have a light yellow to a dark red-brown color, due mostly to color bodies originating from nonenzymatic browning during evaporation. Because invert sugar browns at a lower temperature than sucrose, the level of glucose and fructose in sap, as well as rate of processing are the primary determinants of syrup color and light transmission. Maple syrup produced by vacuum or microwave processing is nearly colorless as a result of there being no or low thermal degradation of the sugars (Favreau et al., 2001; Willits et al., 1952). Chen et al. (2001) analyzed the colorimetric properties of maple syrup and found that total color value and ‘‘redness’’ increased strongly with increasing color grade, but that color was not an adequate tool to discern compositional differences between authentic maple syrup and table syrups composed of other sugars. Most likely, this is because table syrups utilize burnt caramel as a color agent in the formulation of these products.
F. Rheology The rheological properties of Canadian maple syrup have been described by Ngadi and Yu (2004). Although maple syrup is primarily Newtonian in its flow characteristics, grade, temperature, and density can independently and interactively influence the apparent viscosity, with a range of 0.138–0.160 Pa s at 25 C. Colder, more dense, and darker grade syrups have higher viscosity than warmer, thinner, and lighter grades. Chen et al. (2001) found a slightly higher range of 0.128–0.247 Pa s at 25 C, with a mean of 0.164 Pa s at 25 C, probably as a result of a higher variation in syrup density and geographic origin. They did not find a relationship between syrup grade and viscosity. Corn syrup
128
Timothy D. Perkins and Abby K. van den Berg
containing cellulose gum as a thickening agent had a vastly different rheological signature, making structural analysis a possible method to detect some forms of maple syrup adulteration.
G. Inorganic composition A number of studies have described the range of minerals and metals found in maple syrup (Dumont, 1996; Morselli, 1975; Stuckel and Low, 1996). These studies and other unpublished works are summarized in Che´nard and Gagnon (1997), and by Perkins et al. (2006) (Table 4.7). While many of these constituents originate in the composition of the sap as it comes directly from the tree (which can vary widely), other elements may be introduced during the collection and evaporation process. Potassium is by far the mineral found in highest concentration in maple syrup, with a range from various studies of 541–4031 ppm. Typically, the average values are around 2000–3300 ppm. Calcium is also found in substantial quantities, with a mean of 630–1470 ppm. The mean magnesium concentration is 130–375 ppm. Lesser amounts of manganese, sodium, and phosphorus are usually found. The content of various metals in maple syrup fluctuates widely, often as a result of contact with sap collection and processing equipment. Iron, zinc, copper, and tin are generally present to varying degrees. Tin often originates from uncoated metal containers sometimes used to market maple syrup. Upon opening a metal container, oxidation can cause tin levels to rise to the point where syrup begins to appear slightly greenish, and acquire a distinct ‘‘metallic’’ off-flavor over several months time. Little research has focused on the differences in syrup composition by grade. While light transmittance is different (by grade definition), the trends in other aspects of syrup chemistry are not as simple. Two unpublished studies at the University of Vermont Proctor Maple Research Center show that there are no consistently predictable trends in bulk syrup chemistry related to grade (color class). Maple sap collection and processing equipment produced prior to 1994 was frequently made with lead solder, terneplate, or galvanizing material containing lead. Lead is not detectable in sap collected directly from the tree, but fairly rapidly dissolves in sap. The composition of the sap collection and processing materials, surface area, and length of contact time are important factors in determining the degree of lead contamination of sap. Upon boiling, lead is concentrated; however, a great deal of lead is sequestered in scale and sugar sand, and is thus filtered out. The dissolved lead fraction cannot be filtered, and thus may contaminate the syrup. The U.S., Canada, and some states have set maximum permissible lead levels ranging from 250 to 500 ppb, and the vast majority of syrup produced falls well under those limits. As equipment
TABLE 4.7
Reported values and ranges for composition of maple syrup (adapted from Perkins et al., 2006, used with permission)
Morsellia
Stuckel and Lowb
Dumontc
Perkinsd
van den Berg, Perkins and Isselhardte
Typical mean
66.5 33.5 – –
63.2–69.5 26.5–39.4 5.6–7.9 –
– – 5.7–8.5 –
– – – –
62.0–68.0 – 5.5–7.3 96–318
65 – 6.4 189
Carbohydrates Sucrose (%)
58.5–65.8
51.7–75.6
–
59.4–73.8
66
Glucose (%) Fructose (%)
0–7.3 Trace
0.0–9.6 0.0–4.0
42.3– 74.0 0.0–7.2 0.0–6.8
– –
0.0–1.6 0.0–1.1
0.4 0.5
– 1055–2990
– 1600– 2590 600–1250
0.03 2283
266–1707
278–2494
911
Magnesium (ppm) Manganese (ppm) Sodium (ppm)
12–360 2–220 0–6
10–380 – –
– 541– 4031 183– 1943 11–575 <1–252 <1–261
0.001–0.077 963–3319
Calcium (ppm)
– 1300– 3900 400–2800
0–198 0–117 0–27
25–543 0.01–223 0.01–492
177 39.8 36
General Total Solids (%) Moisture (%) pH Conductivity mS/cm2
Minerals and Metals Nitrogen (%) Potassium (ppm)
(continued)
TABLE 4.7 (continued)
a b c d e
van den Berg, Perkins and Isselhardte
Typical mean
20–113 0–18 0–96 – – – 0–6 0–24 0–0.35 0–0.07
0.01–91 0.01–61 0.01–527 0.01–18 0.01–3 0.01–100 – – – – 20–400
37 10 22 2 0.2 18.2 1.2 8 0.20 0.02 93
–
–
– – –
– – –
Morsellia
Stuckel and Lowb
Dumontc
Perkinsd
Phosphorus (ppm) Iron (ppm) Zinc (ppm) Aluminum (ppm) Boron (ppm) Sulfur (ppm) Copper (ppm) Tin (ppm) Lead (ppm) Cadmium (ppm) Chloride (ppm)
79–183 0–36 0–90 – – – 0–2 0–33 0–0.25 – 31–191
– – – – – – – – – –
<2–2335 0–18 2–43 – – – 0–8 – 0–0.49 0–0.09
Organic Acids Malic Acid (%)
0.141
0.06–0.66
Fumaric Acid (%) Citric Acid (%) Succinic Acid (%)
0.006 0.015 0.012
0.001–0.012 – –
0.32– 0.90 0.0–0.13 – 0.0–0.26
Morselli (1975) Stuckel and Low (1996) Dumont (1996) Perkins (1998, unpublished data) van den Berg, Perkins and Isselhardt (unpublished data).
Maple Syrup—Production, Composition, Chemistry, and Sensory Characteristics
131
containing lead reaches the end of its service lifetime, the overall level of lead in maple syrup should decrease. Only a few studies have examined the minor inorganic constituents of maple syrup. Both sodium and chloride are sometimes found in maple syrup, occasionally at fairly high levels. Morselli and Whalen (1987) found a mean chloride concentration of 93 ppm in syrup, with a range of 31–191 ppm. Trees growing in areas of heavy road-salt use had higher concentrations of salt in sap, and eventually in syrup. Sodium ranges from 0.1 to 492 ppm, with a typical mean of 2–36 ppm. Sodium can be elevated due to roadside salt, or more commonly, to the use of sanitizers (sodium hypochlorite) in maple tubing. More recently, the ratio of sodium to chloride has been investigated as an indicator of artificial decolorization of syrup (an illegal activity) with anion exchange resins (Perkins and van den Berg, unpublished data).
H. Organic acids Several organic acids are typically present in maple syrup (Table 4.7). Malic acid is the predominant form, comprising on average about 0.5% of the total weight of syrup. Various studies have yielded a range of 0.06–0.90%, but typically average about 0.5 ppm. Citric acid, succinic acid, and fumaric acid are also present in trace amounts in syrup (Stuckel and Low, 1996), but typically are found in far lower concentrations than malic acid. Malic acid is a dominant constituent of sugar sand, and malic and citric acids are sometimes used to detect different forms of adulteration, as the signatures of these compounds are quite different in some other types of sugars.
I. Flavor compounds Over 130 volatile flavor compounds have been identified in maple syrup, although in widely varying levels and combinations depending upon the grade of syrup as influenced by the type and intensity of processing, the time of season (probably as a result of microbial influence on sap chemistry), and possibly by geographic location. Many of the flavor compounds are phenol derivatives. Several published studies describe some of these (Akochi et al., 1994, 1997; Alli et al., 1992; Belford et al., 1991; Deslauriers, 2000; Filipic et al., 1969; Kermasha et al., 1995; Potter et al., 1995; Underwood, 1971). The dominant types of flavor compounds include: phenolics, pyrazines, carbonyl-based molecules, and others. Analysis of the primary flavor contributors to maple syrup is complicated by the large number, combinations, and levels of flavor compounds present, as well as the detection threshold and saturation nature of the taste reaction of these compounds.
132
Timothy D. Perkins and Abby K. van den Berg
In general, lighter-color (higher grade) syrups tend to contain lower levels of flavor compounds than darker (lower grade) syrups. Some extremely light syrups have very little flavor (other than sweet). In the maple industry, these are sometimes termed ‘‘techno-syrup,’’ as they are considered to be more prevalent in operations using a higher level of processing technology (reverse osmosis, steam- or vapor compression evaporators). The veracity of this connection has not been established, however, it is enough of a concern that researchers are beginning to investigate the effects of processing equipment and intensity on maple syrup chemistry and flavor. Phenolic flavor compounds present in syrup are largely the degradation products of lignin components of sap. These compounds include: vanillin, syringaldehyde, dehydroconiferyl alcohol, syringoyl methyl ketone, and 2,6-dimethoxyphenol. Vanillin is generally a strong contributor to the flavor of light-colored, early-season maple syrup (Potter et al., 1995), whereas a variety of furanones, cyclotene, and other caramelization-derived compounds appear to dominate in darker syrups (Belford et al., 1991; Potter et al., 1995). Pyrazines are generally products of Maillard reactions, and tend to increase fairly steadily during boiling (Akochi et al., 1997). The presence and concentration of several forms of pyrazine in maple syrup are quite variable, and they contribute both to the flavor and off-flavor attributes of maple syrup depending upon the form and concentration of both the pyrazine and other flavor compounds present which may mask any off-flavor. In particular, moderately high levels of 2,5-dimethylpyrazine are thought to result in the objectionable ‘‘metabolism’’ off-flavor found occasionally in maple syrup (van den Berg et al., 2009b). In some years, ‘‘metabolism’’ (also called ‘‘woody’’) can affect up to 25% of the total annual maple syrup crop. At lower concentrations, especially when combined with strong carmelization flavors, 2,5-dimethylpyrazine likely contributes acceptable flavors to maple syrup. Research is ongoing to investigate processes to reduce or mitigate ‘‘metabolism’’ off-flavor in maple syrup. Reverse osmosis of sap increases the relative proportion of phenolic compounds, while decreasing aldehydes and alcohols (Kermasha et al., 1995). Most maple producers concentrate sap to around 8–10%, reducing the amount of water to be evaporated by 75%. Maple syrup produced from concentrated sap tends to be lighter in flavor and color, due to the far reduced boiling time required to achieve the proper density. Given the increasing cost of fuel, the tendency among some producers in recent years has been to increase the level of concentration. Sap concentration by RO up to 15–25% is not unheard of at the present time. Many producers and packers in the maple industry feel that high concentration (> 10%) of sap by RO produces a ‘‘flat’’ syrup, lacking in flavor. This topic is also the focus of ongoing research.
Maple Syrup—Production, Composition, Chemistry, and Sensory Characteristics
133
Overall, dozens of compounds are involved in the development of the characteristic flavors found in maple syrup. The number of compounds, the different sap to syrup processing involved, the wide range of maple syrup flavor and intensity, and somewhat frequent off-flavors all complicate the understanding of the flavor of maple syrup.
J. Sensory evaluation of flavor Although the dominant taste is usually sweet, maple syrup can range from a very light, sweet, but otherwise nearly tasteless substance, to a very dark, bitter, burnt taste. Most syrups used for table use on pancakes or in cooking are intermediate in color and have a moderate to strong maple-caramel flavor, without objectionable off-flavors. Light colored, low-flavor syrups are generally used for blending with darker syrups to achieve the correct color and flavor profile. Moderate syrups can be used without blending. Dark syrups usually have strong flavors (and sometimes slight-moderate off-flavors that are masked by strong ‘‘good’’ flavors) and are used for blending, or as ingredients in recipes requiring a maple flavor. Very dark, commercial syrups (which are not sold for table use) range from very strong tasting syrups with good flavor, to very strongly bitter syrups with mild to moderate off-flavors. These syrups are used in commercial food manufacturing processes where a strong maple flavor is required. The degree of off-flavor deemed acceptable is determined by the user and the application. For most producers and consumers, the description of syrup flavor is generally fairly straightforward, focusing on a simple scale of light flavor to strong (dark) flavor. The lightest syrups (Canada No. 1 Extra Light and U.S.D.A. Grade A Light Amber) are generally sweet, with only a slight hint of maple flavor. As one tastes progressively darker syrups, the taste becomes dominated less by sweet alone, and more by a maple-caramel flavor. The strongest table grades (Canada No. 2 Amber and Vermont Grade B) are dominated by an intense caramel flavor (although not enough to cause a bite or bitter taste), generally with a mix of several other flavors in lesser amounts. Often the darker color and strong flavors cause people to believe that these syrups are thicker, although all syrups must legally meet the same density requirements. A relatively recent addition to the maple industry is the ‘‘Flavor Wheel for Maple Products’’ (Agriculture and AgriFood Canada, 2004), patterned after similar classification systems in wines and other food products. The Flavor Wheel for Maple Products (hereafter FWMP) contains 13 flavor families, including: maple, confectionary, vanilla, milky, empyreumatic, floral, fruity, spicy, foreign (deterioration/fermentation), foreign (environment), plant—herbaceous, plant—humus/forest/cereals, and plant— ligneous. Within each family are one or more subfamilies of flavors,
134
Timothy D. Perkins and Abby K. van den Berg
which on the outer ring are further divided into individual flavors. The FWMP is a useful tool in the maple industry, and although it is fairly new, and requires substantial training to be used correctly, is likely to become more widespread in maple research and large-scale marketing as it provides a common lexicon for those in the industry to describe the sensory characteristics of maple in a very detailed way. Typically, there is a progression from lighter-colored and flavored syrups (Canadian No. 1 AA or U.S.D.A. Grade A Light Amber) early in the maple production season, dominated by delicate maple and vanilla flavors, to moderately colored and tasting syrups (Canadian No. 1 A and U.S.D.A. Grade A Medium Amber) in the middle of the season, with light-moderate confectionary (light brown sugar) and empyreumatic (light caramel) flavors, to dark-colored and strong flavored syrups (Canadian No. 2 B or U.S.D.A. Grade B), characterized by heavy confectionary (molasses) and empyreumatic (burnt-sugar) notes. There has been some recent research investigating the ‘‘Terroir’’ of maple. Preliminary work has shown that the underlying bedrock composition may influence the chemistry of the sap and resulting syrup (Corbett and Munroe, 2006), but the effects on maple syrup flavor are not as clear (Costanza-Robinson et al., 2007).
K. Off-flavors in maple syrup Within the maple industry, more attention is sometimes given to recognizing off-flavors, as these can result in syrups being down-graded, or placed into the commercial or substandard class. The origin of off-flavors may be either intrinsic to the sap, or arise from external factors. Offflavors regularly found in maple syrup include: metabolism (a poorly understood woody-cardboard flavor arising from environmental conditions), buddy (made from late season sap, characterized by a strong, bitter chocolate flavor), salty (inadequate rinsing of equipment), chemical (inadequate cleaning/rinsing procedures), scorch/burnt (overheating damage), moldy or fermentation (microbial contamination), and several others (Heiligmann et al., 2006). Producers, packers, and agricultural inspectors generally learn to recognize these and are able to diagnose production and/or storage issues causing the off-flavor. Although typically color and flavor development progress hand-in-hand during evaporation, those in the maple industry must also assess whether the flavor of the syrup is typical of the color. A syrup found to have a flavor not characteristic of the grade is usually down-graded (placed into the next lowest syrup grade). Although primarily designed as a tool to describe positive sensory attributes of maple flavor, the FWMP does include some off-flavors as well. Some off-flavors are dealt with by blending, an attempt to reduce the off-flavor to undetectable or unobjectionable levels.
Maple Syrup—Production, Composition, Chemistry, and Sensory Characteristics
135
Other off-flavors (metabolism, buddy, chemical) are often too severe to allow the syrup to be used in this manner, and are thus relegated to commercial status and used for industrial flavoring, or to make sugar (during which the process reduces or eliminates most of the volatile offflavor compounds). Due to improper packing or storage conditions, more frequently found in syrups packed at too low a density, microbial contamination can sometimes cause off-flavors. Because of the low water activity of maple syrup, few organisms are generally able to successfully colonize maple syrup. When contamination does occur, it will result in a moldy offflavor. Bulk storage of syrup low in density will result in a fermentation off-flavor (fruity). Generally these barrels are easy to spot, as the top and bottom of the drum will bulge outward due to the buildup of carbon dioxide. Both moldy and fermented syrup, unless severely damaged, can be reheated, filtered, and blended with a good-flavored syrup to achieve an acceptable result. Bulk storage of syrup in barrels in good condition, especially stainless steel barrels, will maintain syrup color and flavor for a long period of time (1–2 years). Syrup is packed into retail containers at a temperature of 180–185 F. Exceeding this temperature may promote the development of nitre, whereas packing too cool does not result in complete kill of any microorganisms that may be in the container. After packing, containers are usually turned on their sides to sterilize the container top as well. If containers are not cooled rapidly (especially when large orders are packed tightly together before cooling), they are subject to ‘‘stack burn,’’ wherein the syrup darkens rapidly. Maple syrup is commonly packaged in tin, glass, and plastic containers. Glass retains syrup color and flavor well for 6þ months, especially if kept cold. Syrup packed into plastic may begin to darken rapidly, although flavor is generally unaffected. For this reason, maple producers and packers will often pack a syrup that is somewhat lighter than the grade to account for darkening on the retail shelf. Alternatively, packing is done frequently with only a small amount of syrup in plastic kept in storage or on retail shelves. Tin, although less common presently, is particularly good for short-term maple syrup storage. Tin retains syrup color very well. Prolonged storage of syrup in tin, especially after the container has been opened, may result in a ‘‘tin’’ or ‘‘metallic’’ off-flavor. With long-term storage in tin containers a slightly green off-color may develop.
L. Nutritional aspects of maple syrup Although predominantly sugar, maple syrup does contain some amount of several minerals and vitamins (Tables 4.7–4.8). A one-fourth cup serving contains 217 calories, but supplies 100% of the Canadian
136
Timothy D. Perkins and Abby K. van den Berg
TABLE 4.8 Vitamins in maple syrup (adapted from Perkins et al., 2006, used with permission)
Thiamin (ppm) Niacin (ppm) Riboflavin (ppm) Folic Acid Biotin Pyridoxine (B6) A a b
Morsellia
CANDIb
– 0.16 0.046 Trace Trace Trace Trace
1.3 1 0.6 – – – –
Morselli (1975). CANDI (1997).
and U.S. F.D.A. daily recommended value of manganese, 34% riboflavin (vitamin B2), 11% zinc, and lesser amounts of magnesium, calcium, and potassium. Frequently it is said that darker syrups contain higher levels of minerals and vitamins, and while some limited studies have demonstrated this to some degree, the rather high variability in syrup chemistry often overwhelms this trend. Of more interest, recent work has shown that the phenolic compounds in maple sap and syrup confer some degree of antioxidant, antiradical, and antimutagenic properties to maple products (The´riault et al., 2006). Consumer preference of maple syrup appears to be trending towards darker (stronger tasting) grades.
XI. OTHER MAPLE PRODUCTS A variety of other ‘‘value-added’’ maple products are made from maple syrup. These include maple cream (also known as maple butter), which despite the name, does not contain any dairy product. It is made by heating syrup to 12–13 C above the temperature of boiling water, then cooling it rapidly (usually in a water bath), and then stirring it slowly, either mechanically or by hand, to encourage the formation of fine crystals. To prevent separation, the enzyme invertase is used in some areas (it is not legal in every U.S. state) to encourage the conversion of a portion of the sucrose to glucose and fructose. Similarly, a small amount of potassium sorbate is used in some areas to inhibit mold formation to lengthen shelf-life of the product. Maple candy, maple block or cake sugar, loose maple sugar (Indian sugar), and other types of maple products are made by boiling syrup to different densities followed by pouring into molds, pans, or grinding to a
Maple Syrup—Production, Composition, Chemistry, and Sensory Characteristics
137
fine texture. Specific procedures to follow can be found in Heiligmann et al. (2006). A new product consisting of maple syrup in which a portion of the sucrose has been converted to invert sugar (using invertase) was introduced recently. The advantages are that higher density syrup can be sold with a greatly reduced tendency to crystallize. In some areas this product is not allowed to be sold as ‘‘maple syrup,’’ due to the addition of invertase and because some areas have upper limits on the density of maple syrup (which this product exceeds). Due to the high level of invert sugar in this syrup, this product has a somewhat different flavor profile as well, with a more pronounced caramel-like flavor. Maple syrup and maple sugar are often used as flavoring agents in commercial and retail cooking. When added as an ingredient, typically a dark and strong tasting syrup will be used. Maple can add a distinctive flavor to foods, and is also often used as a humectant in some recipes.
XII. CONTAMINATION Due to the large concentration factor involved in converting sap to syrup (typically 40:1), any slight contamination of sap can result in off-flavors in syrup. All sap collection and processing materials must be kept very clean and free from cleaning or other residues, as these may impact the flavor of the resultant syrup. Soap is used very sparingly, if at all, in maple sugaring operations, as most detergents leave a residue that is almost impossible to entirely remove. Most cleaning is done with copious amounts of hot water. Chlorine or other special-purpose agents are used for sanitizing tubing and processing equipment, with repeated rinsing necessary to remove all residues. Given the increasing use of RO machines and the use of strong acids to clean evaporators, strict attention to chemical safety, particularly in using proper procedures to eliminate the risk of inadvertent contamination, is very important.
XIII. ADULTERATION Due to the wide differences in the price of maple and sugars from other origins, there is a considerable incentive for adulteration. The simplest is the addition of cane, corn, or beet sugar to maple syrup. A great deal of research has been conducted on this problem. As a general screening method for addition of nonmaple sugars, the conductivity of maple syrup can be checked. Syrups with conductivity readings above 1200 mS/cm are suspect, and thus subjected to additional testing.
138
Timothy D. Perkins and Abby K. van den Berg
Determination of the carbohydrate profile in suspect syrups may also be used. One historical method was to examine the specific rotation induced in linearly polarized light by the syrup using a polarimeter. Sucrose has an optical rotation of þ66.5 , glucose þ52.7 , and fructose 92 . Addition of sugars with a proportion of sucrose to invert sugars different from maple is readily detectible. Currently, HPLC can be used to readily determine the sucrose/invert ratios. These methods are only able to detect gross adulteration of maple syrup, due to the large natural variation in invert sugars found in maple syrup (0–12%). More recently, the carbon stable isotope ratio test (SIRA) has become an easy method to detect adulteration with cane and corn syrup (Carro et al., 1980). Because maple trees are C3 plants with a somewhat different photosynthetic pathway for carbon fixation, the ratio of 13C/12C in the sugar produced is different than cane or corn. Maple has a d13C of approximately 24.5, whereas corn and cane are closer to a d13C of 8 to 12. Thus, even a small addition of cane or corn syrup is readily detectable. Because beets are also C3 plants, the SIRA test is not able to detect adulteration with beet sugar. Improvement of the SIRA method is possible using malic acid as an internal standard (Tremblay and Paquin, 2007). Due to the inability to reliably detect beet sugar additions, the sitespecific natural isotope fractionation nuclear magnetic resonance (SNIF– NMR) method used widely in the wine industry was adapted for the maple industry (Martin et al., 1996). This method determines the site specific isotope concentrations of organic compounds by nuclear magnetic resonance of ethanol fermented from the suspect sample. Other methods to detect adulteration include anion-exchange liquid chromatography in combination with pulsed amperometric detection (Stuckel and Low, 1995) and FTIR, FT-Ramen, and NIR Spectroscopy (Paradkar et al., 2002, 2003). At times, a large price differential between light and dark grades of maple syrup has given rise to another type of adulteration in the form of bleaching of syrup. Such a practice is illegal throughout the U.S. and Canada. A variety of methods can be employed (Fig. 4.10). Hydrogen peroxide serves as a direct bleaching agent which acts by oxidation of color bodies in syrup. Use of hydrogen peroxide also causes serious flavor defects, the effect wears off somewhat over time, and residual hydrogen peroxide is readily detectible using simple test strips. Acid, which is added to the backpan during evaporation, can also result in syrup lightening, probably by lowering the pH enough to reduce the rate of alkaline degradation, and thus color body precursor formation. Carbon filtration is another method of syrup lightening. Formerly somewhat difficult to do, as it required charcoal or bone char, lightening via carbon filtration is somewhat easier now given the prevalence of a wide selection of resins that are designed specifically for sugar decolorization. Although adulteration via decolorization with
Maple Syrup—Production, Composition, Chemistry, and Sensory Characteristics
139
FIGURE 4.10 Effect of treatment of syrup with decolorizing resin (Top: from left to right, control syrup, decolorized, further decolorized) and hydrogen peroxide (Bottom: control syrup, treated syrup).
resins can be difficult to detect, there have been some methods developed to screen suspect syrup. One promising method is to examine syrup for increased level of anions substituted for the highly positively charged syrup color bodies (Perkins and van den Berg, unpublished data).
140
Timothy D. Perkins and Abby K. van den Berg
XIV. SUMMARY Although pure maple sap is predominately a dilute solution of sucrose and water, and maple syrup is mostly concentrated sap, the mixture of minor amounts of minerals, amino acids, phenolic compounds, and organic acids found in sap in varying amounts during the short sap exudation season, along with substances added, formed, or sequestered during the collection and processing phases of maple syrup manufacturing, and the interactions of sap and syrup with microorganisms at many stages of the process, make the chemistry of maple syrup rather complicated. In all cases, regardless of how sap is collected or processed, the end result is usually a pure product—maple syrup—with a unique and distinctive flavor.
REFERENCES Agriculture and Agri-Food Canada/Centre Acer (2004). Flavor Wheel for Maple Products. Food Research and Development Center. http://www.agr.gc.ca/maple_wheel. Akochi, K. E., Alli, I., Kermasha, S., Yaylayan, V., and Dumont, J. (1994). Quantitation of alkylpyrazines in maple syrup, maple flavors and non-maple syrups. Food Res. Int. 27, 451–457. Akochi, K. E., Alli, I., and Kermasha, S. (1997). Characterization of the pyrazines formed during the processing of maple syrup. J. Agric. Food Chem. 45, 3368–3373. Alli, I., Akochi, K. E., and Kermasha, S. (1992). ‘‘Flavor Compounds in Maple Syrup. Developments in Food Science’’ Vol. 29, pp. 131–140. Elsevier, Amsterdam. Bazinet, L., Gaudreau, H., Lavigne, D., and Martin, N. (2007). Partial demineralization of maple sap by electrodialysis: Impact on syrup sensory and physicochemical characteristics. J. Sci. Food Agric. 87, 1691–1698. Belford, A. L., Lindsay, R. C., and Ridley, S. C. (1991). Contributions of selected flavor compounds to the sensory properties of maple syrup. J. Sens. Stud. 6, 101–118. Belford, A. L., Lindsay, R. C., and Ridley, S. C. (2006). Bound vanillin in maple sap. Flavour Fragrance J. 7, 9–13. Canadian Nutritional Data Information (CANDI). (1997). ‘‘Nutritional Labeling Evaluation.’’ Health Canada, Ottawa, Ontario, Canada. Carro, O., Hilliaire-Marcel, C., and Gagnon, M. (1980). Detection of adulterated maple products by stable carbon isotope ratio. JAOAC 63, 840–844. Chen, H., Li, J., and Perkins, T. (2001). Rheological and colorimetry properties of maple syrup. American Society of Agricultural Engineers, Annual International Meeting, Sacramento, California, July-August 2001. Che´nard, C. and Gagnon, R. (1997). ‘‘Composition du sirop d’e´rable selon une revue de litte´rature.’’ Cintech AA Inc., St. Hyacinthe, Quebec, Canada. Cirelli, D., Jagels, R., and Tyree, M. T. (2008). Towards an improved model of maple sap exudation: The location and role of osmotic barriers in sugar maple, butternut and white birch. Tree Physiol. 28, 1145–1155. Corbett, L. B. and Munroe, J. S. (2006). Vermont terroir: Investigating the relationship between maple syrup chemistry and bedrock lithography (abstract). Geol. Soc. Am. Annual Meeting, Philadelphia, PA, October 22–25, 2006.
Maple Syrup—Production, Composition, Chemistry, and Sensory Characteristics
141
Cortes, P. M. and Sinclair, T. R. (1965). The role of osmotic potential in spring sap flow of mature sugar maple trees (Acer saccharum Marsh). J. Exp. Bot. 36, 12–24. Costanza-Robinson, M. S., Burwell, T., Chavis, A., Dickerson, N., Kelley, E., and Lovell, A. (2007). Geochemical investigation of Vermont maple syrup in the context of terroir. Environmental Science Practicum research report. Environmental Studies Program, Middlebury College, Middlebury, Vermont. Davis, D. R., Gallander, J. F., Hacskaylo, J., Gould, W. A., and Willits, C. O. (1963). The chemical composition of maple sugar sand. J. Food Sci. 28, 182–190. Deslauriers, L. (2000). Recovery, Separation and Characterization of Phenolic Compounds and Flavonoids from Maple Products. Masters Thesis. McGill University, Montreal. p. 105. Dumont, J. (1994). ‘‘L’eau d’e´rable.’’ Centre Acer Research Note. Publication no: 300-NTR1094. Dumont, J. (1996). ‘‘Projet de Recherche´: Inte´grite´ de Produits durable.’’ Rapport scientifique, Centre de recherche´ ace´ricole. Quebec, Canada. Edson, H. A. (1910). The influence of microorganisms upon the quality of maple syrup. J. Ind. Eng. Chem. 2, 325–327. Edson, H. A. and Jones, C. H. (1912). ‘‘Micro-Organisms of Maple Sap.’’ Vermont Agric. Exp. Sta. Bull. 167. University of Vermont and State Agricultural College, Burlington, VT. Favreau, D., Sosle, V., and Raghavan, G. S. (2001). Microwave processing of maple sap to maple syrup and maple syrup products. J. Microw. Power Electromagn. Energy 36, 25–35. Filipic, V. J., Underwood, J. C., and Dooley, C. J. (1969). Trace components of the flavor fraction of maple syrup. J. Food Sci. 34, 105. Haq, S. and Adams, G. A. (1961). Oligosaccharides from the sap of sugar maple (Acer saccharum Marsh). Can. J. Chem. 39, 1165–1170. Heiligmann, R. B., Koelling, M. R., and Perkins, T. D. (eds.) (2006). ‘‘North American Maple Syrup Producers Manual.’’ Ohio State University Extension Bulletin 856. Ohio State University. Isselhardt, M. L., van den Berg, A. K., and Perkins, T. D. (2007). Temperature patterns within an oil-fired maple evaporator. Maple Syrup Dig. 19A(3), 20–25. Kermasha, S., Goetghebeur, M., and Dumont, J. (1995). Determination of phenolic compound profiles in maple products using high-performance liquid chromatography. J. Agric. Food Chem. 43, 708–716. Labbe, R. G., Kinsley, M., and Wu, J. (2001). Limitation in the use of ozone to disinfect maple sap. J. Food Prot. 64, 104–107. LMEA (Les Manufacturiers d’E´quipements Ace´ricoles) (2001). ‘‘Standards on Maple Equipment Intended for the Production of Maple Syrup.’’ Saint-Adrien, Que´bec. Martin, G. G., Martin, Y., Naulet, N., and McManus, H. J. D. (1996). Application of 2H SNIFNMR and 13C SIRA-MS analyses to maple syrup: Detection of added sugars. J. Agric. Food Chem. 44, 3206–3213. Marvin, J. W. and Greene, M. T. (1959). ‘‘Some Factors Affecting the Yield from Maple Tapholes.’’ University of Vermont, Agricultural Experiment Station Bulletin 611. Milburn, J. A. and O’Mally, P. E. R. (1984). Freeze-induced sap absorption in Acer pseudoplantanus: A possible explanation. Can. J. Bot. 62, 2101–2106. Mollica, J. N. and Morselli, M. F. (1984). Gas chromatographic determination of nonvolatile organic acids in sap of sugar maple (Acer saccharum Marsh). JAOAC 67, 1125–1129. Morselli, M. F. (1975). Nutritional value of pure maple syrup. Maple Syrup Dig. 14, 12. Morselli, M. and Whalen, M. L. (1986). Amino acids increase in the xylem sap of Acer saccharum prior to budbreak. Am. J. Bot. 73, 722–723. Morselli, M. F. and Whalen, M. L. (1987). Salty syrup from roadside sugar maples in decline. Maple Syrup Dig. 27, 23–24.
142
Timothy D. Perkins and Abby K. van den Berg
Naghski, J. and Willits, C. O. (1957). Maple sirup. XI. Relationship between the type and origin of reducing sugars in sap and the color and flavor of maple sirup. J. Food Sci. 22, 567–571. Ngadi, M. O. and Yu, L. J. (2004). Rheological properties of Canadian maple syrup. Can. Biosys. Eng. 46, 3.15–3.18. Paradkar, M. M., Sakhamuri, S., and Irudayaraj, J. (2002). Comparison of FTIR, FT-Raman, and NIR spectroscopy in a maple syrup adulteration study. J. Food Sci. 67, 2009–2015. Paradkar, M. M., Sivakesava, S., and Irudayaraj, J. (2003). Discrimination and classification of adulterants in maple syrup with the use of infrared spectroscopic techniques. J. Sci. Food Agric. 83, 714–721. Perkins, T., Wilmot, T., and Zando, M. (2004a). Fertilization of sugarbushes—Part I. Physiological effects. Maple Syrup Dig. 16A, 23–27. Perkins, T., Wilmot, T., and Zando, M. (2004b). Fertilization of sugarbushes—Part II. Sap volume and sweetness. Maple Syrup Dig. 16A, 15–18. Perkins, T. D., Morselli, M. F., van den Berg, A. K., and Wilmot, T. R. (2006). Maple chemistry and quality. Appendix 2. In ‘‘North American Maple Syrup Producers Manual,’’ (R. B. Heiligmann, M. R. Koelling, and T. D. Perkins, eds.). Ohio State University Extension Bulletin 856. Ohio State University. Pollard, J. K. and Sproston, T. (1954). Nitrogenous constituents of sap exuded from the sapwood of Acer saccharum. Plant Physiol. 29, 360–364. Potter, T. L. and Fagerson, I. S. (1992). Phenolic compounds in maple syrup. In ‘‘Phenolic Compounds in Food and Their Effects on Health 1’’ (M. T. Huang, C. T. Ho, and C. Y. Lee, eds.), pp. 193–199. Am. Chem. Soc., Washington, DC. Potter, T. L., Fagerson, I. S., and Goldsmith, J. (1995). Mysteries of maple syrup flavor. Maple Syrup Dig. 7A, 9–13. Stilwell, D. E. and Musante, C. L. (1996). Lead in maple syrup produced in Connecticut. J. Agric. Food Chem. 44, 3153–3158. Stinson, E. E., Dooley, C. J., Purcell, J. M., and Ard, J. S. (1967). Quebrachitol. A new component of maple sap and syrup. J. Agric. Food Chem. 15, 394–397. Stuckel, J. G. and Low, N. H. (1995). Maple syrup authenticity analysis by anion-exchange liquid chromatography with pulsed amperometric detection. J. Agric. Food Chem. 43, 3046–3051. Stuckel, J. G. and Low, N. H. (1996). The chemical composition of 80 pure maple syrup samples produced in North America. Food Res. Int. 29, 373–379. Taylor, F. H. (1956). ‘‘Variation in Sugar Content of Maple Sap.’’ University of Vermont, Agricultural Experiment Station Bulletin 587. The´riault, M., Caillet, S., Kermasha, S., and Lacroix, M. (2006). Antioxidant, antiradical and antimutagenic activities of phenolic compounds present in maple products. Food Chem. 98, 490–501. Tremblay, P. and Paquin, R. (2007). Improved detection of sugar addition to maple syrup using malic acid as internal standard and in 13C isotope ratio mass spectrometry (IRMS). J. Agric. Food Chem. 55, 197–203. Tyree, M. T. (1983). Maple sap uptake, exudation and pressure changes correlated with freezing exotherms and thawing endotherms. Plant Physiol. 73, 277–285. Underwood, J. C. (1971). Effect of heat on the flavoring components of maple sirup. J. Food Sci. 36, 228–230. van den Berg, A., Perkins, T., and Isselhardt, M. (2006). Sugar profiles of maple syrup grades. Maple Syrup Dig. 18A(4), 12–13. van den Berg, A. K., Perkins, T. D., Isselhardt, M. L., Godshall, M. A. I., and Lloyd, S. W. (2009a). Effects of air injection during sap processing on maple syrup color, chemical composition and flavor volatiles. Int. Sugar J. 111, 37–42.
Maple Syrup—Production, Composition, Chemistry, and Sensory Characteristics
143
van den Berg, A. K., Perkins, T. D., Godshall, M. A., Lloyd, S. W., and Isselhardt, M. L. (2009b). Metabolism off-flavor in maple syrup Part I: Identification of the compound responsible for metabolism off-flavor. Maple Syrup Digest 21A(1), 15–18. Willits, C. O., Porter, W. L., and Buch, M. L. (1952). Maple sirip. V. Formation of color during evaporation of maple sap to sirip. J. Food Sci. 17, 482–486. Wilmot, T. R., Perkins, T. D., Stowe, B., and van den Berg, A. K. (2007a). Comparison of the ‘‘Small’’ Spout with the Traditional 7/16’’ Spout. Maple Syrup Dig. 19A(2), 20–26. Wilmot, T. R., Perkins, T. D., and van den Berg, A. K. (2007b). Vacuum sap collection—How High—or Low—Should You Go?. Maple Syrup Dig. 19A(3), 27–32. Wong, B. L., Baggett, K. L., and Rye, A. H. (2003). Seasonal patterns of reserve and soluble carbohydrates in mature sugar maple (Acer saccharum). Can. J. Bot. 81, 780–788.
CHAPTER
5 Maternal Fumonisin Exposure as a Risk Factor for Neural Tube Defects J. Gelineau-van Waes,* K. A. Voss,† V. L. Stevens,‡ M. C. Speer,§ and R. T. Riley†
Contents
I. Introduction II. Neural Tube Defects A. Neural tube defects: Overview B. Environmental risk factors for neural tube defects: Fumonisins C. Nutritional risk factors for neural tube defects: Folic acid D. Genetic risk factors for neural tube defects III. Fumonisin Exposures A. Fumonisins and regulatory policy B. Measurements of fumonisin exposure IV. Reproductive Toxicology of Fumonisins A. Animal studies: Overview B. Mouse models of fumonisin-induced neural tube defects V. Mechanisms of Fumonisin Toxicity A. Structural considerations B. Role of sphingolipids in fumonisin toxicity C. Fumonisin inhibition of de novo sphingolipid metabolism
146 148 148 148 149 151 153 153 154 155 155 156 158 158 160 162
* Department of Genetics, Cell Biology & Anatomy, University of Nebraska Medical Center, { { }
Omaha, Nebraska, USA Toxicology & Mycotoxin Research Unit, USDA Agricultural Research Service, Athens, Georgia, USA Department of Epidemiology and Surveillance Research, American Cancer Society, Atlanta, Georgia, USA Center for Human Genetics, Duke University Medical Center, Durham, North Carolina, USA
Advances in Food and Nutrition Research, Volume 56 ISSN 1043-4526, DOI: 10.1016/S1043-4526(08)00605-0
#
2009 Elsevier Inc. All rights reserved.
145
146
J. Gelineau-van Waes et al.
D. Fumonisin and accumulation of bioactive sphingoid base-1-phosphates E. Fumonisin depletion of glycosphingolipids and disruption of folate transport VI. Conclusions References
Abstract
164 166 170 171
Fumonisins are mycotoxins produced by the fungus F. verticillioides, a common contaminant of maize (corn) worldwide. Maternal consumption of fumonisin B1-contaminated maize during early pregnancy has recently been associated with increased risk for neural tube defects (NTDs) in human populations that rely heavily on maize as a dietary staple. Experimental administration of purified fumonisin to mice early in gestation also results in an increased incidence of NTDs in exposed offspring. Fumonisin inhibits the enzyme ceramide synthase in de novo sphingolipid biosynthesis, resulting in an elevation of free sphingoid bases and depletion of downstream glycosphingolipids. Increased sphingoid base metabolites (i.e., sphinganine-1-phosphate) may perturb signaling cascades involved in embryonic morphogenesis by functioning as ligands for sphingosine-1-P (S1P) receptors, a family of G-proteincoupled receptors that regulate key biological processes such as cell survival/proliferation, differentiation and migration. Fumonisininduced depletion of glycosphingolipids impairs expression and function of the GPI-anchored folate receptor (Folr1), which may also contribute to adverse pregnancy outcomes. NTDs appear to be multifactorial in origin, involving complex gene-nutrientenvironment interactions. Vitamin supplements containing folic acid have been shown to reduce the occurrence of NTDs, and may help protect the developing fetus from environmental teratogens. Fumonisins appear to be an environmental risk factor for birth defects, although other aspects of maternal nutrition and genetics play interactive roles in determining pregnancy outcome. Minimizing exposures to mycotoxins through enhanced agricultural practices, identifying biomarkers of exposure, characterizing mechanisms of toxicity, and improving maternal nutrition are all important strategies for reducing the NTD burden in susceptible human populations.
I. INTRODUCTION Fumonisins are mycotoxins produced by the fungus Fusarium verticillioides, F. proliferatum, and more rarely, other Fusarium species (Bolger et al., 2001; Gelderblom et al., 1988) (Fig. 5.1A). They are found in variable amounts in maize (corn) and maize-based foods worldwide
147
Fumonisin and Neural Tube Defects
COOH O HOOC O
OH
OH 1
Me O
Me
OH
NH2
HOOC COOH O
Fusarium verticillioides
Fumonisin
FIGURE 5.1 The photograph on the left shows an ear of maize infected with the Fusarium verticillioides fungus. The mold is white to pinkish in color, and produces the fumonisin toxin. Decay often begins with insect-damaged kernels, however, levels of fumonisins capable of causing toxicity in rodents can occur in asymptomatic maize. The chemical structure shown on the right is that of the mycotoxin fumonisin B1. Fumonisin was first isolated in 1988 by investigators at PROMEC (program on mycotoxins and experimental carcinogenesis) in Tygerberg, South Africa.
(Bolger et al., 2001; Humpf and Voss, 2004) and their presence in other commodities has occasionally been reported (Castoria et al., 2005; da Silva et al., 2000; Kritzinger et al., 2003). Consumption of fumonisin B1-contaminated maize has been associated with a variety of different diseases in animals, including leukoencephalomalacia in horses (Smith et al., 2002) and pulmonary edema in swine (Constable et al., 2000, 2003; Haschek et al., 2001). Independent studies have also established a causal relationship between fumonisin B1 exposure and liver and kidney toxicity (Bolger et al., 2001; Voss et al., 2001), and/or liver and kidney carcinogenicity in rodents (Gelderblom et al., 1991; Howard et al., 2001a,b). The International Agency for Research on Cancer has evaluated fumonisin B1 as a possible human carcinogen (Group 2B) (IARC, 2002). Although the human health effects of fumonisins are not proved, the consumption of foods made from F. verticillioides- or fumonisin B1contaminated maize has been associated with high rates of esophageal cancer in parts of southern Africa, China, and northeastern Italy (Bolger et al., 2001, Shephard et al., 2007). Elevated levels of fumonisin have also been found in maize from Mazandaran Province in Iran, another region with a high incidence of esophageal cancer (Shephard et al., 2002; Yazdanapanah et al., 2006). However, fumonisin exposures in this and other areas of Iran in which esophageal cancer is less frequent were low ( 0.22 mg/kg body weight/day) due to the relatively small amount of maize consumed (Yazdanpanah et al., 2006). In terms of human health risks, maternal consumption of fumonisin B1-contaminated maize or maize-based food products during early gestation has recently been associated with increased risk for birth defects, specifically, neural tube
148
J. Gelineau-van Waes et al.
defects (NTDs) (Hendricks, 1999; Kromberg and Jenkins, 1982; Marasas et al., 2004; Melnick and Marazita, 1998; Missmer et al., 2006; Moore et al., 1997; Ncayiyana, 1986; Xiao et al., 1990).
II. NEURAL TUBE DEFECTS A. Neural tube defects: Overview Approximately 3–4% of all newborns have a significant abnormality in body structure or function. Birth defects (congenital anomalies) are the leading cause of death in children under 1 year of age. NTDs, which occur with a frequency of approximately 1/1000 live births (Nakano, 1973), are among the most common of all human birth defects, yet their etiologic basis and embryology remain poorly understood. NTDs are common congenital malformations that occur when the embryonic neural tube, which ultimately forms the brain and spinal cord, fails to properly close during the first few weeks of development. Anencephaly, which is essentially the absence of the brain, is invariably fatal, and results from failure of the anterior neural folds to properly elevate and fuse along the dorsal midline during early embryogenesis. Spina bifida refers to incomplete development of the posterior neural tube, and failure of fusion of one or more vertebral arches, often accompanied by protrusion of the spinal cord and its associated membranes. Patients with spina bifida can have a variety of associated conditions, the severity of which is often determined by the level of the lesion. Most commonly, patients with spina bifida have difficulty walking and require either braces or a wheelchair, have hydrocephalus requiring shunting, and/or have difficulty with bowel and bladder control. Empirical risk figures, along with numerous clinical and experimental studies, suggest that NTDs are multifactorial in origin, having genetic, environmental, and nutritional components that contribute to their prevalence (Campbell et al., 1986).
B. Environmental risk factors for neural tube defects: Fumonisins Maternal fumonisin exposure has been proposed as a potential risk factor for human NTDs among populations consuming large amounts of fumonisin-contaminated maize or maize-based products (Marasas et al., 2004). Globally, NTDs present a tremendous burden to human populations in rural areas of the world where maize is a dietary staple, and fumonisin contamination is common (Hendricks, 1999; Kromberg and Jenkins, 1982; Marasas et al., 2004; Melnick and Marazita, 1998; Moore et al., 1997;
Fumonisin and Neural Tube Defects
149
Ncayiyana, 1986; Torres et al., 2007; Xiao et al., 1990). The incidence of NTDs in these regions, including Guatemala, northern China, and the Transkei of South Africa, is often 6–10 times higher than the average global neural tube defect rate (Hendricks, 1999; Kromberg and Jenkins, 1982; Marasas et al., 2004; Melnick and Marazita, 1998; Moore et al., 1997; Ncayiyana, 1986; Xiao et al., 1990). Since fumonisins are ubiquitous, and maize-based meals are often the primary food commodity available, human exposures can be significant (Marasas, 1996; Meredith et al., 1999; Miller, 2001; Shephard et al., 1996, 2007). In Guatemala, a recent analysis of fumonisin B1 in maize samples, coupled with data on daily maize intake, demonstrated that human consumption of maize products could frequently result in fumonisin exposures exceeding the recommended World Health Organization (WHO) provisional maximal tolerable daily intake (Torres et al., 2007). In the United States, the Texas Department of Health reported a neural tube defect cluster among Mexican-American women along the south Texas border in 1990–1991 (Hendricks, 1999). Corn crops in the Lower Rio Grande Valley registered unusually high levels of fumonisin B1 during this period, and epidemiological studies revealed that Cameron County women who conceived during 1990–1991 had a substantially higher neural tube defect rate (2.7/1000 live births) than those who conceived during the period from 1986 to 1989 (1.5/1000 live births). MexicanAmericans in Texas consume large quantities of maize, primarily in the form of tortillas, and therefore may be exposed to high levels of fumonisins. Results of a follow-up population-based case-control study conducted among the south Texas Mexican-American population suggested that fumonisin exposure increased the risk of NTDs, proportionate to dose, up to a threshold level, at which point fetal death was postulated to be more likely to occur (Missmer et al., 2006).
C. Nutritional risk factors for neural tube defects: Folic acid The etiology of NTDs is largely unknown, but appears to be multifactorial in origin, involving complex gene-nutrient-environment interactions. Consumption of large quantities of fumonisin-contaminated maize or maize-based food products during early gestation may represent an environmental risk factor for having a child with a neural tube defect. However, other aspects of maternal nutrition are also important in ensuring a positive pregnancy outcome, and adequate maternal folate appears to play a critical role in protecting the developing fetus from the harmful effects of environmental teratogens. Diets heavily dependent on maize are likely to be folate deficient since maize contains only low levels of this vitamin (Burton et al., 2008). In Mexico and Central America, maizebased foods are typically prepared using an alkaline process known as
150
J. Gelineau-van Waes et al.
nixtamalization (Palencia et al., 2003) which has been shown to further reduce folate, riboflavin, and other vitamins (Burton et al., 2008; Ca´rdenas et al., 2001). Folate is an essential vitamin derived from plant sources that plays an important role in DNA biosynthesis and amino acid metabolism. Multiple transport systems play a role in mediating internalization of folates through the plasma membrane into the cell for utilization in a variety of critical housekeeping functions. The primary mechanisms for folate delivery into the cell are through (1) carrier-mediated (reduced folate carrier; RFC1) or (2) receptor-mediated (folate receptor a, b, d, g) processes. These two transport systems are distinguished by their unique patterns of tissue expression, divergent specificities for oxidized vs. reduced folates, and differing protein structures and mechanisms for the transmembrane transport of folates. As mammalian cells are not capable of synthesizing folates de novo, these two transport systems play a critical role in mediating folate uptake for the biosynthesis of purines, pyrimidines, and some amino acids that are necessary for cell survival and proliferation. The murine folate binding protein 1 (Folr1) is homologous to the human folate receptor-alpha (FRa), a membrane bound protein with high binding affinity for folic acid. During early development, Folr1 is highly expressed in the placenta, yolk sac membrane, and dorsal neural tube, and is later observed in the choroid plexus, ependymal cells, and peripheral edge of the retina (Saitsu et al., 2003). The glycosylphosphatidylinositol (GPI)-anchored Folr1 internalizes folate via endocytosis, and is localized in unique sphingolipid-enriched plasma membrane microdomains known as lipid rafts (Elortza et al., 2003; Foster et al., 2003). Lipid rafts function as specialized platforms for coordinating protein–protein interactions involved in the initiation of signal transduction cascades and intracellular signaling (Brown and London, 1998). Since most embryonic cells do not express Folr1 (Page et al., 1993), the presence of these receptors in neuroepithelial cells (Barber et al., 1999; Saitsu et al., 2003) suggests a critical role for Folr1-mediated folate transport in the normal morphogenetic events involved in neural tube closure (Piedrahita et al., 1999; Spiegelstein et al., 2004). Human clinical and epidemiological studies have demonstrated that maternal use of folic acid in early pregnancy can significantly reduce both the occurrence (Czeizel and Dudas, 1992), as well as the recurrence (MRC, 1991) of neural tube defect-affected pregnancies. These findings have been further validated by observational studies of women taking daily periconceptional multivitamin supplements containing folic acid (Shaw et al., 1995; Werler et al., 1993). However, while the epidemiologic and experimental data support the hypothesis that this apparent
Fumonisin and Neural Tube Defects
151
reduction in neural tube defect risk may be specifically attributable to folic acid, the mechanisms underlying the protective effects of folic acid are not fully understood. The incidence of NTDs is elevated in regions of the world (Guatemala, South Africa, China) where maize consumption with fumonisin is documented or plausible and where diets are also likely to be deficient in folate (Chu and Li, 1994; Cifuentes, 2002; Marasas, 2001; Marasas et al., 2004; Shannon and Fenichel, 1990; Yoshizawa et al., 1994). The high incidence of NTDs among Mexican-American women living in the U.S.–Mexico border region in Texas is likely related to maternal fumonisin exposure coupled with folate deficiency (Hendricks, 1999; Marasas et al., 2004; Missmer et al., 2006). Interestingly, fumonisin content in Guatemalan and U.S. maize samples has been shown to be similar in some years (Marasas et al., 2004); however, Guatemalans typically consume more maize than Americans, resulting in significantly higher fumonisin exposures. Finally, other studies found a 50% decrease in the incidence of NTDs with folic acid supplementation in Mexico (Martinez de Villarreal et al., 2002), suggesting that folic acid supplementation may offer some protection against the adverse effects of fumonisin exposure.
D. Genetic risk factors for neural tube defects In addition to environmental and nutritional risk factors, there is clearly a genetic susceptibility component that contributes to the incidence of congenital malformations. Neural tube defect risk has been shown to vary across ethnic groups, suggesting that genetic variation (polymorphisms) may predispose particular individuals, or groups of individuals to NTDs. Several lines of evidence suggest a genetic component to this type of birth defect. First, NTDs are associated with known genetic syndromes, including Meckel syndrome, anterior sacral meningomyelocele, and anal stenosis. Secondly, in NTDs occurring without other syndromes, the recurrence risk for siblings is approximately 2–5%, which is up to a 50-fold increase over that observed in the general population. Khoury et al. (1988) have shown that for a recurrence risk to be this high, an environmental teratogen would have to increase the risk at least 100-fold to exhibit the same degree of familial aggregation, making a genetic component essentially required. Additionally, a variety of mutations leading to NTDs in mice have been documented, providing further evidence in support of a genetic factor in humans. To date, more than 80 different murine loci predisposing to NTDs have been identified (Copp et al., 2003; Juriloff and Harris, 2000). The complex, early embryological development of NTDs presents fascinating challenges to the geneticist. One traditional approach to identifying disease genes in families has been
152
J. Gelineau-van Waes et al.
through genomic screening of multiplex pedigrees. Advances in genetic marker availability, including the development of highly polymorphic microsatellite repeat markers (STRPs) and single nucleotide repeat polymorphisms (SNPs), the dense genetic maps of microsatellites (Broman et al., 1998) and of SNPs (Matise et al., 2003), the development of highthroughput genotyping methods (Ben Othmane et al., 1998), and innovations in statistical genetic linkage analysis (Gudbjartsson et al., 2000; Kruglyak et al., 1996; Kong and Cox, 1997; Martin et al., 2000; O’Connell and Weeks, 1995) have paved the way for large-scale investigations of complex human diseases. Recently, a genomic screen of 44 multiplex pedigrees (those with at least two sampled affected individuals) was performed using highly polymorphic microsatellite repeat markers and identified regions of interest on chromosomes 7 and 10 (Rampersaud et al., 2005). Occasionally, large multiplex pedigrees—those with three or more affected individuals—are identified and such families can be invaluable for identifying a gene associated with NTDs (Stamm et al., 2006). Presumably, the affected individuals, since related to one another, all share the same genetic variant. Once a putative causative genetic variant is identified in such a family, its relevance in other families can be quickly assessed. Another common approach to characterizing the genetic influence of a condition is to assess biologically plausible candidate genes—genes whose involvement in a process is suggested by the known function of a gene. In NTDs, biologically plausible candidate genes might be those known to be involved in neural tube closure in other animal systems, are known to be important generally in early development, or are involved with folate metabolism. Folate metabolism pathway genes have often been examined for association with NTDs due to the well-established ability of folic acid to reduce the risk of this debilitating birth defect by 50–70%. Some studies have shown that the protective effect of folic acid may be decreased in populations of Hispanic descent (Shaw et al., 1995; Suarez et al., 2000). Blood folate levels have a strong genetic component with an estimated heritability of 46% (Morrison et al., 1998), yet maternal folate supplementation can only prevent 50–70% of NTDs (Chatkupt et al., 1994). Folic acid supplementation does not entirely eliminate recurrence risk (Chatkupt et al., 1994; Milunsky et al., 1989), suggesting that additional, genetic factors are responsible for the development of this type of birth defect. These non-folate responsive cases may represent highly genetic cases of NTDs (Scriver, 1985). The ability to identify those women genetically predisposed to having a child with a neural tube defect, and whose risk could be minimized by the use of folic acid supplements, would allow genotype-directed pharmacogenetic interventions. The mechanism by which dietary folate supplementation prevents NTDs is not understood (Prevention of Neural Tube Defects: Results of
Fumonisin and Neural Tube Defects
153
the Medical Research Council Vitamin Study, MRC Vitamin Study Research Group, 1991). Folic acid derivatives are essential for the synthesis of DNA, as well as for cell division, tissue growth, and DNA methylation (Morrison et al., 1998), and may also protect cells against oxidative damage. DNA methylation enables proper gene expression and chromosome structure maintenance, both of which are critical in the developing embryo (Razin and Kantor, 2005). Low dietary folate levels could directly limit folate availability to cells or indirectly disrupt methionine metabolism, thereby increasing homocysteine in the maternal serum (Rosenquist and Finnell, 2001). Either mechanism implicates folate receptor and methionine–homocysteine regulatory genes. Thus, the study of variation in genes involved in folate metabolism makes good biologic sense. NTDs in humans result from the combined effects of genetic, nutritional, and environmental influences, and as such are a classic example of a multifactorial disorder. Identifying the genetic factors will be critical for characterizing the interactions between genes and the environment, and understanding these interactions will provide the basis for identifying individuals at greatest risk for having a child with a birth defect, as well as facilitating the design of novel preventive strategies.
III. FUMONISIN EXPOSURES A. Fumonisins and regulatory policy Based on the adverse health effects observed in animals, the Center for Food Safety and Nutrition, U.S. Food and Drug Administration (FDA) issued industry guidance levels for fumonisins in maize in 2001 (‘‘Guidance for Industry on Fumonisin Levels in Human Foods and Animal Feeds’’; http://cfsan.fda.gov/dms/fumongu2.html) stating that human health risks associated with exposure to fumonisins were possible. Recommended levels for foods range from 2 to 4 ppm total fumonisins (fumonisin B1 plus fumonisin B2 plus fumonisin B3) depending on the intended use of the maize and whether or not the kernels are whole or degermed. Proposed limits (fumonisin B1 plus fumonisin B2) in Europe, (effective October 2007), range from 0.2 to 2 ppm depending upon the commodity and the targeted consumer (i.e., adults or infants) (Commission of European Communities, 2005). The International Agency for Research on Cancer (IARC, 2002) classified fumonisin B1 as ‘‘possibly carcinogenic to humans (class 2B)’’ and in 2001 the Joint FAO/WHO Expert Committee of Food Additives (JECFA) recommended a provisional maximum tolerated daily intake (PMTDI) of 2 mg/kg body weight for fumonisin B1, fumonisin B2 and fumonisin B3, ‘‘alone or in
154
J. Gelineau-van Waes et al.
combination’’ (Bolger et al., 2001), and suggested that further research was needed to examine the teratogenic potential of this compound.
B. Measurements of fumonisin exposure One of the difficulties in studying the environmental component to birth defects is that the time of ‘‘injury’’ precedes the occurrence of the condition by weeks or months. In humans, neural tube closure is complete by approximately 28 days after conception, frequently before a woman even knows she is pregnant. Prospective studies of exposure would require extensive population screening to identify the occasional affected case and would require years to develop a large enough sample size for the analysis of gene–environment interactions. Thus, use of biomarkers at time points after the event of interest (e.g., pregnancy) may be a useful approach for characterizing environmental influences (Krapels et al., 2004; Leck, 1983; Murray et al., 1997; Prakash et al., 2002; Weis et al., 2005). Fumonisin B1 is rapidly cleared from the system and can only be detected in human urine and feces for up to 3–5 days after ingestion (Sewram et al., 2003; Riley, unpublished data). In the absence of modern technology for prenatal monitoring (i.e., ultrasound), fetal malformations may not be detected until the child is born. At this point, it is too late to accurately determine the level of maternal fumonisin exposure that occurred during the critical gestational window spanning neural tube closure (i.e., 3–4 weeks post-conception). A possible surrogate for direct fumonisin exposure is a rise in the sphinganine/sphingosine ratio, but this measurement is also limited to short-term detection (Sewram et al., 2003) unless exposure is fairly constant. Several studies in rodents have shown that once elevated by a high dose, the levels of sphinganine in either urine or kidney will remain significantly elevated by exposure to a low dose of fumonisin that would not ordinarily elicit an increase in sphinganine (Enongene et al., 2002; Wang et al., 1999). For example, rats were fed diets containing 10 ppm of fumonisin B1 for 10 days and then changed to either diets containing 0 or 1 ppm fumonisin B1 (Wang et al., 1999). In the rats consuming the 0 ppm diet, the urinary sphinganine returned to control values (0.2 nmol/ml) by day 20, whereas in the animals changed to the 1 ppm fumonisin B1 diet, sphinganine was still significantly elevated on day 20 (2.5 nmol/ml). Chemical analysis of hair samples may also provide a method for examining chronic mycotoxin exposures. In 2003, Sewram et al. (2003) reported that human hair testing could be used to detect fumonisins. After extraction and clean up, high performance liquid chromatography coupled to electrospray ionization-mass spectrometry (HPLC-ESI-MS) was able to detect fumonisin B1, fumonisin B2, and fumonisin B3 from human hair samples (Sewram et al., 2003). However, these were
Fumonisin and Neural Tube Defects
155
composite samples taken from barbershops in various regions of South Africa. Thus, hair testing has not yet been employed to measure individual fumonisin exposure level. Hair analysis has many potential advantages. Compared to urine and blood testing, which are only able to detect exposures from 2 to 4 days past, hair analysis provides a larger window of xenobiotic detection. Exposure detection ranges from weeks to months depending upon the length of the hair (Kintz, 2004) and assessment of the hair in segments can be performed so that the segment most closely reflecting time-at-exposure is selected for study. Nonetheless, the method for measuring fumonisin in hair has not been validated so that levels in hair can be used to predict past or ongoing exposure. More recently, a simple method for measuring fumonisin in urine samples from women in a cohort recruited in Morelos County, Mexico has correlated urinary fumonisin B1 with tortilla consumption and suggests that urinary fumonisin B1 could be a valuable tool in investigating the associated health effects of exposure (Gong et al., 2008). As part of the evaluations of the cluster of NTDs along the Texas– Mexico border in 1990–1991 (Missmer et al., 2006), three different metrics to evaluate fumonisin exposure were considered. These metrics included (1) number of tortillas consumed during the first trimester (collected via questionnaire); (2) post-partum sphinganine/sphingosine ratio to estimate fumonisin exposure from maternal blood; and (3) nanograms of fumonisin ingested per day during the periconceptional period, estimated from grouped 6-month averages for sampled tortillas from homes and markets. The outcome of this study suggested that ingestion of maize-based products in the amounts observed in Mexican-American diets were associated with increased risk for NTDs. All of the experimental findings were robust across exposure metrics, suggesting that the questionnaire data were a realistic measure of fumonisin exposure, particularly when coupled with estimates of seasonal variation in fumonisin levels.
IV. REPRODUCTIVE TOXICOLOGY OF FUMONISINS A. Animal studies: Overview Previous reproductive toxicology studies in laboratory animals examining the effects of prenatal exposure to fumonisin demonstrated a potential risk to the developing fetus. Studies using an aqueous extract of contaminated maize-culture material of F. verticillioides reported that fumonisin was developmentally toxic in hamsters (Floss et al., 1994; Penner et al., 1998). In addition, purified fumonisin B1 was shown experimentally to cause fetal toxicity in rats and mice (Collins et al., 1998; Reddy et al., 1996). In another study, pregnant CD1 mice treated with a semipurified extract
156
J. Gelineau-van Waes et al.
of F. verticillioides culture material exhibited a dose-dependent increase in the number of resorptions and fetal abnormalities (Gross et al., 1994). However, in several of the studies, developmental abnormalities also occurred in the presence of maternal toxicity. In animal model systems, cultured rat embryos exhibited an increase in the incidence of NTDs (Flynn et al., 1997) following exposure to the hydrolyzed fumonisin B1, also known as aminopentol 1 (AP1), which is produced during the nixtamalization process used to make maize flour for masa and tortillas (Dombrink-Kurtzman et al., 2000). Sadler et al. (2002) reported an association between fumonisin B1 exposure and NTDs in murine embryo culture and showed that folinic acid supplementation was able to ameliorate the teratogenic effects of fumonisin. Early gestational exposure to the mycotoxin also induces a high incidence of NTDs in offspring of LM/Bc mice (Gelineau-van Waes et al., 2005), and a somewhat lower, but biologically significant incidence of NTDs in embryos of CD1 mice following maternal intraperitoneal injections of purified fumonisin B1 (Voss et al., 2006). In the inbred LM/Bc mouse strain, maternal administration of 20 mg/kg/day fumonisin B1 on gestational days 7.5 and 8.5 (intraperitoneal injection) resulted in NTDs (exencephaly, which is the mouse equivalent of anencephaly in humans) in all litters examined (n ¼ 10 litters), and in 79% of the exposed embryos. Similar to the results obtained in murine embryo culture (Sadler et al., 2002), maternal supplementation with folic acid was able to reduce the in vivo incidence of fumonisin-induced NTDs in LM/Bc mice by approximately 50% (Gelineau-van Waes et al., 2005). Preliminary results using the LM/Bc mouse model (n ¼ 3 l) indicate that the oral route of fumonisin exposure (20 mg/kg of fumonisin B1 on E7.5 and E8.5; equivalent to 80–120 mg of fumonisin B1/kg diet) also results in NTDs (20%) in exposed embryos.
B. Mouse models of fumonisin-induced neural tube defects 1. Fumonisin B1-induced neural tube defects in cultured mouse embryos
Sadler et al. (2002) assessed the effect of fumonisin B1 treatment on development of neurulating mouse embryos in culture. Fumonisin B1 treatment resulted in a dose-dependent increase in the percentage of embryos exhibiting NTDs (Sadler et al., 2002). Concurrent alterations in levels of sphingoid bases were also found. Supplementation of the embryos with high levels of folate during the fumonisin B1 treatment reduced the incidence of NTDs without correcting the sphingolipid abnormalities, suggesting that the mycotoxin influenced neural tube maturation by affecting the function of the folate receptor. Cumulatively, these findings provided a conceptual framework to account for how
Fumonisin and Neural Tube Defects
157
exposure to fumonisin B1 could increase the risk of NTDs by disrupting lipid rafts through sphingolipid depletion and, consequently, impaired folate receptor function.
2. Maternal fumonisin exposure and neural tube defects Further evidence of a causal relationship between gestational fumonisin exposure and altered embryonic morphogenesis was provided by the establishment of an in vivo mouse model (Gelineau-van Waes et al., 2005). Early gestational administration of the purified fumonisin B1 mycotoxin to pregnant mice of the inbred SWV and LM/Bc mouse strains resulted in an increased incidence of fetal resorptions and/or malformations (Gelineau-van Waes et al., 2005). Although reproductive toxicology studies in which purified fumonisin B1 was administered to hamsters (Penner et al., 1998) or fumonisin-contaminated culture material was administered to mink (Powell et al., 1996) reported a decrease in birth weight and/or an increase in fetal death as the primary finding(s), very few LM/Bc embryos/fetuses were resorbed, even at the highest dose of fumonisin administered. Rather, a significant proportion of fumonisinexposed LM/Bc embryos failed to complete neurulation. The number of affected fetuses per litter increased as the dose of fumonisin was increased, and at the highest dose administered (20 mg/kg/day on E7.5 and E8.5), the litters of all treated dams were positive for NTDs, and 79% of the exposed LM/Bc fetuses were exencephalic (Gelineau-van Waes et al., 2005) (Fig. 5.2).
A
B
Control
Exencephalic
FIGURE 5.2 (A) Image of a normal (vehicle control) E10.5 embryo; (B) image of an E10.5 exencephalic embryo collected from an LM/Bc dam that was treated with purified fumonisin B1 early in gestation (20 mg/kg i.p. on E7.5 and E8.5).
158
J. Gelineau-van Waes et al.
The same fumonisin treatment regime performed in mice from the highly inbred SWV strain, however, yielded very different results. At the highest fumonisin B1 dose administered (20 mg/kg/day on E7.5 and E8.5), approximately 15% of the implants were resorbed, and only one exencephalic fetus was observed in the 10 SWV litters examined. Sphinganine levels were elevated in the placentas and embryos of both strains, although to a much greater extent in the LM/Bc mice, suggesting that genetics plays a role in the observed differences in sphingolipid metabolites and pregnancy outcome after fumonisin exposure. Observations from preliminary feeding studies (summarized in Voss et al., 2006) are consistent with the concept of strain-dependent differences. In these studies, LM/Bc and CD1 mice were fed fumonisin-contaminated diets beginning 5 weeks before mating and continuing throughout gestation. At a dietary fumonisin B1 concentration of 150 ppm, late fetal deaths occurred in two of the nine (22%) CD1 litters, and the incidence of late fetal deaths in affected litters ranged from 40 to 64%. No NTDs were found in the CD1 mice. In contrast, no late fetal deaths were found in the five LM/Bc litters, but one LM/Bc litter was positive for NTDs. The results from the studies in LM/Bc and SWV mice, in conjunction with the preliminary results of the feeding studies comparing the LM/Bc and CD1 strains clearly point to an as yet unknown genetic susceptibility component with respect to risk for NTDs following maternal fumonisin B1 exposure. In the LM/Bc mouse strain, maternal fumonisin exposure resulted in decreased expression of folate receptor (Folr1) protein in the yolk sac membrane and neuroepithelial cells of developing embryos. In addition, reduced levels of 3H-folate uptake by neurulating embryos demonstrated that fumonisin depletion of glycosphingolipids (gangliosides) had an impact on both expression and function of the GPI-anchored folate receptor. Moreover, similar to the results observed in the embryo culture model of fumonisin exposure, supplementing pregnant dams with high levels of folic acid (50 mg/kg/day on E0.5–E9.5) afforded some protection against fumonisin teratogenicity, and reduced the incidence of NTDs from 79 to 50% in exposed LM/Bc fetuses (Gelineau-van Waes et al., 2005).
V. MECHANISMS OF FUMONISIN TOXICITY A. Structural considerations The chemical structure of fumonisin (Fig. 5.1B) is remarkably similar to that of the sphingoid bases sphinganine (Sa) and sphingosine (So), and fumonisin has been shown to inhibit the enzyme ceramide synthase in de novo sphingolipid metabolism (Wang et al., 1991). Ceramide synthase
Fumonisin and Neural Tube Defects
159
catalyzes the formation of ceramide from the condensation of fatty-acyl CoA and sphinganine (Sa) or sphingosine (So). Inhibition of ceramide synthase by fumonisins is competitive (Merrill et al., 2001; Riley and Voss, 2006; Riley et al., 2001, 2006) and likely involves the tricaballylic acid and amine groups, which occupy the enzyme’s respective fatty-acyl CoA and the sphingoid base binding sites (Merrill et al., 2001). Experimental evidence indicates that the primary amine is essential for the fumonisin molecule’s ceramide synthase inhibitory activity (Lemke et al., 2001; Norred et al., 2001), which is believed to be the mechanistic ‘‘trigger’’ for toxicity. Inhibition of ceramide synthase results in an accumulation of upstream sphingoid bases and sphingoid base-1-phosphates, and a depletion of downstream complex glycosphingolipids (Fig. 5.3). To date, at least 28 analogues of fumonisin have been identified (Reeder et al., 2002). Fumonisin B1 (FB1) is the most common and, from a toxicological standpoint, the most thoroughly studied. Structures of the less common
Fumonisin inhibits de novo sphingolipid biosynthesis Palmitoyl CoA + serine
Sphinganine Fumonisin
Sphinganine-1-P
Ceramide synthase Ceramide
Sphingosine
S1P receptors Sphingosine-1-P
Glucosylceramide GPI-anchored folate receptor (folr1) Lactosylceramide
Lipid raft Gangliosides
FIGURE 5.3 Simplified schematic of the de novo sphingolipid biosynthetic pathway. Fumonisin inhibits the enzyme ceramide synthase, resulting in (1) an accumulation of sphinganine, which is subsequently phosphorylated to form bioactive sphinganine-1phosphate, a ligand for the G protein-coupled S1P receptors; and (2) a depletion of downstream glycosphingolipids (gangliosides), important components of ‘‘lipid rafts’’ in which GPI-anchored receptors such as the folate receptor (Folr1) are found.
160
J. Gelineau-van Waes et al.
fumonisin B2, and fumonisin B3, differ from that of fumonisin B1 by the number and position of hydroxyl groups on the backbone. Base hydrolysis removes the tricarballylic acid groups yielding the corresponding hydrolyzed fumonisins (HFB1, HFB2) (DombrinkKurtzman et al., 2000; Merrill et al., 2001). Hydrolyzed fumonisins form during the traditional corn cooking (in alkaline water) method known as nixtamalization that is used to make masa for tortillas (DombrinkKurtzman et al., 2000; Palencia et al., 2003). Hydrolyzed fumonisin B1 inhibits ceramide synthase less effectively in vitro (Merrill et al., 2001) and, in contrast to fumonisin B1, was not toxic when fed to female mice for 28 days (Howard et al., 2002). Likewise, N-(acetyl)-fumonisin B1 and N-(carboxymethyl)-fumonisin B1 which, like N-(1-deoxy-D-fructose-1-yl) fumonisin B1, lack a free primary amine group, were not toxic and did affect sphingolipid metabolism when fed to female mice (Howard et al., 2002). Heating fumonisin B1 with reducing sugars such as glucose yields browning reaction products such as N-(1-deoxy-D-fructose-1-yl) fumonisin B1 and N-(carboxymethyl)-fumonisin B1 (Lu et al., 2002). Fumonisin B1 also forms conjugates when heated with model starch or protein compounds (Humpf and Voss, 2004). In this case, however, binding likely occurs through the tricarballylic group. The extent to which fumonisin binds to food matrices and the bioavailability of matrix-bound fumonisins is unknown. However, the presence of matrix bound (or ‘‘hidden’’) fumonisins in maize-based cereal (Kim et al., 2003), and alkali-processed food (Park et al., 2004) has been demonstrated.
B. Role of sphingolipids in fumonisin toxicity Several biochemical modes of action have been postulated to explain fumonisin-induced toxicity and carcinogenicity in animals, but the primary hypothesis involves perturbation of sphingolipid metabolism (Bolger et al., 2001; IARC, 2002; Wang et al., 1991; WHO, 2000). Sphingolipids are important structural components of the plasma membrane and also function as intra- and intercellular signaling molecules (Chalfant and Spiegel, 2005; Merrill et al., 2001). The toxic effects of fumonisin appear to depend on disruption of various aspects of lipid metabolism, membrane structure, and signal transduction pathways mediated by lipid second messengers. The demonstrated effects include altered rates of cell proliferation vs. apoptosis, altered cell–cell communication and cell adhesion, oxidative stress, and modulation of gene expression (Abel and Gelderblom, 1998; Bhandari et al., 2002; Seefelder et al., 2003). The biomolecular events linking ceramide synthase inhibition, disruption of sphingolipid metabolism, and toxicity have not been elucidated and mechanistic roles for oxidative damage (Abel and Gelderblom, 1998; Kouadio et al., 2005; Lemmer et al., 1999; Sahu et al., 1998), critical changes
Fumonisin and Neural Tube Defects
161
in cell fatty acid composition (Gelderblom et al., 2001), up- or downregulation of transcription or post-transcriptional stabilization of cytokines and other signaling molecules (Bhandari et al., 2002; Bondy et al., 2000; Gelderblom et al., 2001; Merrill et al., 2001; Sharma et al., 2003) have also been proposed. The extent to which such events occur independently or secondary to sphingolipid metabolism disruption is unknown. Aside from influencing apoptosis and mitosis (Dragan et al., 2001; Howard et al., 2001a), fumonisin interferes with cell–cell or cell–matrix adhesion and membrane permeability, perhaps through sphingolipid dependent mechanisms (Gon et al., 2005; Pelagalli et al., 1999; Ramasamy et al., 1995). It is noteworthy that detachment and sloughing of tubule epithelial cells is a feature of fumonisin-induced renal disease (Hard et al., 2001). In addition, complex sphingolipids are important for the uptake and trafficking of folate (Stevens and Tang, 1997) via the GPI-anchored folate receptor. Apoptosis and mitosis in liver (hepatocytes) and kidney (proximal tubule epithelium of the outer medulla) are early consequences of fumonisin exposure (Bolger et al., 2001; Dragan et al., 2001; Howard et al., 2001a; Voss et al., 2001). They occur simultaneously and it is proposed that an imbalance between cell death and increased regenerative pressure through continuous mitosis is a critical component of fumonisin’s carcinogenic mode of action, especially in the kidney (Dragan et al., 2001; Howard et al., 2001a). In this regard, the survival and replication of damaged cells escaping apoptosis is likely critical for renal carcinogenesis. Sphingoid bases and other sphingolipids are signaling molecules for apoptosis and mitosis (Merrill et al., 2001; Taha et al., 2006). Specifically, sphingosine and ceramide are pro-apoptotic, growth inhibitory and cytotoxic whereas sphingosine-1-phosphate (S1P) exerts pro-growth and anti-apoptotic effects. Sphinganine causes apoptosis in LLC PK1 kidney cells in vitro (Kim et al., 2001; Riley et al., 1999; Yu et al., 2001), possibly through a calmodulin-dependent mechanism (Kim et al., 2001). The importance of elevated levels of free sphinganine in fumonisin-induced apoptosis has been demonstrated using myriocin, a potent inhibitor of serine palmitoyltransferase that abolishes the fumonisin-induced increase in free sphinganine and the increased apoptosis (Riley et al., 1999) Addition of exogenous sphinganine to the incubation medium also enhanced the toxicity of fumonisin to Chinese hamster ovary (CHO) cells, which are relatively resistant to the sphingolipid and pro-apoptotic effects of fumonisin (Yu et al., 2001). Thus, the overall metabolic balance of the various sphingolipids is likely a critical determinant of whether cells undergo apoptosis, remain quiescent, or replicate and, in this regard, the ratio of S1P to ceramide is likely to be important (Spiegel and Milstien, 2002). Sphingolipid metabolism is also affected by cytokines and other signaling molecules. Tumor necrosis factor a (TNFa) is a cytokine that
162
J. Gelineau-van Waes et al.
exerts both pro- or anti-apoptotic effects and has been shown to modify the extent of fumonisin-induced liver injury in mice (Sharma et al., 2002), perhaps through a mechanism that involves NF-kB (Gopee and Sharma, 2004). Enhancement of liver apoptosis (compared to wild-type mice) in TNFa knockout mice has been observed and was likely affected through over-expression of Fas ligand, indicating that Fas-dependent pathways also mediate fumonisin-induced apoptosis (Sharma et al., 2003). Activation of neutral sphingomyelinase by TNFa could also enhance apoptosis by increasing cell ceramide or sphingosine levels or, alternatively, promote cell survival in the event that significant amounts of sphingosine and, ultimately, S1P were formed from the sphingomyelinase-generated ceramide pool. Like TNFa and possibly Fas ligand, other cell surface receptor agonists influence cell survival and replication by activating sphingomyelinase, ceramidase or sphingosine kinases, thereby increasing the production of ceramide, sphingosine, and S1P, respectively (Merrill et al., 2001; Spiegel and Milstien, 2002).
C. Fumonisin inhibition of de novo sphingolipid metabolism Fumonisin exposure has been shown to result in subsequent alterations in the major pools of sphingolipids, including increased concentrations of free sphingoid bases and their 1-phosphate metabolites, and decreased biosynthesis of ceramide and downstream glycosphingolipids (Merrill et al., 2001; Riley et al., 1996) (Fig. 5.3). Sphinganine levels accumulate rapidly following inhibition of ceramide synthase, providing a biomarker for fumonisin exposure that has been validated in vitro, as well as in tissue, serum, and urine (Norred et al., 1997; Riley et al., 1993; Wang et al., 1991). The significantly elevated levels of sphinganine observed in LM/Bc maternal and fetal tissues following fumonisin exposure are indicative of fumonisin inhibition of the enzyme ceramide synthase. The marked increase in sphinganine in embryonic tissue following early gestational exposure to fumonisin (E7.5 and E8.5) suggests that the toxin crosses the placenta and that there is an effect at the level of the embryo that cannot simply be attributed to maternal toxicity. At this early time point, the placenta is not well developed, leaving the embryo potentially vulnerable to teratogenic insult. Fumonisin inhibition of ceramide synthase leads to the accumulation of free sphinganine and sphingosine in tissues, serum and urine in a wide range of species including catfish, trout, poultry cattle, horses, rabbits, and mice (Bolger et al., 2001; Merrill et al., 2001; Riley et al., 2001). Free sphingosine or sphinganine are subsequently phosphorylated by sphingosine kinase to form the corresponding sphingoid base-1-phosphates (S1P or sphinganine-1-P), which may then be dephosphorylated by either sphingosine phosphate phosphatases (Sgpp1, Sgpp2) or lipid phosphate
Fumonisin and Neural Tube Defects
163
phosphatases (Ppap2a, Ppap2b), or alternatively, irreversibly degraded by sphingosine phosphate lyase (Sgpl) to form phosphoethanolamine and hexadecenal (Fig. 5.4). Phosphoethanolamine is a precursor for the biosynthesis of phosphatidylethanolamine, a compound that has also been shown to accumulate following fumonisin exposure both in vitro and in vivo. Accumulation of bioactive S1P or sphinganine-1-P in serum or tissues of horse (Constable et al., 2005), swine (Piva et al., 2005), rats (Riley and Voss, 2006), adult mice (Suzuki et al., 2007), ducks (Tardieu et al., 2006), and livers of mouse embryos (Riley et al., 2006) has been reported after fumonisin exposure. The in vivo sphingolipid effects of fumonisins are reversible (Merrill et al., 2001). Within 3 weeks after replacing fumonisin-contaminated diet with control feed, the elevated tissue sphinganine (Sa) and sphingosine (So) concentrations and Sa/So ratios (used as a biomarker of fumonisin exposure) of rats decreased markedly, almost reaching pretest values, and the histological appearance of the liver and kidney returned (loss of fumonisin-linked lesions) to normal (Voss et al., 1998). In ponies, serum sphingoid base concentrations and serum chemistry indicators of hepatic injury (alanine and aspartate transaminase activities) rose and fell together as the fumonisin contaminated feed was given to and withdrawn from the animals (Wang et al., 1992).
TNFa, IFNg PDGF, VEGF Estrogen
Stored in RBCs, platelets
Sphingosine kinase 1 (sphk1) Sphingosine kinase 2 (sphk2) Sphingosine (sphinganine)
Sphingosine phosphate lyase Sphingosine-1-P (sphinganine-1-P)
Ethanolamine-P
Sphingosine phosphate phosphatases (Sgpp1, Sgpp2) Lipid phosphate phosphatases (Ppap2a, Ppap2b)
S1P establishes chemokine gradients important for cell migration
FIGURE 5.4 Biochemical pathway illustrating (1) phosphorylation of sphingosine or sphinganine by the enzymes sphingosine kinase 1 (sphk1) and/or sphingosine kinase 2 (sphk2); (2) dephosphorylation of sphingosine-1-phosphate (S1P) or sphinganine-1phosphate by sphingosine phosphate phosphatases (sgpp1, sgpp2) and/or lipid phosphate phosphatases (ppap2a, ppap2b); and (3) irreversible degradation of S1P or sphinganine-1-phosphate by sphingosine phosphate lyase (sgpl) to form phosphoethanolamine. Both S1P and sphinganine-1-phosphate are stored in red blood cells and platelets, and function as ligands for the G protein-coupled S1P receptors.
164
J. Gelineau-van Waes et al.
D. Fumonisin and accumulation of bioactive sphingoid base-1-phosphates Although sphingolipids are important structural components of the exoplasmic leaflet of the plasma membrane, their role in mediating cell signaling processes has also recently been recognized. Sphingoid base metabolites, such as sphingosine-1-phosphate (S1P) are involved in signal transduction cascades that regulate key biological processes (Futerman and Hannun, 2004; Spiegel and Milstien, 2002, 2003; Stunff et al., 2004; Watterson et al., 2003). S1P turnover is mediated either by reversible dephosphorylation to sphingosine or by irreversible cleavage to ethanolamine phosphate and fatty aldehyde (Zhou and Saba, 1998). The dynamic balance between S1P (pro-survival) and ceramide/sphingosine (proapoptosis) has been proposed to form a ‘‘cellular rheostat’’ that determines cell survival vs. cell death (Spiegel and Milstien, 2002). S1P has both extracellular and intracellular signaling functions, and functions as a ligand on cell membrane S1P receptors, a family of five G-protein-coupled receptors (S1P1–5; previously endothelial differentiation gene (Edg) receptors) that play important roles in such diverse functions as cell survival/ proliferation, differentiation and migration, as well as angiogenesis and endothelial barrier integrity (Brinkmann, 2007; Dev et al., 2008; Takabe et al., 2008) (Fig. 5.5). SIP gradients and S1P receptors also play an important role in mediating immune cell migration to sites of inflammation (Brinkmann, 2007; Cyster, 2005). The phenotypes observed in S1P receptor knockout mice indicate a certain degree of functional redundancy between the various receptor isoforms, but also establish a critical role for S1P receptors in angiogenesis and neurogenesis during embryonic development (Kono et al., 2004). Inactivation of S1P1 in mouse models results in embryolethality, and the offspring die between E12.5 and E13.5 due to severe hemorrhage. S1P1,2,3 are expressed in neural tissue (especially the telencephalon), S1P5 is expressed in CNS white matter (oligodendrocytes), and both sphingosine kinases (Sphk1, Sphk2) are expressed in the developing brain and spinal cord (Dev et al., 2008; Kono et al., 2004; Mizugishi et al., 2005), suggesting a role for S1P receptor-mediated signaling in neural tube closure. Although there is no obvious phenotype when either Sphk1 or Sphk2 alone is inactivated in a knockout mouse model, Sphk1/Sphk2 double knockout embryos die by E12.5 with severe vascular and neural abnormalities, including failure of neural tube closure (Mizugishi et al., 2005). Interestingly, Sphk1-/-/Sphk2 þ/ mutant mice are infertile, and follow-up experiments suggest a critical role for early gestational activation of de novo sphingolipid metabolism in the maintenance of pregnancy (Mizugishi et al., 2007). Sphinganine-1-P also functions as a ligand for S1P receptors, although the binding affinity for the different S1P receptor isoforms differs from
Fumonisin and Neural Tube Defects
S1P1
S1P2
S1P3
S1P4
165
S1P5
Gi/o G12/13
Gq AC
c-Src
Pi3K
Rho PLC Ca2+
cAMP
ERK
Rac Cdc 42
Akt PKC
Rock, PLD cadherin
eNOS Cell proliferation Adhesion, extension cell migration
Cell shape changes
Detachment, retraction cell Migration
Cell cycle regulation
FIGURE 5.5 There are five isoforms of S1P receptors (S1P1–5), a family of G proteincoupled membrane receptors previously known as ‘‘Edg’’ receptors (endothelial differentiation gene). Ligand binding and activation of S1P receptors initiates multiple intracellular signaling cascades involved in cellular proliferation, differentiation, survival and migration.
that of S1P (Im et al., 2001). The high levels of sphinganine and sphinganine-1-P that accumulate after fumonisin exposure may therefore elicit different downstream signaling cascades than S1P due to differential binding and activation of the various S1P receptor isoforms. Moreover, high concentrations of ligand have been shown to cause internalization and downregulation of S1P cell surface receptors, thereby acting as a ‘‘functional antagonist’’ rather than as an agonist (Brinkmann, 2007). Further differences in downstream signaling effects include a role for S1P in intracellular signaling, which apparently is not observed for sphinganine-1-phosphate (Taha et al., 2006). The activation of S1P receptors reduces renal and mesenteric blood flow in rats (Bischoff et al., 2001), and it can be speculated that decreased blood flow and hypoxia contribute to the extreme sensitivity of rat kidneys to fumonisins. Fumonisin exposure has also been shown to decrease cardiac output, heart rate, and mean arterial pressure as well as increase pulmonary arterial pressure and pulmonary resistance in pigs (Constable et al., 2000, 2003), findings which suggest that left-sided cardiac insufficiency underlies porcine pulmonary edema. Plasma sphingoid base
166
J. Gelineau-van Waes et al.
concentrations (Constable et al., 2003) and serum S1P (and sphinganine-1-P) (Piva et al., 2005) increase in fumonisin-exposed pigs. Sphingosine blocks L-type calcium channels and ryanodine receptors in rabbits (Sabbadini et al., 1992). Therefore, one possible mechanism for fumonisin-induced porcine pulmonary edema involves interference with calcium ion transport, which in turn is critical for cardiac function and vascular tone (Constable et al., 2003). Both sphinganine and sphingosine relaxed phenylephrin-contracted and uncontracted thoracic aortic and pulmonary arterial rings in vitro (Hsiao et al., 2005) and, while S1P had little effect on the aortic rings, it increased the tension of uncontracted pulmonary arterial rings. From these observations, it can be further speculated that plasma S1P also contributes to pulmonary edema by increasing pulmonary arterial tension, likely through a mechanism involving S1P receptors. S1P1,2,3 receptors are known to mediate endothelial barrier integrity and vascular permeability (Brinkmann, 2007; Gon et al., 2005; Sanchez et al., 2007; Singleton et al., 2005), while agonism of S1P3 receptors results in sinus bradycardia (Sanna et al., 2004). The process of neural tube closure requires the integration of multiple signaling cascades involved in cell proliferation, migration, and differentiation. It is therefore likely that fumonisin inhibition of ceramide synthase, resulting in elevated levels of sphinganine-1-phosphate (as opposed to tightly regulated levels of S1P) may alter the dynamics of S1P receptor-mediated signaling pathways involved in coordinating these events, thereby contributing to the failure of neural tube closure.
E. Fumonisin depletion of glycosphingolipids and disruption of folate transport In addition to an accumulation of upstream sphingoid bases and sphingoid base-1-phosphates, depletion of the downstream, complex sphingolipid metabolites of ceramide (i.e., gangliosides) has also been shown to occur after fumonisin exposure (Fig. 5.3). Stevens and Tang (1997) reported in an in vitro study that fumonisin-induced depletion of glycosphingolipids inhibited vitamin uptake via the GPI-anchored folate receptor and suggested that dietary exposure to fumonisin could therefore adversely affect folate uptake, and potentially compromise cellular processes dependent on this vitamin. The high-affinity folate receptor (murine Folr1; human FRa) is a GPI-anchored protein found associated with detergent-insoluble membrane microdomains rich in cholesterol and sphingolipids known as ‘‘lipid rafts’’ (Elortza et al., 2003). These rafts create domains of reduced fluidity in membranes and arise from the shared biophysical properties of sphingolipids and cholesterol (Brown and London, 2000; Rietveld and Simons, 1998). They are hypothesized to function in membrane and endocytic sorting of proteins in polarized
Fumonisin and Neural Tube Defects
167
epithelial cells, as foci for recruiting signaling molecules to the plasma membrane, and in cell adhesion (Ilangumaran et al., 1999; Lin et al., 1998; Simons and Ikonen, 1997). A role for membrane rafts in regulating the folate receptor was suggested by studies in which the depletion of cellular cholesterol was found to impair the uptake of 5-methyltetrahydrofolate into cells (Chang et al., 1992). Gangliosides stabilize the association of GPI-anchored proteins with the outer leaflet of cell membranes (Watanabe et al., 2002) and are important components of specialized membrane microdomains, or ‘‘lipid rafts’’ in which GPI-anchored proteins cluster. Ganglioside GM1 has been shown to be highly expressed in the developing brain and spinal cord (Silani et al., 1993), and decreased levels of ganglioside GM1a and GD1a have previously been reported in anencephalic human fetal brain (Cacic, 1995). In the LM/Bc in vivo mouse model of fumonisin exposure, maternal supplementation with ganglioside GM1 was more effective than folate in preventing NTDs. Expression of ganglioside GM1 in plasma membrane microdomains of the developing neuroepithelium may be necessary for coordinating/orchestrating protein–lipid interactions and signal transduction events involved in normal neural tube closure. GPI-anchored proteins such as the folate receptor associate with gangliosides in the trans-golgi network, prior to sorting and transport to the plasma membrane. Association of Folr1 with ganglioside GM1 may be important for delivery of the GPI-anchored folate receptor to the apical plasma membrane in neuroepithelial cells. In addition, co-localization of Folr1 with GM1 in lipid rafts appears to facilitate receptor function, and promote normal neural tube closure. Different gangliosides associate with different proteins in ‘‘lipid rafts’’ in a cell-type specific manner, and the rafts function as specialized platforms for the initiation of signal transduction (Brown and London, 1998; Hoessli et al., 2000). Although the specific interactions between ganglioside GM1 and Folr1 and their precise role in neural tube closure are unknown at this time, several hypotheses are currently under investigation. Since the GPI-anchored folate receptor has previously been isolated from lipid raft microdomains enriched in signaling molecules, including kinases/phosphatases, heterotrimeric G proteins, and small GTP-binding proteins (Elortza et al., 2003; Foster et al., 2003; Miotti et al., 2000), it seems reasonable to hypothesize that Folr1 may play a critical role in signal transduction pathways necessary for cell survival, proliferation, and activation of the actin cytoskeleton necessary for normal neural tube closure. Fumonisin inhibition of de novo sphingolipid biosynthesis (Merrill et al., 2001; Riley et al., 1996), resulting in depletion of complex glycosphingolipids, disruption of lipid rafts, and compromised folate transport via Folr1, which may have significant implications with respect to developmental anomalies such as NTDs (Hansen et al., 2003; Piedrahita
168
J. Gelineau-van Waes et al.
et al., 1999; Rothenberg et al., 2004; Saitsu et al., 2003). The inhibition of folate uptake by cells is expected to have the same consequences as a dietary deficiency in this vitamin. Therefore, the possibility that inhibition of folate receptor-mediated vitamin uptake contributes to the development of NTDs has been investigated. Mice in which expression of the gene encoding the folate receptor (Folr1) was knocked out were found to have defects in neural tube development and die before birth (Piedrahita et al., 1999). This phenotype could be reversed by supplementing the pregnant knockout mice with excess folate, suggesting that the loss of vitamin uptake mediated by the folate receptor was responsible for the developmental defects (Piedrahita et al., 1999). Reduction of folate receptor expression in mouse embryos with antisense technology (Hansen et al., 2003) or blocking Folr1 activity in pregnant rats with antibodies to this transporter (da Costa et al., 2003) both resulted in impaired embryogenesis that could be reversed with excess folate supplementation. Rothenberg et al. (2004) found autoantibodies against this transporter in serum from a subset of women with neural tube defect-affected pregnancies, suggesting that antibody-induced inactivation of the folate receptor may be responsible for some NTDs in humans. Cumulatively, these findings indicate that impairment of folate receptor function could result in NTDs. Fumonisin B1-induced depletion of the major sphingolipids to 40–65% of their normal levels in Caco-2 cells (which express high levels of the folate receptor) was found to significantly impair 5-methyltetrahydrofolate uptake (Stevens and Tang, 1997). While this result demonstrates that sphingolipid depletion inhibits vitamin uptake, it does not indicate a specific effect on the folate receptor because 5-methyltetrahydrofolate can be taken up by both the RFC1 and the GPI-anchored folate receptor (Folr1) (Corona et al., 1998; Miotti et al., 1997). Therefore, to specifically determine the effects of fumonisin B1 treatment on the GPI-anchored folate receptor, uptake was measured with folic acid. This synthetic folate is bound by the folate receptor (Folr1) with an affinity that is approximately 100,000 times greater than that of the RFC1, and folic acid is transported almost exclusively by the former (Goldman, 1971; Kamen and Capdevila, 1986). Folate receptor function was found to be inhibited by 20–35% in cells derived from a human placental choriocarcinoma ( JAR), monkey kidney epithelium (MA-104), and a human adenocarcinoma (Caco-2) that had been depleted of roughly 40% of their sphingolipids by fumonisin B1 treatment. Other uptake processes, including receptor-mediated and facilitative transport, were not affected by this treatment (Stevens and Tang, 1997). Thus, depletion of cellular sphingolipids specifically inhibits folate receptor-mediated vitamin uptake. Folate receptor function can be altered by changing the level of this protein, its activity, or both. To determine if folate receptor levels were
Fumonisin and Neural Tube Defects
169
affected by sphingolipid depletion, the amount of this transporter was measured in fumonisin B1-treated cells. Total cellular folate receptor was measured by quantifying specific folic acid binding in detergent solubilized cells while the amount of receptor active in endocytosis and available for folate uptake was measured in intact cells under conditions where endocytosis was maintained. Fumonisin B1-induced sphingolipid depletion did not change total cellular levels of folate receptor significantly in Caco-2, JAR and MA-104 cells. However, this treatment did decrease the amount of receptor active in endocytosis by between 25 and 40%. This decrease in the amount of folate receptor available for vitamin uptake could contribute to the fumonisin B1-induced inhibition of this transporter. The effect of fumonisin B1 treatment on folate receptor function was investigated by determining if sphingolipid depletion altered the endocytic kinetics of this transporter. These experiments were done using FRaTb-1 cells, which were generated by stably transfecting CHO cells with the GPI-anchored folate receptor and the transferrin receptor (Mayor et al., 1998). The movement of the folate receptor through the endocytic pathway was followed by using a fluorescent folic acid analog that remained tightly bound to this protein. Fumonisin B1-treatment did not significantly alter the rate at which the folate receptor was internalized. However, the recycling of this receptor to the cell surface was accelerated approximately threefold by sphingolipid depletion (Chatterjee et al., 2001). The recycling rate was also faster in cells depleted of cholesterol through treatment with the HMG-CoA reductase inhibitor compactin (Chatterjee et al., 2001). The fact that reduction of either of the major lipid species found in membrane rafts specifically perturbed the recycling of the folate receptor suggests that association with these domains regulates the intracellular sorting and trafficking of this GPI-anchored protein. Thus, the inhibition of folate receptor-mediated vitamin uptake caused by fumonisin B1-induced sphingolipid depletion appears to result from a combination of altered endocytic trafficking and reduced levels of this receptor available for folate uptake. The molecular mechanisms by which sphingolipid depletion causes these changes are unknown. In summary, fumonisin mycotoxins exert a wide range of diverse and species-specific toxicological effects. The weight of evidence suggests that inhibition of ceramide synthase and disruption of sphingolipid metabolism is the initial mechanistic event. The ensuing disruption of the metabolism of the sphingoid bases, sphingoid base 1-phosphates, and complex glycosphingolipids and the physiological processes modulated by these molecules are critical for fumonisin toxicity.
170
J. Gelineau-van Waes et al.
VI. CONCLUSIONS NTDs present a tremendous burden to human populations in rural areas of the world where maize is a dietary staple (Hendricks, 1999; Kromberg and Jenkins, 1982; Marasas et al., 2004; Melnick and Marazita, 1998; Moore et al., 1997; Ncayiyana, 1986; Xiao et al., 1990). The observed association between consumption of fumonisin-contaminated maize and increased incidence of NTDs in several world communities, coupled with the existing experimental evidence in animal models strongly supports the need for further investigation into the teratogenic potential of this compound. Epidemiologic data on dietary fumonisin exposures, including levels in foods, and daily food consumption, coupled with measurements of reliable biomarkers of exposure are needed. These exposure assessments, along with information on maternal genetics and nutrition (i.e., maternal folate status) as they relate to pregnancy outcome will help to establish a comprehensive understanding of the impact of fumonisin as a human health hazard, and risk factor for birth defects. Animal feeding trials incorporating dietary administration of the mycotoxin will help to establish relevant ‘‘no observable adverse effect levels’’ (NOAEL), and mechanistic studies in animal models will help to identify and validate biomarkers of exposure, and establish the complex biochemical, genetic, immune, and nutritional factors that contribute to increased susceptibility to fumonisin-induced NTDs. Establishing the basic mechanisms underlying fumonisin toxicity and identifying polymorphisms in candidate genes that contribute to susceptibility will facilitate our understanding of neural tube defect risk in women exposed to this mycotoxin during early pregnancy. Collectively, this information will also help in developing strategies for prevention. The experimental evidence suggests that maternal folate supplementation is effective in ameliorating the teratogenic effects of fumonisin. Maternal diet and nutritional status play a key role in fetal development, and strategies such as folate supplementation or folate fortification of foods may prove effective in reducing the incidence of NTDs in communities likely to be consuming high levels of fumonisincontaminated maize. Fumonisins are present at low levels in maize throughout the world, but high levels of the toxin may occur, depending on the environmental conditions and genetics of the host plant. Strategies for reducing the risk of fumonisin contamination in maize supplied to the market include improved crop management practices, and improved breeding strategies, as well as transgenic approaches for increased earmold resistance (Betz et al., 2000; Duvick, 2001). Minimizing exposures to mycotoxins through enhanced agricultural practices and biotechnology, characterizing potential health risks and mechanisms of toxicity, and improving maternal nutrition are all important strategies or considerations for reducing the neural tube defect burden in high risk human populations.
Fumonisin and Neural Tube Defects
171
REFERENCES Abel, S. and Gelderblom, W. C. (1998). Oxidative damage and fumonisin B1-induced toxicity in primary rat hepatocytes and rat liver in vivo. Toxicology 131(2–3), 121–131. Barber, R. C., Bennett, G. D., Greer, K. A., and Finnell, R. H. (1999). Expression patterns of folate binding proteins one and two in the developing mouse embryo. Mol. Genet. Metab. 66(1), 31–39. Ben Othmane, K., Rochelle, J. M., Ben Hamida, M., Slotterbeck, B., Rao, N., Hentati, F., Pericak-Vance, M. A., and Vance, J. M. (1998). Fine localization of the CMT4A locus using a PAC contig and haplotype analysis. Neurogenetics 2(1), 18–23. Betz, F. S., Hammond, B. G., and Fuchs, R. L. (2000). Safety and advantages of Bacillus thuringiensis-protected plants to control insect pests. Regul. Toxicol. Pharmacol. 32(2), 156–173. Bhandari, N., Enongene, E. N., Riley, R. T., Meredith, F. I., and Sharma, R. P. (2002). Temporal expression of fumonisin B(1)-induced tumor necrosis factor-alpha and interferon gamma in mice. Comp. Biochem. Physiol. C Toxicol. Pharmacol. 131(2), 113–122. Bischoff, A., Finger, J., and Michel, M. C. (2001). Nifedipine inhibits sphinogosine-1-phosphate-induced renovascular contraction in vitro and in vivo. Naunyn. Schmiedebergs Arch. Pharmacol. 364(2), 179–182. Bolger, M. C. R. D., Dinovi, M., Gaylor, D., Gelderblom, W. C., Olsen, M., Paster, N., Riley, R. T., Shephard, G., and Spijers, G. J. A. (2001). ‘‘Fumonisins. Safety evaluation of certain mycotoxin in food. International Programme on Chemical Safetty World Health Organization.’’ Vol. 47, pp. 103–279. FAO FOOD and Nutrition Paper 74, WHO – Food Additives Series, Geneva. Bondy, G. S., Barker, M. G., Lombaert, G. A., Armstrong, C. L., Fernie, S. M., Gurofsky, S., Huzel, V., Savard, M. E., and Curran, I. H. (2000). A comparison of clinical, histopathological and cell-cycle markers in rats receiving the fungal toxins fumonisin B1 or fumonisin B2 by intraperitoneal injection. Food Chem. Toxicol. 38(10), 873–886. Brinkmann, V. (2007). Sphingosine 1-phosphate receptors in health and disease: Mechanistic insights from gene deletion studies and reverse pharmacology. Pharmacol. Ther. 115(1), 84–105. Broman, K. W., Murray, J. C., Sheffield, V. C., White, R. L., and Weber, J. L. (1998). Comprehensive human genetic maps: Individual and sex-specific variation in recombination. Am. J. Hum. Genet. 63(3), 861–869. Brown, D. A. and London, E. (1998). Functions of lipid rafts in biological membranes. Annu. Rev. Cell Dev. Biol. 14, 111–136. Brown, D. A. and London, E. (2000). Structure and function of sphingolipid- and cholesterolrich membrane rafts. J. Biol. Chem. 275(23), 17221–17224. Burton, K. E., Steele, F. M., Jefferies, L., Pike, O. A., and Dunn, M. L. (2008). Effect of micronutrient fortification on nutritional and other properties of nixtamal tortillas. Cereal Chem. 85(1), 70–75. Cacic, M. (1995). Gangliosides of anencephalic and fetal brain – immunostaining on thinlayer chromatograms. Neuroreport 6(2), 389–393. Campbell, L. R., Dayton, D. H., and Sohal, G. S. (1986). Neural tube defects: A review of human and animal studies on the etiology of neural tube defects. Teratology 34(2), 171–187. Ca´rdenas, J. D. F., Godinez, M. G. A., Me´ndez, N. L. V., Guzma´n, A. L., Acosta, L. M. F., and Gonza´lez-Herna´ndez, J. (2001). Fortificacion y evaluacion de tortillas de nixtamal. Arch. Latinoam. Nutr. 51(3), 293–302. Castoria, R., Lima, G., Ferracane, R., and Ritieni, A. (2005). Occurrence of mycotoxin in Farro samples from southern Italy. J. Food Prot. 68(2), 416–420.
172
J. Gelineau-van Waes et al.
Chalfant, C. E. and Spiegel, S. (2005). Sphingosine 1-phosphate and ceramide 1-phosphate: Expanding roles in cell signaling. J. Cell Sci. 118(Pt 20), 4605–4612. Chang, W. J., Rothberg, K. G., Kamen, B. A., and Anderson, R. G. (1992). Lowering the cholesterol content of MA104 cells inhibits receptor-mediated transport of folate. J. Cell Biol. 118(1), 63–69. Chatkupt, S., Skurnick, J. H., Jaggi, M., Mitruka, K., Koenigsberger, M. R., and Johnson, W. G. (1994). Study of genetics, epidemiology, and vitamin usage in familial spina bifida in the United States in the 1990s. Neurology 44(1), 65–70. Chatterjee, S., Smith, E. R., Hanada, K., Stevens, V. L., and Mayor, S. (2001). GPI anchoring leads to sphingolipid-dependent retention of endocytosed proteins in the recycling endosomal compartment. EMBO J. 20(7), 1583–1592. Chu, F. S. and Li, G. Y. (1994). Simultaneous occurrence of fumonisin B1 and other mycotoxins in moldy corn collected from the People’s Republic of China in regions with high incidences of esophageal cancer. Appl. Environ. Microbiol. 60(3), 847–852. Cifuentes, G. (2002). ‘‘Perfil epidemiologico de law anomalias del tubo neural en Guatemala, durante el ano 2000.’’ Graduation Thesis of the School of Medicine of Universidad San Carlos de Guatemala. Collins, T. F., Shackelford, M. E., Sprando, R. L., Black, T. N., Laborde, J. G., Hansen, D. K., Eppley, R. M., Trucksess, M. W., Howard, P. C., Bryant, M. A., Ruggles, D. I., Olejnik, N., et al. (1998). Effects of fumonisin B1 in pregnant rats. Food Chem. Toxicol. 36(5), 397–408. Commission of European Communities (2005). Commission Regulation (EC) No. 856/2005 of 6 June 2005 amending Regulation (EC) No 466/2001 as regards Fusarium toxins: pp. 3–8. Constable, P. D., Smith, G. W., Rottinghaus, G. E., and Haschek, W. M. (2000). Ingestion of fumonisin B1-containing culture material decreases cardiac contractility and mechanical efficiency in swine. Toxicol. Appl. Pharmacol. 162(3), 151–160. Constable, P. D., Smith, G. W., Rottinghaus, G. E., Tumbleson, M. E., and Haschek, W. M. (2003). Fumonisin-induced blockade of ceramide synthase in sphingolipid biosynthetic pathway alters aortic input impedance spectrum of pigs. Am. J. Physiol. Heart Circ. Physiol. 284(6), H2034–2044. Constable, P. D., Riley, R. T., Waggoner, A. L., Hsiao, S. H., Foreman, J. H., Tumbleson, M. E., and Haschek, W. M. (2005). ‘‘Serum sphingosine–1-phosphate and sphinganine-1-phosphate are elevated in horses exposed to fumonisin B1.’’ AOAC International Midwest Section Final Program, pp. 63–64. Copp, A. J., Greene, N. D., and Murdoch, J. N. (2003). The genetic basis of mammalian neurulation. Nat. Rev. Genet. 4(10), 784–793. Corona, G., Giannini, F., Fabris, M., Toffoli, G., and Boiocchi, M. (1998). Role of folate receptor and reduced folate carrier in the transport of 5-methyltetrahydrofolic acid in human ovarian carcinoma cells. Int. J. Cancer 75(1), 125–133. Cyster, J. G. (2005). Chemokines, sphingosine-1-phosphate, and cell migration in secondary lymphoid organs. Annu. Rev. Immunol. 23, 127–159. Czeizel, A. E. and Dudas, I. (1992). Prevention of the first occurrence of neural-tube defects by periconceptional vitamin supplementation. N. Engl. J. Med. 327(26), 1832–1835. da Costa, M., Sequeira, J. M., Rothenberg, S. P., and Weedon, J. (2003). Antibodies to folate receptors impair embryogenesis and fetal development in the rat. Birth Defects Res. A Clin. Mol. Teratol. 67(10), 837–847. da Silva, J. B., Pozzi, C. R., Mallozzi, M. A., Ortega, E. M., and Correa, B. (2000). Mycoflora and occurrence of aflatoxin B(1) and fumonisin B(1) during storage of Brazilian sorghum. J. Agric. Food Chem. 48(9), 4352–4356. Dev, K. K., Mullershausen, F., Mattes, H., Kuhn, R. R., Bilbe, G., Hoyer, D., and Mir, A. (2008). Brain sphingosine-1-phosphate receptors: Implication for FTY720 in the treatment of multiple sclerosis. Pharmacol. Ther. 117(1), 77–93.
Fumonisin and Neural Tube Defects
173
Dombrink-Kurtzman, M. A., Dvorak, T. J., Barron, M. E., and Rooney, L. W. (2000). Effect of nixtamalization (Alkaline cooking) on fumonisin-contaminated corn for production of masa and tortillas. J. Agric. Food. Chem. 48(11), 5781–5786. Dragan, Y. P., Bidlack, W. R., Cohen, S. M., Goldsworthy, T. L., Hard, G. C., Howard, P. C., Riley, R. T., and Voss, K. A. (2001). Implications of apoptosis for toxicity, carcinogenicity, and risk assessment: Fumonisin B(1) as an example. Toxicol. Sci. 61(1), 6–17. Duvick, J. (2001). Prospects for reducing fumonisin contamination of maize through genetic modification. Environ. Health Perspect. 109(Suppl 2), 337–342. Elortza, F., Nuhse, T. S., Foster, L. J., Stensballe, A., Peck, S. C., and Jensen, O. N. (2003). Proteomic analysis of glycosylphosphatidylinositol-anchored membrane proteins. Mol. Cell Proteomics 2(12), 1261–1270. Enongene, E. N., Sharma, R. P., Bhandari, N., Miller, J. D., Meredith, F. I., Voss, K. A., and Riley, R. T. (2002). Persistence and reversibility of the elevation in free sphingoid bases induced by fumonisin inhibition of ceramide synthase. Toxicol. Sci. 67, 173–181. Floss, J. L., Casteel, S. W., Johnson, G. C., Rottinghaus, G. E., and Krause, G. F. (1994). Development toxicity of fumonisin in Syrian hamsters. Mycopathologia 128(1), 33–38. Flynn, T. J., Stack, M. E., Troy, A. L., and Chirtel, S. J. (1997). Assessment of the embryotoxic potential of the total hydrolysis product of fumonisin B1 using cultured organogenesisstaged rat embryos. Food Chem. Toxicol. 35(12), 1135–1141. Foster, L. J., De Hoog, C. L., and Mann, M. (2003). Unbiased quantitative proteomics of lipid rafts reveals high specificity for signaling factors. Proc. Natl. Acad. Sci. USA 100(10), 5813–5818. Futerman, A. H. and Hannun, Y. A. (2004). The complex life of simple sphingolipids. EMBO Rep. 5(8), 777–782. Gelderblom, W. C., Jaskiewicz, K., Marasas, W. F., Thiel, P. G., Horak, R. M., Vleggaar, R., and Kriek, N. P. (1988). Fumonisins – Novel mycotoxins with cancer-promoting activity produced by Fusarium moniliforme. Appl. Environ. Microbiol. 54(7), 1806–1811. Gelderblom, W. C., Kriek, N. P., Marasas, W. F., and Thiel, P. G. (1991). Toxicity and carcinogenicity of the Fusarium moniliforme metabolite, fumonisin B1, in rats. Carcinogenesis 12(7), 1247–1251. Gelderblom, W. C., Seier, J. V., Snijman, P. W., van Schalkwyk, D. J., Shephard, G. S., and Marasas, W. F. (2001). Toxicity of culture material of Fusarium verticillioides strain MRC 826 to nonhuman primates. Environ. Health Perspect. 109(Suppl 2), 267–276. Gelineau-van Waes, J., Starr, L., Maddox, J., Aleman, F., Voss, K. A., Wilberding, J., and Riley, R. T. (2005). Maternal fumonisin exposure and risk for neural tube defects: Mechanisms in an in vivo mouse model. Birth Defects Res. A Clin. Mol. Teratol. 73(7), 487–497. Goldman, I. D. (1971). The characteristics of the membrane transport of amethopterin and the naturally occurring folates. Ann. NY Acad. Sci. 186, 400–422. Gon, Y., Wood, M. R., Kiosses, W. B., Jo, E., Sanna, M. G., Chun, J., and Rosen, H. (2005). S1P3 receptor-induced reorganization of epithelial tight junctions compromises lung barrier integrity and is potentiated by TNF. Proc. Natl. Acad. Sci. USA 102(26), 9270–9275. Gong, Y. Y., Sanchez, L. T., Carrillo, L. L., Peng, J. H., Sutcliffe, A. E., White, K. L., Humpf, H. U., Turner, P. C., and Wild, C. P. (2008). Association between tortilla consumption and human urinary fumonisin B1 levels in a Mexican population. Cancer Epidemiol. Biomarkers Prev. 17(3), 688–694. Gopee, N. V. and Sharma, R. P. (2004). The mycotoxin fumonisin B1 transiently activates nuclear factor-kappaB, tumor necrosis factor alpha and caspase 3 via protein kinase C alpha-dependent pathway in porcine renal epithelial cells. Cell Biol. Toxicol. 20(4), 197–212. Gross, S. M., Reddy, R. V., Rottinghaus, G. E., Johnson, G., and Reddy, C. S. (1994). Developmental effects of fumonisin B1-containing Fusarium moniliforme culture extract in CD1 mice. Mycopathologia 128(2), 111–118.
174
J. Gelineau-van Waes et al.
Gudbjartsson, D. F., Jonasson, K., Frigge, M. L., and Kong, A. (2000). Allegro, a new computer program for multipoint linkage analysis. Nat. Genet. 25(1), 12–13. Hansen, D. K., Streck, R. D., and Antony, A. C. (2003). Antisense modulation of the coding or regulatory sequence of the folate receptor (folate binding protein-1) in mouse embryos leads to neural tube defects. Birth Defects Res. Part A Clin. Mol. Teratol. 67(7), 475–487. Hard, G. C., Howard, P. C., Kovatch, R. M., and Bucci, T. J. (2001). Rat kidney pathology induced by chronic exposure to fumonisin B1 includes rare variants of renal tubule tumor. Toxicol. Pathol. 29(3), 379–386. Haschek, W. M., Gumprecht, L. A., Smith, G., Tumbleson, M. E., and Constable, P. D. (2001). Fumonisin toxicosis in swine: An overview of porcine pulmonary edema and current perspectives. Environ. Health Perspect. 109(Suppl 2), 251–257. Hendricks, K. (1999). Fumonisins and neural tube defects in South Texas. Epidemiology 10(2), 198–200. Hoessli, D. C., Ilangumaran, S., Soltermann, A., Robinson, P. J., Borisch, B., and Nasir Ud, D. (2000). Signaling through sphingolipid microdomains of the plasma membrane: The concept of signaling platform. Glycoconj J. 17(3–4), 191–197. Howard, P. C., Warbritton, A., Voss, K. A., Lorentzen, R. J., Thurman, J. D., Kovach, R. M., and Bucci, T. J. (2001a). Compensatory regeneration as a mechanism for renal tubule carcinogenesis of fumonisin B1 in the F344/N/Nctr BR rat. Environ. Health Perspect 109(Suppl 2), 309–314. Howard, P. C., Eppley, R. M., Stack, M. E., Warbritton, A., Voss, K. A., Lorentzen, R. J., Kovach, R. M., and Bucci, T. J. (2001b). Fumonisin b1 carcinogenicity in a two-year feeding study using F344 rats and B6C3F1 mice. Environ. Health Perspect 109(Suppl 2), 277–282. Howard, P. C., Couch, L. H., Patton, R. E., Eppley, R. M., Doerge, D. R., Churchwell, M. I., Marques, M. M., and Okerberg, C. V. (2002). Comparison of the toxicity of several fumonisin derivatives in a 28-day feeding study with female B6C3F(1) mice. Toxicol. Appl. Pharmacol. 185(3), 153–165. Hsiao, S. H., Constable, P. D., Smith, G. W., and Haschek, W. M. (2005). Effects of exogenous sphinganine, sphingosine, and sphingosine-1-phosphate on relaxation and contraction of porcine thoracic aortic and pulmonary arterial rings. Toxicol. Sci. 86(1), 194–199. Humpf, H. U. and Voss, K. A. (2004). Effects of thermal food processing on the chemical structure and toxicity of fumonisin mycotoxins. Mol. Nutr. Food Res. 48(4), 255–269. IARC (2002). ‘‘Some Traditional Herbal Medicines, Some Mycotoxins, Naphthalene and Styrene,’’ IARC Working Group on the Evaluation of Carcinogenic Risk to Humans, IARC Press, Lyon. Ilangumaran, S., Arni, S., van Echten-Deckert, G., Borisch, B., and Hoessli, D. C. (1999). Microdomain-dependent regulation of Lck and Fyn protein-tyrosine kinases in T lymphocyte plasma membranes. Mol. Biol. Cell 10(4), 891–905. Im, D. S., Clemens, J., Macdonald, T. L., and Lynch, K. R. (2001). Characterization of the human and mouse sphingosine 1-phosphate receptor, S1P5 (Edg-8): Structure-activity relationship of sphingosine1-phosphate receptors. Biochemistry 40(46), 14053–14060. International Agency for Research on Cancer (2002). ‘‘IARC Monographs on the Evaluation of Carcinogenic Risks to Humans,’’ pp. 301–366, IARC Press, Lyon. Juriloff, D. M. and Harris, M. J. (2000). Mouse models for neural tube closure defects. Hum. Mol. Genet. 9(6), 993–1000. Kamen, B. A. and Capdevila, A. (1986). Receptor-mediated folate accumulation is regulated by the cellular folate content. Proc. Natl. Acad. Sci. 83(16), 5983–5987. Khoury, M. J., Beaty, T. H., and Liang, K. Y. (1988). Can familial aggregation of disease be explained by familial aggregation of environmental risk factors? Am. J. Epidemiol. 127(3), 674–683. Kim, E. K., Scott, P. M., and Lau, B. P. (2003). Hidden fumonisin in corn flakes. Food Addit. Contam. 20(2), 161–169.
Fumonisin and Neural Tube Defects
175
Kim, M. S., Lee, D. Y., Wang, T., and Schroeder, J. J. (2001). Fumonisin B(1) induces apoptosis in LLC-PK(1) renal epithelial cells via a sphinganine- and calmodulin-dependent pathway. Toxicol. Appl. Pharmacol. 176(2), 118–126. Kintz, P. (2004). Value of hair analysis in postmortem toxicology. Forensic Sci. Int. 142(2–3), 127–134. Kong, A. and Cox, N. J. (1997). Allele-sharing models: LOD scores and accurate linkage tests. Am. J. Hum. Genet. 61(5), 1179–1188. Kono, M., Mi, Y., Liu, Y., Sasaki, T., Allende, M. L., Wu, Y. P., Yamashita, T., and Proia, R. L. (2004). The sphingosine-1-phosphate receptors S1P1, S1P2, and S1P3 function coordinately during embryonic angiogenesis. J. Biol. Chem. 279(28), 29367–29373. Kouadio, J. H., Mobio, T. A., Baudrimont, I., Moukha, S., Dano, S. D., and Creppy, E. E. (2005). Comparative study of cytotoxicity and oxidative stress induced by deoxynivalenol, zearalenone or fumonisin B1 in human intestinal cell line Caco-2. Toxicology 213(1–2), 56–65. Krapels, I. P., Rooij, I. A., Wevers, R. A., Zielhuis, G. A., Spauwen, P. H., Brussel, W., and Steegers-Theunissen, R. P. (2004). Myo-inositol, glucose and zinc status as risk factors for non-syndromic cleft lip with or without cleft palate in offspring: A case-control study. BJOG 111(7), 661–668. Kritzinger, Q., Aveling, T. A., Marasas, W. F., Rheeder, J. P., van der Westhuizen, L., and Shephard, G. S. (2003). Mycoflora and fumonisin mycotoxins associated with cowpea [Vigna unguiculata (L.) Walp] seeds. J. Agric. Food Chem. 51(8), 2188–2192. Kromberg, J. G. and Jenkins, T. (1982). Common birth defects in South African Blacks. S. Afr. Med. J. 62(17), 599–602. Kruglyak, L., Daly, M. J., Reeve-Daly, M. P., and Lander, E. S. (1996). Parametric and nonparametric linkage analysis: A unified multipoint approach. Am. J. Hum. Genet. 58(6), 1347–1363. Leck, I. (1983). Spina bifida and anencephaly: Fewer patients, more problems. Br. Med. J. (Clin. Res. Ed.) 286(6379), 1679–1680. Lemke, S. L., Ottinger, S. E., Ake, C. L., Mayura, K., and Phillips, T. D. (2001). Deamination of fumonisin B(1) and biological assessment of reaction product toxicity. Chem. Res. Toxicol. 14(1), 11–15. Lemmer, E. R., Gelderblom, W. C., Shephard, E. G., Abel, S., Seymour, B. L., Cruse, J. P., Kirsch, R. E., Marasas, W. F., and Hall, P. M. (1999). The effects of dietary iron overload on fumonisin B1-induced cancer promotion in the rat liver. Cancer Lett. 146(2), 207–215. Lin, S., Naim, H. Y., Rodriguez, A. C., and Roth, M. G. (1998). Mutations in the middle of the transmembrane domain reverse the polarity of transport of the influenza virus hemagglutinin in MDCK epithelial cells. J. Cell Biol. 142(1), 51–57. Lu, Y., Clifford, L., Hauck, C. C., Hendrich, S., Osweiler, G., and Murphy, P. A. (2002). Characterization of fumonisin B(1)-glucose reaction kinetics and products. J. Agric. Food Chem. 50(16), 4726–4733. Marasas, W. F. (1996). Fumonisins: History, world-wide occurrence and impact. Adv. Exp. Med. Biol. 392, 1–17. Marasas, W. F. (2001). Discovery and occurrence of the fumonisins: A historical perspective. Environ. Health Perspect. 109(Suppl 2), 239–243. Marasas, W. F., Riley, R. T., Hendricks, K. A., Stevens, V. L., Sadler, T. W., Gelineau-van Waes, J., Missmer, S. A., Cabrera, J., Torres, O., Gelderblom, W. C., Allegood, J., Martinez, C., et al. (2004). Fumonisins disrupt sphingolipid metabolism, folate transport, and neural tube development in embryo culture and in vivo: A potential risk factor for human neural tube defects among populations consuming fumonisin-contaminated maize. J. Nutr. 134(4), 711–716. Martin, E. R., Lai, E. H., Gilbert, J. R., Rogala, A. R., Afshari, A. J., Riley, J., Finch, K. L., Stevens, J. F., Livak, K. J., Slotterbeck, B. D., Slifer, S. H., Warren, L. L., et al. (2000). SNPing away at complex diseases: Analysis of single-nucleotide polymorphisms around APOE in Alzheimer disease. Am. J. Hum. Genet. 67(2), 383–394.
176
J. Gelineau-van Waes et al.
Martinez de Villarreal, L., Perez, J. Z., Vazquez, P. A., Herrera, R. H., del Campos, M. R., Lopez, R. A., Ramirez, J. L., Sanchez, J. M., Villarreal, J. J., Garza, M. T., Limon, A., Lopez, A. G., et al. (2002). Decline of neural tube defects cases after a folic acid campaign in Nuevo Leon, Mexico. Teratology 66(5), 249–256. Matise, T. C., Sachidanandam, R., Clark, A. G., Kruglyak, L., Wijsman, E., Kakol, J., Buyske, S., Chui, B., Cohen, P., de Toma, C., Ehm, M., Glanowski, S., et al. (2003). A 3.9centimorgan-resolution human single-nucleotide polymorphism linkage map and screening set. Am. J. Hum. Genet. 73(2), 271–284. Mayor, S., Sabharanjak, S., and Maxfield, F. R. (1998). Cholesterol-dependent retention of GPI-anchored proteins in endosomes. EMBO J. 17(16), 4626–4638. Melnick, M. and Marazita, M. L. (1998). Neural tube defects, methylenetetrahydrofolate reductase mutation, and north/south dietary differences in China. J. Craniofac Genet. Dev. Biol. 18(4), 233–235. Meredith, F. I., Torres, O. R., Saenz de Tejada, S., Riley, R. T., and Merrill, A. H., Jr. (1999). Fumonisin B1 and hydrolyzed fumonisin B1 (AP1) in tortillas and nixtamalized corn (Zea mays L.) from two different geographic locations in Guatemala. J. Food Prot. 62(10), 1218–1222. Merrill, A. H., Jr., Sullards, M. C., Wang, E., Voss, K. A., and Riley, R. T. (2001). Sphingolipid metabolism: Roles in signal transduction and disruption by fumonisins. Environ. Health Perspect 109(Suppl 2), 283–289. Miller, J. (2001). Occurrence of Fusarium and fumonisins on food grains and in foods. Environ. Health Perspect 109(Suppl 2), 321–324. Milunsky, A., Jick, H., Jick, S. S., Bruell, C. L., MacLaughlin, D. S., Rothman, K. J., and Willett, W. (1989). Multivitamin/folic acid supplementation in early pregnancy reduces the prevalence of neural tube defects. JAMA 262(20), 2847–2852. Miotti, S., Bagnoli, M., Ottone, F., Tomassetti, A., Colnaghi, M. I., and Canevari, S. (1997). Simultaneous activity of two different mechanisms of folate transport in ovarian carcinoma cell lines. J. Cell Biochem. 65(4), 479–491. Miotti, S., Bagnoli, M., Tomassetti, A., Colnaghi, M. I., and Canevari, S. (2000). Interaction of folate receptor with signaling molecules lyn and G(alpha)(i-3) in detergent-resistant complexes from the ovary carcinoma cell line IGROV1. J. Cell Sci. 113(Pt 2), 349–357. Missmer, S. A., Suarez, L., Felkner, M., Wang, E., Merrill, A. H., Jr., Rothman, K. J., and Hendricks, K. A. (2006). Exposure to fumonisins and the occurrence of neural tube defects along the Texas–Mexico border. Environ. Health Perspect 114(2), 237–241. Mizugishi, K., Yamashita, T., Olivera, A., Miller, G. F., Spiegel, S., and Proia, R. L. (2005). Essential role for sphingosine kinases in neural and vascular development. Mol. Cell Biol. 25(24), 11113–11121. Mizugishi, K., Li, C., Olivera, A., Bielawski, J., Bielawski, A., Deng, C. X., and Proia, R. L. (2007). Maternal disturbance in activated sphingolipid metabolism causes pregnancy loss in mice. J. Clin. Invest. 117(10), 2993–3006. Moore, C. A., Li, S., Li, Z., Hong, S. X., Gu, H. Q., Berry, R. J., Mulinare, J., and Erickson, J. D. (1997). Elevated rates of severe neural tube defects in a high-prevalence area in northern China. Am. J. Med. Genet. 73(2), 113–118. Morrison, K., Papapetrou, C., Hol, F. A., Mariman, E. C., Lynch, S. A., Burn, J., and Edwards, Y. H. (1998). Susceptibility to spina bifida; an association study of five candidate genes. Ann. Hum. Genet. 62(Pt 5), 379–396. MRC (1991). Prevention of neural tube defects: Results of the Medical Research Council Vitamin Study. MRC Vitamin Study Research Group. Lancet 338(8760), 131–137. Murray, J. C., Daack-Hirsch, S., Buetow, K. H., Munger, R., Espina, L., Paglinawan, N., Villanueva, E., Rary, J., Magee, K., and Magee, W. (1997). Clinical and epidemiologic studies of cleft lip and palate in the Philippines. Cleft Palate Craniofac J. 34(1), 7–10. Nakano, K. K. (1973). Anencephaly: A review. Dev. Med. Child Neurol. 15(3), 383–400.
Fumonisin and Neural Tube Defects
177
Ncayiyana, D. J. (1986). Neural tube defects among rural blacks in a Transkei district. A preliminary report and analysis. S Afr. Med. J. 69(10), 618–620. Norred, W. P., Plattner, R. D., Meredith, F. I., and Riley, R. T. (1997). Mycotoxin-induced elevation of free sphingoid bases in precision-cut rat liver slices: Specificity of the response and structure-activity relationships. Toxicol. Appl. Pharmacol. 147, 63–70. Norred, W. P., Riley, R. T., Meredith, F. I., Poling, S. M., and Plattner, R. D. (2001). Instability of N-acetylated fumonisin B1 (FA1) and the impact on inhibition of ceramide synthase in rat liver slices. Food Chem. Toxicol. 39, 1071–1078. O’Connell, J. R. and Weeks, D. E. (1995). The VITESSE algorithm for rapid exact multilocus linkage analysis via genotype set-recoding and fuzzy inheritance. Nat. Genet. 11(4), 402–408. Page, S. T., Owen, W. C., Price, K., and Elwood, P. C. (1993). Expression of the human placental folate receptor transcript is regulated in human tissues. Organization and full nucleotide sequence of the gene. J. Mol. Biol. 229(4), 1175–1183. Palencia, E., Torres, O., Hagler, W., Meredith, F. I., Williams, L. D., and Riley, R. T. (2003). Total fumonisins are reduced in tortillas using the traditional nixtamalization method of Mayan communities. J. Nutr. 133, 3200–3203. Park, J. W., Scott, P. M., Lau, B. P., and Lewis, D. A. (2004). Analysis of heat-processed corn foods for fumonisins and bound fumonisins. Food Addit. Contam. 21(12), 1168–1178. Pelagalli, A., Belisario, M. A., Squillacioti, C., Della Morte, R., d’Angelo, D., Tafuri, S., Lucisano, A., and Staiano, N. (1999). The mycotoxin fumonisin B1 inhibits integrinmediated cell-matrix adhesion. Biochimie 81(10), 1003–1008. Penner, J. D., Casteel, S. W., Pittman, L., Jr., Rottinghaus, G. E., and Wyatt, R. D. (1998). Developmental toxicity of purified fumonisin B1 in pregnant Syrian hamsters. J. Appl. Toxicol. 18(3), 197–203. Piedrahita, J. A., Oetama, B., Bennett, G. D., van Waes, J., Kamen, B. A., Richardson, J., Lacey, S. W., Anderson, R. G., and Finnell, R. H. (1999). Mice lacking the folic acidbinding protein Folbp1 are defective in early embryonic development. Nat. Genet. 23, 228–232. Piva, A., Casadei, G., Pagliuca, G., Cabassi, E., Galvano, F., Solfrizzo, M., Riley, R. T., and Diaz, D. E. (2005). Activated carbon does not prevent the toxicity of culture material containing fumonisin B1 when fed to weanling piglets. J. Anim. Sci. 83(8), 1939–1947. Powell, D. C., Bursian, S. J., Bush, C. R., Render, J. A., Rottinghaus, G. E., and Aulerich, R. J. (1996). Effects of dietary exposure to fumonisins from Fusarium moniliforme culture material (M-1325) on the reproductive performance of female mink. Arch. Environ. Contam. Toxicol. 31(2), 286–292. Prakash, K., Pirozzi, G., Elashoff, M., Munger, W., Waga, I., Dhir, R., Kakehi, Y., and Getzenberg, R. H. (2002). Symptomatic and asymptomatic benign prostatic hyperplasia: Molecular differentiation by using microarrays. Proc. Natl. Acad. Sci. 99(11), 7598–7603. Ramasamy, S., Wang, E., Hennig, B., and Merrill, A. H., Jr. (1995). Fumonisin B1 alters sphingolipid metabolism and disrupts the barrier function of endothelial cells in culture. Toxicol. Appl. Pharmacol. 133(2), 343–348. Rampersaud, E., Bassuk, A. G., Enterline, D. S., George, T. M., Siegel, D. G., Melvin, E. C., Aben, J., Allen, J., Aylsworth, A., Brei, T., Bodurtha, J., Buran, C., et al. (2005). Whole genome-wide linkage screen for neural tube defects reveals regions of interest on chromosomes 7 and 10. J. Med. Genet. 42(12), 940–946. Razin, A. and Kantor, B. (2005). DNA methylation in epigenetic control of gene expression. Prog. Mol. Subcell Biol. 38, 151–167. Reddy, R. V., Johnson, G., Rottinghaus, G. E., Casteel, S. W., and Reddy, C. S. (1996). Developmental effects of fumonisin B1 in mice. Mycopathologia 134(3), 161–166. Reeder, J. P., Marasas, W. F., and Vismer, H. F. (2002). Production of fumonisin analogs by Fusarium species. Appl. Environ. Microbiol. 68(5), 2101–2105.
178
J. Gelineau-van Waes et al.
Rietveld, A. and Simons, K. (1998). The differential miscibility of lipids as the basis for the formation of functional membrane rafts. Biochim. Biophys. Acta 1376(3), 467–479. Riley, R. T. and Voss, K. A. (2006). Differential sensitivity of rat kidney and liver to fumonisin toxicity: Organ-specific differences in toxin accumulation and sphingoid base metabolism. Toxicol Sci. 92(1), 335–345. Riley, R. T., An, N. H., Showker, J. L., Yoo, H. S., Norred, W. P., Chamberlain, W. J., Wang, E., Merrill, A. H., Jr., Motelin, G., and Beasley, V. R. (1993). Alteration of tissue and serum sphinganine to sphingosine ratio: An early biomarker of exposure to fumonisincontaining feeds in pigs. Toxicol. Appl. Pharmacol. 118(1), 105–112. Riley, R. T., Wang, E., Schroeder, J. J., Smith, E. R., Plattner, R. D., Abbas, H., Yoo, H. S., and Merrill, A. H., Jr. (1996). Evidence for disruption of sphingolipid metabolism as a contributing factor in the toxicity and carcinogenicity of fumonisins. Nat. Toxins 4(1), 3–15. Riley, R. T., Voss, K. A., Norred, W. P., Bacon, C. W., Meredith, F. I., and Sharma, R. P. (1999). Serine palmitoyltransferase inhibition reverses antiproliferative effects of ceramide synthase inhibition in cultured renal cells and suppresses free sphingoid base accumulation in kidney of BALBc mice. Environ. Toxicol. Pharmacol. 7, 109–118. Riley, R. T., Enongene, E., Voss, K. A., Norred, W. P., Meredith, F. I., Sharma, R. P., Spitsbergen, J., Williams, D. E., Carlson, D. B., and Merrill, A. H., Jr. (2001). Sphingolipid perturbations as mechanisms for fumonisin carcinogenesis. Environ. Health Perspect 109(Suppl 2), 301–308. Riley, R. T., Voss, K. A., Speer, M., Stevens, V. L., and Gelineau-van Waes, J. (2006). Fumonisin inhibition of ceramide synthase: A possible risk factor for human neural tube defects. In ‘‘Sphingolipid Biology,’’ (Y. Hirabayashi, A. H. Merrill, Jr., and Y. Igarashi, Eds.), pp. 345–362. Springer, Toyko. Rosenquist, T. H. and Finnell, R. H. (2001). Genes, folate and homocysteine in embryonic development. Proc. Nutr. Soc. 60(1), 53–61. Rothenberg, S. P., Da Costa, M. P., Sequeira, J. M., Cracco, J., Roberts, J. L., Weedon, J., and Quadros, E. V. (2004). Autoantibodies against folate receptors in women with a pregnancy complicated by a neural tube defect. Obstet. Gynecol. Surv. 59(6), 410–411. Sabbadini, R. A., Betto, R., Teresi, A., Fachechi-Cassano, G., and Salviati, G. (1992). The effects of sphingosine on sarcoplasmic reticulum membrane calcium release. J. Biol. Chem. 267(22), 15475–15484. Sadler, T. W., Merrill, A. H., Jr., Stevens, V. L., Sullards, M. C., Wang, E., and Wang, P. (2002). Prevention of fumonisin B1-induced neural tube defects by folic acid. Teratology 66(4), 169–176. Sahu, S. C., Eppley, R. M., Page, S. W., Gray, G. C., Barton, C. N., and O’Donnell, M. W. (1998). Peroxidation of membrane lipids and oxidative DNA damage by fumonisin B1 in isolated rat liver nuclei. Cancer Lett. 125(1–2), 117–121. Saitsu, H., Ishibashi, M., Nakano, H., and Shiota, K. (2003). Spatial and temporal expression of folate-binding protein 1 (Folbp1) is closely associated with anterior neural tube closure in mice. Dev. Dyn. 226(1), 112–117. Sanchez, T., Skoura, A., Wu, M. T., Casserly, B., Harrington, E. O., and Hla, T. (2007). Induction of vascular permeability by the sphingosine-1-phosphate receptor-2 (S1P2R) and its downstream effectors ROCK and PTEN. Arterioscler. Thromb. Vasc. Biol. 27(6), 1312–1318. Sanna, M. G., Liao, J., Jo, E., Alfonso, C., Ahn, M. Y., Peterson, M. S., Webb, B., Lefebvre, S., Chun, J., Gray, N., and Rosen, H. (2004). Sphingosine 1-phosphate (S1P) receptor subtypes S1P1 and S1P3, respectively, regulate lymphocyte recirculation and heart rate. J. Biol. Chem. 279(14), 13839–13848. Scriver, C. R. (1985). Vitamins: An evolutionary perspective. J. Inherit. Metab. Dis. 8(Suppl 1), 2–7.
Fumonisin and Neural Tube Defects
179
Seefelder, W., Humpf, H. U., Schwerdt, G., Freudinger, R., and Gekle, M. (2003). Induction of apoptosis in cultured human proximal tubule cells by fumonisins and fumonisin metabolites. Toxicol. Appl. Pharmacol. 192(2), 146–153. Sewram, V., Mshicileli, N., Shephard, G. S., and Marasas, W. F. (2003). Fumonisin mycotoxins in human hair. Biomarkers 8(2), 110–118. Shannon, K. M. and Fenichel, G. M. (1990). Pimozide treatment of Sydenham’s chorea. Neurology 40(1), 186. Sharma, R. P., He, Q., Meredith, F. I., Riley, R. T., and Voss, K. A. (2002). Paradoxical role of tumor necrosis factor alpha in fumonisin-induced hepatotoxicity in mice. Toxicology 180(3), 221–232. Sharma, R. P., He, Q., Johnson, V. J., and Voss, K. A. (2003). Increased expression of CD95ligand and other apoptotic signaling factors by fumonisin B1, a hepatotoxic mycotoxin, in livers of mice lacking tumor necrosis factor alpha. Cytokine 24(5), 226–236. Shaw, G. M., Schaffer, D., Velie, E. M., Morland, K., and Harris, J. A. (1995). Periconceptional vitamin use, dietary folate, and the occurrence of neural tube defects. Epidemiology 6(3), 219–226. Shephard, G. S., Thiel, P. G., Stockenstrom, S., and Sydenham, E. W. (1996). Worldwide survey of fumonisin contamination of corn and corn-based products. J. AOAC Int. 79(3), 671–687. Shephard, G. S., Marasas, W. F., Yazdanpanah, H., Rahimian, H., Safavi, N., Zarghi, A., Shafaati, A., and Rasekh, H. R. (2002). Fumonisin B(1) in maize harvested in Iran during 1999. Food Addit. Contam. 19(7), 676–679. Shephard, G. S., van der Westhuizen, L., and Sewram, V. (2007). Biomarkers of exposure to fumonisin mycotoxins: A review. Food Addit. Contam. 24(10), 1196–1201. Silani, V., Bonifati, C., Buscaglia, M., Sampietro, A., Ghezzi, C., and Scarlato, G. (1993). Ganglioside GM1 expression during human spinal cord and neural crest development. Neuroreport 4(6), 767–770. Simons, K. and Ikonen, E. (1997). Functional rafts in cell membranes. Nature 387(6633), 569–572. Singleton, P. A., Dudek, S. M., Chiang, E. T., and Garcia, J. G. (2005). Regulation of sphingosine 1-phosphate-induced endothelial cytoskeletal rearrangement and barrier enhancement by S1P1 receptor, PI3 kinase, Tiam1/Rac1, and alpha-actinin. FASEB J. 19(12), 1646–1656. Smith, G. W., Constable, P. D., Foreman, J. H., Eppley, R. M., Waggoner, A. L., Tumbleson, M., and Haschek, W. M. (2002). Cardiovascular changes associated with intravenous administration of fumonisin B1 in horses. Am. J. Vet. Res. 63(4), 538–545. Spiegel, S. and Milstien, S. (2002). Sphingosine 1-phosphate, a key cell signaling molecule. J. Biol. Chem. 277(29), 25851–25854. Spiegel, S. and Milstien, S. (2003). Sphingosine-1-phosphate: An enigmatic signalling lipid. Nat. Rev. Mol. Cell Biol. 4(5), 397–407. Spiegelstein, O., Mitchell, L. E., Merriweather, M. Y., Wicker, N. J., Zhang, Q., Lammer, E. J., and Finnell, R. H. (2004). Embryonic development of folate binding protein-1 (Folbp1) knockout mice: Effects of the chemical form, dose, and timing of maternal folate supplementation. Dev. Dyn. 231(1), 221–231. Stamm, D. S., Rampersaud, E., Slifer, S. H., Mehltretter, L., Siegel, D. G., Xie, J., Hu-Lince, D., Craig, D. W., Stephan, D. A., George, T. M., Gilbert, J. R., and Speer, M. C. (2006). Highdensity single nucleotide polymorphism screen in a large multiplex neural tube defect family refines linkage to loci at 7p21.1-pter and 2q33.1-q35. Birth Defects Res. A Clin. Mol. Teratol. 76(6), 499–505. Stevens, V. L. and Tang, J. (1997). Fumonisin B1-induced sphingolipid depletion inhibits vitamin uptake via the glycosylphosphatidylinositol-anchored folate receptor. J. Biol. Chem. 272(29), 18020–18025.
180
J. Gelineau-van Waes et al.
Stunff, H. L., Milstien, S., and Spiegel, S. (2004). Generation and metabolism of bioactive sphingosine-1-phosphate. J. Cell Biochem. 92(5), 882–899. Suarez, L., Hendricks, K. A., Cooper, S. P., Sweeney, A. M., Hardy, R. J., and Larsen, R. D. (2000). Neural tube defects among Mexican Americans living on the US-Mexico border: Effects of folic acid and dietary folate. Am. J. Epidemiol. 152(11), 1017–10123. Suzuki, H., Riley, R. T., and Sharma, R. P. (2007). Inducible nitric oxide has protective effect on fumonisin B1 hepatotoxicity in mice via modulation of sphingosine kinase. Toxicology 229, 42–53. Taha, T. A., Hannun, Y. A., and Obeid, L. M. (2006). Sphingosine kinase: Biochemical and cellular regulation and role in disease. J. Biochem. Mol. Biol. 39(2), 113–131. Takabe, K., Paugh, S. W., Milstien, S., and Speigel, S. (2008). ‘‘Inside-out’’ signaling of sphingosine-1-phosphate: Therapeutic targets. Pharmacol. Rev. 60(2), 181–195. Tardieu, D., Tran, S. T., Auvergne, A., Babile, R., Benard, G., Bailly, J. D., and Guerre, P. (2006). Effects of fumonisins on liver and kidney sphinganine and the sphinganine to sphingosine ratio during chronic exposure in ducks. Chem. Biol. Interact 160(1), 51–60. Torres, O. A., Palencia, E., Lopez de Pratdesaba, L., Grajeda, R., Fuentes, M., Speer, M. C., Merrill, A. H., Jr., O’Donnell, K., Bacon, C. W., Glenn, A. E., and Riley, R. T. (2007). Estimated fumonisin exposure in Guatemala is greatest in consumers of lowland maize. J. Nutr. 137(12), 2723–2729. Voss, K. A., Plattner, R. D., Riley, R. T., Meredith, F. I., and Norred, W. P. (1998). In vivo effects of fumonisin B1-producing and fumonisin B1-nonproducing Fusarium moniliforme isolates are similar: Fumonisins B2 and B3 cause hepato- and nephrotoxicity in rats. Mycopathologia 141(1), 45–58. Voss, K. A., Riley, R. T., Norred, W. P., Bacon, C. W., Meredith, F. I., Howard, P. C., Plattner, R. D., Collins, T. F., Hansen, D. K., and Porter, J. K. (2001). An overview of rodent toxicities: Liver and kidney effects of fumonisins and Fusarium moniliforme. Environ. Health Perspect 109(Suppl 2), 259–266. Voss, K. A., Riley, R. T., and Gelineau-van Waes, J. (2006). Trends in fumonisin research: Recent studies on the developmental effects of fumonisins and Fusarium verticillioides. Mycotoxins 55(2), 91–100. Wang, E., Norred, W. P., Bacon, C. W., Riley, R. T., and Merrill, A. H., Jr. (1991). Inhibition of sphingolipid biosynthesis by fumonisins. Implications for diseases associated with Fusarium moniliforme. J. Biol. Chem. 266(22), 14486–14490. Wang, E., Ross, P. F., Wilson, T. M., Riley, R. T., and Merrill, A. H., Jr. (1992). Increases in serum sphingosine and sphinganine and decreases in complex sphingolipids in ponies given feed containing fumonisins, mycotoxins produced by Fusarium moniliforme. J. Nutr. 122(8), 1706–17016. Wang, E., Riley, R. T., Meredith, F. I., and Merrill, A. H., Jr. (1999). Fumonisin B1 consumption by rats causes reversible dose-dependent increases in urinary sphinganine and sphingosine. J. Nutr. 129, 214–220. Watanabe, R., Funato, K., Venkataraman, K., Futerman, A. H., and Riezman, H. (2002). Sphingolipids are required for the stable membrane association of glycosylphosphatidylinositolanchored proteins in yeast. J. Biol. Chem. 277(51), 49538–49544. Watterson, K., Sankala, H., Milstien, S., and Spiegel, S. (2003). Pleiotropic actions of sphingosine-1-phosphate. Prog. Lipid Res. 42(4), 344–357. Weis, B. K., Balshaw, D., Barr, J. R., Brown, D., Ellisman, M., Lioy, P., Omenn, G., Potter, J. D., Smith, M. T., Sohn, L., Suk, W. A., Sumner, S., et al. (2005). Personalized exposure assessment: Promising approaches for human environmental health research. Environ. Health Perspect 113(7), 840–848. Werler, M. M., Shapiro, S., and Mitchell, A. A. (1993). Periconceptional folic acid exposure and risk of occurrent neural tube defects. JAMA 269(10), 1257–1261.
Fumonisin and Neural Tube Defects
181
WHO (2000). ‘‘Environmental Health Criteria 219: Fumonisin B1’’, United Nations Environmental Programme, the International Labour Organization and the World Health Organization International Programme on Chemical Safety, Vol. 153, Geneva, Switzerland. Xiao, K. Z., Zhang, Z. Y., Su, Y. M., Liu, F. Q., Yan, Z. Z., Jiang, Z. Q., Zhou, S. F., He, W. G., Wang, B. Y., and Jiang, H. P. (1990). Central nervous system congenital malformations, especially neural tube defects in 29 provinces, metropolitan cities and autonomous regions of China: Chinese Birth Defects Monitoring Program. Int. J. Epidemiol. 19(4), 978–982. Yazdanpanah, H., Shephard, G. S., Marasas, W. F., van der Westhuizen, L., Rahimian, H., Safavi, S. N., Eskandari, P., and Ghiasian, S. A. (2006). Human dietary exposure to fumonisin B1 from Iranian maize harvested during 1998–2000. Mycopathologia 161(6), 395–401. Yoshizawa, T., Yamashita, A., and Luo, Y. (1994). Fumonisin occurrence in corn from highand low-risk areas for human esophageal cancer in China. Appl. Environ. Microbiol. 60(5), 1626–1629. Yu, C. H., Lee, Y. M., Yun, Y. P., and Yoo, H. S. (2001). Differential effects of fumonisin B1 on cell death in cultured cells: The significance of the elevated sphinganine. Arch. Pharm. Res. 24(2), 136–143. Zhou, J. and Saba, J. D. (1998). Identification of the first mammalian sphinogosine phosphate lyase gene and its functional expression in yeast. Biochem. Biophys. Res. Commun. 242(3), 502–507.
INDEX A Adherent invasive E. coli (AIEC), 9 Anterior cingulated cortex (ACC), 81 Anxiety disorders, 66 C Carbon stable isotope ratio test (SIRA), 138 C2Bbe1 epithelial cells, 7 Chinese hamster ovary (CHO) cells, 161 chlorinated PVC (cPVC) food-packaging, 38 Conjugated linoleic acids (CLA), 10 Contemporary maple syrup evaporator ancillary equipment, 110 arch, 108–109 fuel source, 108 pans, 109–110 Coping, 76 D De novo sphingolipid metabolism biochemical pathway, 164 sphinganine levels, 163 DNA-microarray analysis, 4 E Enterohaemorrhagic E. coli (EHEC), 4 Extended shelf-life (ESL), 27, 52, 54 F Folate binding protein 1 (Folr1), 150 Folate receptor-alpha (FRa), 150 Food allergy biopsychosocial development, stress impact, 80–81 comorbid anxiety disorders, 66 developmental pathways avoidant and minimizing strategies, 90 awareness, 89 emotional development, 86–87 event control, 88 parental views and management, 91–92
precipitating events, 87 transition points, 90–91 gender impact influence, 83–84 vs. sex impact, 81 HRQL impact, research disease-specific measures, 74–75 generic measures, 73–74 neurocognitive development, 67–68 parental perceptions developmental issues, 78 IAT measures, 78–79 prevalence, mechanisms and clinical manifestations allergic diseases, 69 anaphylaxis, 72 atopy, 68–69 food hypersensitivity, 70–72 psychological burden adaptational processes, 77 coping, 76–77 research biopsychosocial framework, 94 children, 68 literature issues, 93 risk behavior factors, 85 health belief models (HBMs), 86 perception, 85 sex impact vs. gender impact, 81 influence, 82–83 social support, 79–80 Food allergy quality of life-parental burden (FAQL-PB), 74–76 Food-packaging interactions consumer perception interactions, 20–21 sensory problem, 21 evaluation, 18 packaging materials, 19 quality improvement antimicrobial ingredients, 53–54 flavor and odor absorbers, 54–55
183
184
Index
Food-packaging interactions (cont.) oxidation control, 55–56 quality protection light, 50–52 moisture loss prevention, 52–53 package characteristics, 49–50 yogurt package, 50 scalping/sorption color and appearance, 48 polymers and, 49 properties, 47–48 sensory effects discrimination testing, 26 evaluation, 23 flavor chemistry, 22–23 gas chromatography–mass spectroscopy (GC–MS), 25 negative sensory impacts, 24 packaging materials, 22 scalping/sorption, 24–25 sensory methodology books, 26–27 taints, 23–24 taints contact materials, 27–45 potential migrants, 27 technical representatives communication, 20 threshold concept detection/absolute, 22 sensory response, 21 Fumonisin exposure and toxicity animal studies, 155–156 bioactive sphingoid base-1-phosphates fumonisin and accumulation, 164–166 neural tube closure, 166 porcine pulmonary edema, 165–166 signaling functions, 164 SIP gradients and S1P receptors, 164–165 chemical structural considerations ceramide synthase, 159 corn cooking, 160 de novo sphingolipid biosynthetic pathway, 159–160 de novo sphingolipid metabolism biochemical pathway, 163 sphinganine levels, 162 in vitro and in vivo, 162–163 folate transport disruption folate uptake, 168 receptor levels, 168–169 glycosphingolipids fumonisin depletion
GPI-anchored protein, 166 LM/Bc in vivo mouse model, 167 treatments, 168–169 vitamin uptake, 169 measurements, 154–155 and regulatory policy, 153–154 reproductive toxicology animal studies, 155–156 mouse models, 156–158 sphingolipids, 160–161 G GI tract mucins, 5 Glycosylphosphatidylinositol (GPI), 150 H Health belief models (HBMs), 85 Health related quality of life (HRQL) disease-specific measures, 74–76 generic measures, 73–74 Human b-defensin-2 (hBD-2) gene expression, 6 I IBD. See Inflammatory bowel disease Immunomodulation, 8–10 Implicit association test (IAT), 78 Inflammatory bowel disease (IBD), 10 Insulin-dependant diabetes mellitus (IDDM), 74 L Lactobacillus acidophilus R0052, 4–5 Lactobacillus rhamnosus GG (LGG), 11 Lactobacillus rhamnosus R0011, 5 M Maple candy, 136 Maple cream, 136 Maple syrup adulteration bleaching treatment, 137–139 carbon stable isotope ratio test (SIRA), 138 screening method, 137–139 annual production, 110–111 chemistry composition values and ranges, 129–130 conductivity and color, 127
Index
density and carbohydrates, 126 flavor compounds, 131–133 flavor sensory evaluation, 133–134 inorganic composition, 128, 131 nutritional aspects, 135–136 off-flavors, 134–135 organic acids, 131 pH, 126 rheology, 127–128 vitamins, 136 contamination, 137 cream and candy, 136–137 and decolorizing resin, 139 history maple trees sap, 102–103 sap processing, 103–104 maple sap flow phenomenon, 105 physiological process, 104–105 sap chemistry evaporation by heating and transformation, 117–121 inorganic composition, 112–113 organic acids, 113–115 phenolic compounds, 115 pH range, 111 reverse osmosis/nanofiltration transformation, 116–117 storage transformation, 115–116 sugar composition, 111–112 transmission profile, 114 sap collection plastic tubing, 106–107 tapholes, 107 sap processing, evaporation, 107–110 scale/sugar sand formation composition and variation, 122 evaporator surfaces, 123 niter, 121 standards color, flavor and purity, 125 grades, 124 Maple tubing system, 106–107 N Neural tube defects (NTDs) environmental risk factors, fumonisins, 148–149 fumonisin-induced mouse models in cultured mouse embryos, 156–157 maternal fumonisin exposure, 157–158 genetic risk factors
185
DNA methylation, 153 folic acid, 152–153 genes—genes candidate, 152 syndromes, 151 nutritional risk factors, folic acid folate mechanisms, 150 maize consumption, 151 overview, 148 spina bifida, 148 P Paraformaldehyde (PFA), 107 Peripheral blood mononuclear cell (PBMC), 8 Polyamides (PA) food-packaging, 38 Polyethylene terephthalate (PETE) food-packaging, 33–34, 38–39, 49, 54 bottles contaminant analysis, 47 light-exposed and protected milk, 41 ozonated water, 40 high-density poly ethylene (HDPE), 51–52 Polyolefins food-packaging, 38 acetaldehyde, 41 aeration, 39 chlorinated and nonchlorinated water, 42 HDPE and epoxy, 43–44 off-odor development, 44 oxidation products, 40 packaging flavor, 43 sensory characteristics, 39–40 water distribution, 41–42 Polystyrene (PS) food-packaging monomer thresholds, 37 refrigerated storage, 36 signal detection method, 37–38 storage temperature, 35–36 styrene monomer concentration, 35 styrene taint, 38 Polyvinyl chloride (PVC) food-packaging, 38 Probiotics, gastrointestinal pathogens mechanisms, 2, 4 acid production and inhibitory substances secretion, 7–8 epithelial barrier function and probiotic signaling, 4–6 immunomodulation, 8–10 virulence factor expression, 10–11 strains adhesion ability, 6
186
Index
Probiotics, gastrointestinal pathogens (cont.) and beneficial effects, 3 immunomodulation, 9–10 potential mechanism of action, 10–11 VSL#3, 5 Provisional maximum tolerated daily intake (PMTDI), 153 Pyrazines, 132 R Reverse osmosis (RO), 116–117 Robinson test, taints, 28–29 S Shiga toxin-producing E. coli (STEC), 7 Short-chain fatty-acids (SCFAs), 5 Single nucleotide repeat polymorphisms (SNPs), 152 SIRA. See Stable isotope ratio test Small-medium sized enterprises (SMEs), 75 Sphingosine-1-phosphate (S1P), 161, 164 Stable isotope ratio test (SIRA), 138 Study specific questionnaire (SSQ), 79
T Taints, food-packaging additives/noncontacting materials coloring agents and antioxidants, 46 printing inks, varnishes, 45–46 contact materials, 27 epoxy, 44–45 polyamides (PA), 38 polyolefins, 34–44 polystyrene (PS), 35–38 PVC and cPVC, 44 Robinson test, 28–29 sensory active components, 28 sensory descriptors, 29–34 definition, 23–24 recycled materials absorbed odorous, 46 PETE, 47 T84 epithelial cells, 4 Triangle test, food-packaging antioxidant level, 55 basil taint, cheese, 53–54 squashes and fruit beverages, 49 Tumor necrosis factor a (TNFa), 161
