UBC Faculty Research and Publications

Investigation of the Putative Associations between Dairy Consumption and Incidence of Type 1 & Type 2… Lacroix, Isabelle Marie Estelle; Li-Chan, Eunice Chi Yu Oct 9, 2012

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
52383-Lacroix_I_et_al_Investigation_putative_associations_dairy.pdf [ 1.32MB ]
Metadata
JSON: 52383-1.0343572.json
JSON-LD: 52383-1.0343572-ld.json
RDF/XML (Pretty): 52383-1.0343572-rdf.xml
RDF/JSON: 52383-1.0343572-rdf.json
Turtle: 52383-1.0343572-turtle.txt
N-Triples: 52383-1.0343572-rdf-ntriples.txt
Original Record: 52383-1.0343572-source.json
Full Text
52383-1.0343572-fulltext.txt
Citation
52383-1.0343572.ris

Full Text

 1 Investigation of the Putative Associations between Dairy Consumption and Incidence of Type 1 & Type 2 Diabetes     ISABELLE M.E. LACROIX AND EUNICE C.Y. LI-CHAN The University of British Columbia Faculty of Land & Food Systems Food Nutrition & Health Program 2205 East Mall, Vancouver, BC, Canada. V6T 1Z4.       Address correspondence to Eunice Li-Chan, The University of British Columbia, Faculty of Land & Food Systems, Food Nutrition & Health Program, 2205 East Mall, Vancouver, BC, Canada. V6T 1Z4.  Tel: 604 822-6182; Fax: 604 822-5143; E-mail: Eunice.li-chan@ubc.ca  2 ABSTRACT  A growing body of evidence suggests a possible relationship between consumption of dairy products and the incidence of diabetes. A positive correlation between the early introduction of dairy in infancy and the incidence of type 1 diabetes (T1D) in genetically predisposed infants has been suggested by studies on rodents and humans. However, the lines of evidence supporting this association, including epidemiological studies and the observation of antibodies to bovine serum albumin, β-casein and bovine insulin in the serum of patients with T1D, are not without controversy. On the other hand, an inverse relationship between consumption of dairy foods and the development of metabolic syndrome and/or type 2 diabetes (T2D) has been implied by epidemiological studies. Several dairy components, especially milk proteins, are believed to play a role in the beneficial effect of dairy consumption on glucose regulation by modulation of incretin hormones. Other dietary factors have also been associated with the incidence of T1D and T2D, indicating that dairy foods might be only one among many dietary agents possibly implicated in the development of diabetes. The present article critically reviews the evidence and plausible mechanisms for the putative associations between dairy food consumption and type 1 and type 2 diabetes.  Keywords:  diabetogenicity, -cell autoimmunity, glycemic regulation, infant feeding, milk components        3 INTRODUCTION   The incidence of type 1 diabetes (T1D) and type 2 diabetes (T2D) is rising considerably worldwide (Wild et al., 2004). It is estimated that 285 million people are currently affected by diabetes and that each year 7 million new cases of diabetes are diagnosed (Canadian Diabetes Association, 2011). To explain the increasing prevalence of diabetes, several hypotheses have been put forward that involve the consumption of dairy foods. These hypotheses are based on studies that suggest a positive correlation between the consumption of dairy products and T1D on the one hand, and an inverse relationship between dairy consumption and T2D on the other hand, and have fuelled what has been referred to as “the milk debate” (King, 2005).  Type 1 diabetes, also referred to as insulin-dependent diabetes mellitus (IDDM), is an autoimmune disease usually occurring early in life in individuals presenting an inherited genetic susceptibility (Merriman, 2009). This disorder results in the attack and destruction of the insulin-producing β-cells of the pancreas by T lymphocytes (Åkerblom et al., 2002), leading to the incapacity of the pancreas to produce the required insulin to maintain normal blood glucose levels (Merriman, 2009). The occurrence of T1D has escalated in the Westernised world during the past two or three generations, reaching an annual worldwide increase of about 2.8% in the 1990s (The DIAMOND Project Group, 2006). This significant increase is suggested to be indirect evidence that environmental factors play an important role in T1D, and among the major putative environment factors, the early exposure to cow’s milk proteins has particularly attracted the attention of the scientific community (Åkerblom et al., 2002). The current data do not reach the conclusion that cow’s milk as such causes diabetes, but increasing evidence from studies on rodents and humans suggest that the pathogenic mechanisms linking the exposure to cow’s milk proteins with development of T1D could be  4 related to the regulation of oral tolerance (Harrison & Honeyman, 1999; Kolb & Pozzilli, 1999; Wasmuth & Kolb, 2000; Åkerblom et al., 2002).   On the other hand, type 2 diabetes is a chronic metabolic disease representing 90-95% of all cases of diabetes in individuals over 20 years of age, and is characterized by the occurrence of fasting as well as postprandial hyperglycemia (Sebokova et al., 2007). It is believed that hyperglycemia in T2D is caused by several factors, such as a dysfunction of the pancreatic β-cells, insulin resistance or deficiency in insulin secretion, as well as being the result of an increase in hepatic glucose production (Sebokova et al., 2007). T2D is an increasing problem worldwide and it is estimated that the number of people with diabetes will reach 300 million worldwide by 2025 (Choi et al., 2005). An inverse association between the consumption of dairy products and/or dairy components and the incidence of T2D as well as the development of the metabolic syndrome (also known as insulin resistance syndrome), has been observed in several observational studies (for example, Pereira et al., 2002, Elwood et al. 2008 & Tremblay & Gilbert, 2009). This beneficial effect of dairy foods has been attributed to many dairy components, such as its content of medium chain fatty acids, whey proteins, calcium and other minerals (Pfeuffer & Schrezenmeir, 2006), and more recently to monounsaturated fatty acids such as palmitoleic acid (Mozaffarian et al., 2010).  The aim of this review is to examine the reported evidence and plausible mechanisms for the associations between dairy product consumption and incidence of type 1 and type 2 diabetes.    5 COW’S MILK CONSUMPTION AND TYPE 1 DIABETES: POSSIBLE ASSOCIATIONS   While the genetic susceptibility to diabetes is inherited, environmental factors such as diet or enteroviral infections are believed to have an important influence on whether the disease will actually be expressed in genetically pre-disposed individuals (Åkerblom et al., 2002; MacFarlane et al., 2003). One of the dietary factors suggested to play a role in the development of T1D is the early introduction in infancy of cow’s milk into the diet. In fact, it was the observation of an inverse association between incidence of IDDM and breast-feeding from population data and case-control data in Norway, Sweden and Denmark (Borch-Johnsen et al., 1984) that led to further research to explore the role of infant feeding in the etiology of T1D.  Over the last few decades, numerous studies have looked into the effects of early introduction of cow’s milk protein on the risk of T1D and -cell autoimmunity. The findings of these studies, however, have been inconsistent, with some reporting an increased risk of diabetes (Scott, 1990; Dahl-Jørgensen et al., 1991; Virtanen et al., 1993; Fava et al., 1994; Gerstein, 1994; Norris & Scott, 1996; Muntoni et al., 2000; Thorsdottir & Ramel, 2003; Rosenbauer et al., 2007) or -cell autoimmunity (Kimpimäki et al., 2001; Holmberg et al., 2007) in children exposed to cow’s milk, while others have reported no correlation (Norris et al., 1996; Couper et al., 1999; Esfarjani et al., 2001; Strotmeyer et al., 2004). Some of these studies are presented in the following sections.  Selected to represent the range of human studies that have been conducted, including ecological, epidemiological, case control, prospective and retrospective cohort, as well as intervention studies, they serve to illustrate the inconsistent outcomes and complexity underlying the association between dairy product consumption and T1D.    6    DATA FROM HUMAN STUDIES – COW’S MILK EXPOSURE AND INCIDENCE OF TYPE 1 DIABETES  Ecological and Epidemiological Studies  A correlation between the exposure to cow’s milk and the incidence of T1D between and within countries has been found in several epidemiological and ecological studies (Scott, 1990; Dahl-Jorgensen et al. 1991; Fava et al., 1994; Muntoni et al., 2000; Thorsdottir & Ramel, 2003) (Table 1). Scott (1990) investigated the data on consumption of unfermented milk products from 13 countries and found a significant positive correlation with the incidence of T1D (r = 0.86; p < 0.01). Similarly, Dahl-Jørgensen et al. (1991) found a correlation (r = 0.96) between the consumption of fluid milk in 12 countries and age-standardized incidence rates of IDDM in children 0 to 14 years old. The risk of T1D in children 0-14 years of age of nine regions of Italy was also significantly correlated to cow’s milk consumption (r = 0.84; P < 0.004); however, no correlation was found for cheese intake (Fava et al., 1994).  Similar findings were obtained by Muntoni et al. (2000) who investigated the relationship between T1D in children and adolescents under 15 years of age from 40 countries and their dietary energy intake from major food groups. Dairy products were found to be predictors of elevated prevalence rates of T1D (r = 0.80; P < 0.0001) (Muntoni et al., 2000). However, nutritional data in this study were derived from the Food and Agriculture Organization’s food balance sheets and thus, food intake may have been overestimated in wealthy countries. More recently, Thorsdottir and Ramel (2003) studied the possible role of cow’s milk in the etiology of T1D by investigating the food and nutrient intakes of 4,701 children and adolescents ranging from 10 to 16 years in age in relation to the incidence of diabetes in 11  7 European countries. The prevalence of T1D was significantly associated with daily intake of milk (r = 0.829; p 0.042) or animal products (including milk, meat and eggs) (r = 0.999; p 0.001). However, these correlations were observed only when Icelandic data were excluded. The authors justified this exclusion on the controversial premise that Icelandic cow’s milk is less diabetogenic than milk from other countries due to its different protein composition, particularly the β-casein fraction (Thorsdottir & Ramel, 2003).    Although in developed countries milk-based formula are often the first complementary food to which babies are exposed, the proportions of infants first exposed to cow’s milk varies between countries and cultures. Moreover, the type of complementary feeding that infants who are not first introduced to milk-based formula receive also differs between nations (Knip et al., 2010). This may be an important confounder in the ecological and epidemiological studies that have investigated the link between the exposure to cow’s milk and the risk of T1D. Dairy products are not the only dietary factors that have been linked to IDDM; meat products (Muntoni et al., 2000) as well as fruits and vegetables (Thorsdottir & Ramel, 2003) have also been associated with the disease.       Case-Control Studies   A positive association (odds ratio (OR) = 1.57, 95% confidence interval (CI) 1.19-2.07; P 0.10 for homogeneity) between the early exposure to cow’s milk (before 3-4 months of age) and the incidence of T1D was found in a meta-analysis of 13 case-control studies (Gerstein, 1994). A further meta-analysis of 17 case-control studies by Norris and Scott (1996) revealed an analogous relationship between early infant diet and T1D. The summary of all studies showed a moderate effect for exposure to breast-milk substitutes (OR = 1.38, 95% CI, 1.18-1.61) and cow’s milk-based substitutes (OR = 1.61, 95% CI, 1.31-1.98) before  8 the first 3 months of life. The putative diabetogenicity of dairy product consumption in infancy was also observed in a Finnish nationwide case-control study in which children exposed to dairy before the age of 2 months were found to present a greater risk of diabetes compared to those receiving dairy products later (Virtanen et al., 1993) (Table 2). Over the last few decades, a trend towards a longer duration of breastfeeding and a delay in the introduction of supplementary milk feeding has been observed in several countries, including Finland. To assess whether these changes in feeding patterns have an impact on the association between the exposure to dairy foods and the incidence of T1D, Virtanen et al. (1993) also investigated the data from two birth year cohorts (1972-1979 versus 1980-1988). The authors reported no significant difference between the two cohorts in the relationship between the age at introduction of dairy products and the risk of T1D (≥2 mo versus <2 mo; OR = 0.64 and 0.46 for the 1972-1979 and the 1980-1988 cohorts, respectively), which suggests that this association was not affected by the changes in the feeding patterns in early infancy in Finland.  In a recent study, Rosenbauer et al. (2007) investigated early infant feeding patterns as well as other environmental factors on the risk of T1D in children under five years of age in Germany. An increased risk of T1D was observed in children introduced to formula/cow’s milk before 5 months of age compared to those exposed after 5 months of age (OR = 1.34, 95% CI 1.03-1.74; P 0.030). A short breastfeeding period was also found to be associated with an increased risk of T1D (<5 mo versus ≥5 mo; OR = 1.31, 95% CI 1.01-1.69; P 0.039). On the other hand, the influence of the exposure to cow’s milk on the incidence of T1D was not observed in all case-control studies (Table 2). Esfarjani et al. (2001) reported no relationship between early cow’s milk exposure and T1D in 52 IDDM patients and controls from Tehran. Moreover, in a larger study by Strotmeyer et al. (2004), in which the effects of dietary exposure during the first year of life on the risk of diabetes was investigated using data  9 collected for 247 Chinese infants with T1D and 443 controls, cow’s milk consumption was not associated with T1D before 6 months of age, whereas a negative association was observed between the seventh and twelfth month of life (OR = 0.60, 95% CI 0.43-0.85; P 0.004).   Several factors may account for the discrepancy observed in these retrospective studies. The data on breastfeeding duration and age at introduction of cow’s milk or other food products rely on parental recall which can be inaccurate or biased especially when the length of the recall is long (Norris & Scott, 1996; Rosenbauer et al., 2007). Moreover, it has been suggested that mothers of children affected by T1D may have a biased recall of their child’s diet compared with those whose child does not have T1D (Norris & Scott, 1996; Strotmeyer et al., 2004). Differences in the response rate of cases and controls may also introduce variability (Norris & Scott, 1996). Numerous studies failed to define whether “breastfeeding duration” referred to exclusive breastfeeding or overall breastfeeding, and often did not specify whether breastfeeding was stopped completely at the age that cow’s milk was reported to be first introduced into the infant’s diet. These aspects may also be a source of discrepancy.     DATA FROM HUMAN STUDIES – COW’S MILK EXPOSURE AND  -CELL AUTOIMMUNITY  The presence of diabetes-associated autoantibodies, including islet cell antibodies (ICA), insulin autoantibodies (IAA), glutamic acid decarboxylase antibodies (GADA) and tyrosine phosphatase-like protein IA-2 antibodies (IA-2A), is the first noticeable sign of emerging -cell autoimmunity during the period preceding the manifestation of the disease (Virtanen & Knip, 2003). The occurrence of two or more types of these antibodies significantly increases the risk of progression to T1D (Couper et al., 1999; Knip et al., 2010). Positivity for diabetes-associated autoantibodies has been used as a surrogate marker of  10 clinical T1D in prospective studies that have looked at the effects of infant feeding patterns on the incidence of the disease. These studies (Table 3), like those which have investigated the effects of cow’s milk protein on the incidence of T1D, have produced contradictory outcomes (Virtanen & Knip, 2003).  Observational Studies      Kimpimäki et al. (2001) investigated the relation between early infant feeding and signs of -cell autoimmunity in 2949 infants with increased genetic risk of T1D in the Finnish Type 1 Diabetes Prediction and Prevention (DIPP) birth cohort. Analyses of odds ratios with confidence intervals were based on a case control study comprising 65 cases (who showed islet cell antibody positivity before 4 years of age) and 390 controls (who were islet cell antibody-negative) selected from the cohort. Children who first received cow’s milk before 2 months of age or between 2 to 3.9 months of age were found to have a higher risk of seroconversion to positivity for ICA, IAA, GADA and IA-2A antibodies compared to those first introduced to cow’s milk after the age of four months (OR = 9.46; CI 1.50-59.8; p 0.017, and OR =17.9; CI 1.87-172; p 0.012, respectively).  A short duration of both total and exclusive breast-feeding, as well as the early introduction of cow’s milk based formula, were also associated with an increased risk of -cell autoimmunity in children from the All Babies in Southeast Sweden (ABIS) Study (Holmberg et al., 2007). In this prospective study, 3788 children from the general population were followed; 12 of these children developed T1D before 5-6 years of age and were not included in the analysis. The age at which infants were first exposed to formula was correlated with the duration of exclusive breast-feeding (P<0.000), and the early exposure to formula (1-3 months of age) was associated with a higher risk of positivity for GADA, IAA  11 and/or IA-2A above the ninety-ninth percentile (OR = 1.84, 95% CI 1.01, 3.37; P 0.047) in the non-diabetic children at 5 years of age (Holmberg et al., 2007). On the other hand, the early introduction of cow’s milk protein was not found to be significantly associated with -cell autoimmunity in 253 non-diabetic children in The Diabetes Autoimmunity Study in the Young (DAISY) (Norris et al., 1996). In this retrospective study, children selected were from 171 families with a member affected by T1D and between the ages of 9 months and 7 years when screened for -cell autoimmunity (BCA). Among the 253 children studied, 18 cases of BCA were identified at baseline and 153 children without BCA were used as controls. No difference was found in the proportion of cases and controls who were introduced to cow’s milk before 3 months or 6 months of age. Similar findings were reported by Couper et al. (1999) who prospectively followed, from birth for a median of 29 months, 317 children with a first-degree relative with T1D of The Australian Baby DIAB Study. The authors found no association between the development of -cell autoimmunity and the introduction of cow’s milk in these high-risk children.  The possible role of the mother’s diet during pregnancy in the development of -cell autoimmunity in the offspring has also been investigated. In a recent prospective cohort study by Virtanen et al. (2011), the authors assessed whether food consumption of mothers during pregnancy was associated with the development of islet cell antibodies and other diabetes-associated autoantibodies in children with increased human leukocyte antigen (HLA)-DQB1-conferred risk of T1D. No association was found between the maternal intake of milk and dairy products and -cell autoimmunity in the infant. Moreover, no relationship between the maternal diet during pregnancy and the incidence of diabetes-associated autoantibodies was found in a randomized trial in which the mothers’ diet either contained or did not contain cow’s milk during the last trimester of their pregnancy (Ludvigsson, 2003). Similarly, maternal diet during the third trimester was not associated with islet autoimmunity in children  12 from the DAISY birth cohort presenting a high- and moderate-risk HLA genotype (Lamb et al., 2008).   On-Going International Prospective Studies  The Trial to Reduce IDDM in the Genetically at RISK (TRIGR) is the only dietary intervention trial that has been carried out to date to monitor the effects of early nutritional intervention on markers of β-cell autoimmunity and progression to T1D. Although the TRIGR study is still ongoing to follow the outcomes in terms of overt T1D, the effects of the intervention on markers of β-cell autoimmunity have been recently published (Knip et al., 2010). In this double-blind, randomized trial, 230 Finnish infants presenting increased risk of T1D were assigned at weaning either a highly hydrolyzed formula (“intervention”, containing extensively hydrolyzed casein) or a regular cow’s milk-based formula (“control”, containing 80% intact milk protein and 20% hydrolyzed milk protein).  Breast-feeding was encouraged until weaning, and the study formulae were available until the infants were 8 months of age. Study subjects were followed-up with respect to β-cell autoimmunity as well as progression to overt diabetes during a mean period of 4.7 years. In comparison to the control, the intervention was shown to present a protective effect against positivity to at least one diabetes-associated autoantibodies (after adjustment for the duration of exposure to the study formula, HR = 0.51, 95% CI 0.28-0.91; P 0.02). On the other hand, no significant association was found between the risk of T1D and the feeding intervention (after adjustment for the duration of exposure to the study formula, HR = 0.48, 95% CI, 0.14-1.61) (Knip et al., 2010). However, it should be noted that this study was not designed with clinical T1D as the primary end point (Knip et al., 2010), and thus information regarding feasibility of early nutritional intervention to reduce risk of T1D awaits the larger ongoing TRIGR study involving 2160  13 children from 15 countries, the results of which are expected to be available in the year 2017 (TRIGR, 2011).  In addition to the TRIGR study, another international prospective study, The Environmental Determinants of Diabetes in the Young (TEDDY), is currently underway.  The TEDDY study consortium encompasses six clinical centres in the United States, Germany, Sweden and Finland, whose goal is to identify environmental factors (infectious agents, dietary factors, psychosocial and other factors) which either trigger or protect against the development of islet autoimmunity and T1D (TEDDY Study Group, 2008).  To meet this goal, from 2004-2009 the TEDDY group has screened newborns from both the general population and high-risk families; the prospective follow-up study is underway and will continue until the subjects reach 15 years of age.  It is anticipated that the information from the findings of the TRIGR and TEDDY prospective studies, which will become available in the next two decades, will provide useful insights to explain some of the present conflicting evidence on the role of dietary factors and other environmental factors on the incidence of T1D.   DAIRY CONSUMPTION AND TYPE 1 DIABETES: POSSIBLE MECHANISMS    The hypothesis that dairy consumption is associated with risk of developing T1D has evolved from several lines of evidence, including the observation of antibodies to bovine serum albumin (BSA) (Persaud & Barranco-Mendoza, 2004), β-casein (Bell et al., 2006) and bovine insulin (Vaarala et al., 1998) in the serum of patients with T1D, as well as the isolation of T-cell lines specific to bovine β-casein in patients with T1D (Monetini et al, 2003). These observations have suggested that various constituents in bovine milk, including the A1 variant of -casein, bovine serum albumin and insulin may act as triggering agents in the  14 development of T1D. A critical evaluation of these factors as well as the role of the gut mucosal immune function as potential mechanisms underlying the association of milk consumption with the risk of T1D is presented in the following sections.  Bovine A1/A2 -Casein Variant   Caseins comprise approximately 80% of the proteins in bovine milk, and can be classified based on sequence homology into four families, namely αs1-, αs2-, β- and κ-caseins, each with a number of genetic variants (Farrell Jr. et al. 2004). Among these, the A1 variant of cow’s milk -casein has been suggested to play a role in the development of T1D.   Thirteen genetic variants of β-casein exist: A1, A2, A3, B, C, D, E, F, G, H1, H2, I, (Farrell Jr. et al., 2004) and an A4 allele, which has been found in Korean native cattle, but whose nucleotide substitution is not presently recognized (Kamińsky et al., 2007). The A1 and A2 variants of β-casein are the most common in dairy cattle breeds, with the A2 variant being more frequent in breeds such as Simmental, Jersey and Guernsey (Bell et al., 2006), which are especially predominant in Central and Southern Europe. On the other hand, the A1 variant predominates in Northern Europe and North America, where the black- and red-and-white breeds like Ayrshire, Holstein and Friesian, are the most common (Laugesen & Elliott, 2003; Bell et al., 2006; Merriman, 2009). In the United States, the Holstein breed, which produces milk containing about 42-62% A1 variant in the β-casein fraction (De Noni et al., 2009), represents about 92% of the dairy cows, whereas the Guernsey breed, which produces milk containing 96-98% A2 variant in the β-casein fraction (De Noni et al., 2009), account for  only 1% (Bell et al., 2006). The Holstein breed is also the most common dairy breed in Canada, representing about 93% of the dairy herd (Canadian Dairy Information Centre, 2010). Thus, the North American population consumes high levels of the A1 variant of β-casein.  15  The A1 and A2 variants differ according to the amino acid located in position 67 of the β-casein molecule: a proline is present in the A2 variant, whereas a histidine occupies this position in the A1 form (Table 4). The enzymatic digestion of the A1 variant by gastrointestinal proteases leads to the liberation of a 7-residue peptide with the following sequence: 60Tyr-Pro-Phe-Pro-Gly-Pro-Ile66. This β-casomorphin-7 peptide, which is not released from the A2 variant, presents opioid properties and is believed to have an impact on mucus secretion, gut motility and gene expression, as well as to influence lymphocyte proliferation (Kamińsky et al., 2007; Merriman, 2009). It has been suggested that the A1 form of β-casein could play a role in the development of T1D in addition to representing a risk factor for cardiovascular disease, sudden infant death syndrome and neurological disorders such as schizophrenia and autism (Bell et al., 2006; Kamińsky et al., 2007). By acting as an immune suppressant, it has been hypothesized that β-casomorphin-7 disturbs the gut-associated immune tolerance or reduces the patient’s defence towards enterovirus, which is reported to be a possible cause of the development of T1D in genetically predisposed individuals (Kamińsky et al., 2007).  The foundation of the hypothesis that A1 β-casein is linked to T1D rests on epidemiological data that showed a positive correlation between the consumption of cow’s milk containing A1 β-casein and the incidence of T1D in 14-year-old children in 19 countries (Elliott et al., 1999; Laugesen & Elliott, 2003). This epidemiological association is not without controversy (Truswell, 2006; De Noni et al., 2009). One of the major criticisms is the weakening of the correlation by the inclusion of cheese-derived A1 β-casein, instead of considering only milk and cream as sources of A1 β-casein. In fact, a recent study by De Noni and Cattaneo (2010) showed that the in vitro gastrointestinal digestion of cheeses, including varieties such as Cheddar, Gouda and Brie, led to the release of β-casomorphin-7. Moreover, there is an uncertainty surrounding the measurement of A1 β-casein consumption. As pointed  16 out by Truswell (2006), using data based on adult per capita consumption of milk is not ideal in the context of T1D, which usually occurs early in life, since most of the non-breastfed infants are fed with formulas that often are based on a high whey protein content and a reduced casein content. Moreover, the number of countries included in these studies was relatively small, mainly limited by the availability of complete data on breed composition and milk polymorphism, which weakens the robustness of the data. In fact, if Laugesen and Elliott (2003) had included in their analysis the consumption of A1 -casein from Sardinia, the country presenting the second highest worldwide incidence of T1D,  the correlation coefficient between T1D and A1 β-casein consumption would have been reduced from r = 0.92 to r = 0.79 (Merriman, 2009).    The biological evidence supporting the role of A1 β-casein in T1D is also weak. In fact, the animal experiments on non-obese diabetic (NOD) mice failed to confirm that A1 β-casein is more diabetogenic than the A2 form (Beales et al., 2002). In human studies, the levels of specific antibodies to A1 and A2 β-casein were not significantly different in the sera of children with T1D versus their siblings, whereas a significant difference was observed between T1D patients and controls/parents as well as between their siblings and controls/parents. An inverse correlation of A1 and A2 β-casein antibody levels with age was also observed. These findings suggest that the differences in level of A1 and A2 antibodies between various groups of individuals may not be specific to T1D but instead may be linked to age. The latter finding might mirror an infant’s underdeveloped immune system’s poor tolerance of exogenous antigens, in contrast to any specific association between plasma level of β-casein auto-antibodies and T1D (Merriman, 2009).  Human β-casein contains a sequence of seven amino acids (51Tyr-Pro-Phe-Val-Glu-Pro-Ile57) which is homologous to bovine β-casomorphin 7 (Jarmolowska et al. 2007; De Noni et al., 2009). Interestingly, Jarmolowska et al. (2007) have observed the presence of  17 various levels of β-casomorphin 5 and β-casomorphin 7 in human milk depending on the stage of lactation, and this despite the presence of a proline residue in position 58 of the β-casein sequence. It should be noted that human and bovine β-casomorphin 7 differ by two amino acids, a proline and a glycine in the bovine sequence having been replaced by a valine and a glutamine in the human sequence, and that the opioid activity of human β-casomorphin 7 is reported to be weaker than the bovine form (Koch et al., 1985; Chruścińska et al., 1998).  Nevertheless, the natural occurrence of β-casomorphin 7 in human milk challenges the hypothesis that the release of β-casomorphin 7 from bovine milk during digestion is the cause of the positive correlation observed in epidemiological studies between the consumption of cow’s milk containing A1 β-casein and the incidence of T1D, and suggests that other factors might be responsible for the observed association.  Furthermore, the observation of a correlation between the incidence of T1D and latitude has led to the hypothesis that reduced sun exposure, correlated to latitude and associated with a reduction of vitamin D intake, may be a risk factor in the development of T1D (Merriman, 2009). The administration of the active form of vitamin D (1, 25-dihydroxy vitamin D3) was shown to reduce the incidence of diabetes and insulitis in NOD mice. The 1, 25-dihydroxy vitamin D3 was also shown to modify T-cell differentiation, influence dendritic cell maturation toward the direction of a tolerogenic cell and protect human pancreatic islets against cytokine-induce apoptosis (Mathieu & Badenhoop 2005; Merriman, 2009). Several epidemiological studies have shown the protective effect of the active form of vitamin D against T1D (Mathieu & Badenhoop, 2005). Interestingly, the consumption of dairy products (including yogurt, cream, milk powder, as well as whole and skim milk) containing A1 β-casein variant has also been associated with latitude. This raises the question about whether the observed association between A1 β-casein consumption and incidence of T1D, as  18 suggested by the epidemiological data, is confounded with latitude and vitamin D deficiency (Merriman, 2009).   Bovine Serum Albumin (BSA)    Bovine serum albumin is a protein in the whey fraction of cow’s milk which has also been suggested as a triggering agent in the development of T1D. This hypothesis was based on the observation of increased levels of immunoglobulin G (IgG) antibodies against BSA (Karjalainen et al., 1992; Hilger et al., 2001). Experimental studies have suggested that the anti-BSA antibodies seemed to be targeted against a segment of the milk protein called ABBOS, which includes the amino acid residues 152-168 (Persaud & Barranco-Mendoza, 2004). The BSA hypothesis rests on epitope mimicry between the ABBOS fragment and a 69 kDa protein present on the surface of the islet β-cells, known as p69, which induces an autoimmune response causing the destruction of pancreatic β-cells (Alting et al., 1999; Persaud & Barranco-Mendoza, 2004). According to this hypothesis, the introduction of cow’s milk before three months of age, the age at which it is thought that the immune system is able to make the distinction between p69 and BSA, is linked to an increased risk of T1D (Persaud & Barranco-Mendoza, 2004).   This theory has been criticized because antibodies and abnormal T-cell responses to BSA in children with T1D have not been observed in all studies (Alting et al., 1999). Moreover, inconsistencies were found in the amino acid sequence reported as the ABBOS segment. Originally, Karjalainen et al. (1992) identified 17 amino acid residues in the 152-168 region of BSA as the ABBOS section (i.e. 152Glu-Leu-Leu-Tyr-Ala-Asn-Lys-Tyr-Asn-Gly-Val-Phe-Gln-Glu-Cys-Cys-Gln168), whereas Robinson et al. (1993) identified the ABBOS fragment as the 18 amino acids in position 128-145 (i.e. 128Ala-Asp-Glu-Lys-Lys- 19 Phe-Trp-Gly-Lys-Tyr-Leu-Tyr-Glu-Ile-Ala-Arg-Arg-His145) (Figure 1). However, the BSA amino acid sequence was updated following the discovery of a missing tyrosine residue in position 156 and the reversed order of the amino acids in positions 94 and 95 (Hirayama et al., 1990). Thus, the ABBOS peptide reported by Karjalainen et al. (1992) should have contained 18 amino acid residues with the sequence 152Glu-Leu-Leu-Tyr-Tyr-Ala-Asn-Lys-Tyr-Asn-Gly-Val-Phe-Gln-Glu-Cys-Cys-Gln169.  In order to trigger an anti-ABBOS immune response, the ABBOS peptide must remain intact in the gastrointestinal tract. As pointed out by Persaud & Barranco-Mendoza (2004), if the ABBOS peptide corresponds to the amino acids in positions 128-145 it is most likely that the peptide would be broken down by the enzyme trypsin since this enzyme acts on the carboxyl side of Lys and Arg residues and the region 128-145 contains many of these amino acids. Nevertheless, Alting et al. (1999) have found that an increase of the pH of the pepsin-catalysed hydrolysis from 2 to 4 (corresponding to the condition prevailing in adult and young infant stomachs, respectively) significantly reduces the efficiency of the hydrolysis. Thus, it is possible that the ABBOS region of BSA remains intact in the gastrointestinal tract of very young infants because of the relatively high pH of the gut early in life and due to limited accessibility of the ABBOS peptide to proteolytic enzymes.   Three regions of similarity between human p69 and BSA were identified from the results reported by Pietropaolo et al. (1993) and Miyazaki et al. (1994); these homologous sequences were observed in the regions 127-132, 174-179 and 373-380 (Figure 2).   Surprisingly, the regions of similarity between p69 and BSA do not correspond with the ABBOS region (152-169) suggested by Karjalainen et al. (1992), but one region does overlap with the ABBOS fragment (128-145) reported by Robinson et al. (1993).    Human serum albumin also shows some similarity to p69 and BSA, which means, according to the theory of cross reactivity and molecular mimicry proposed by Karjalainen et  20 al. (1992), that antibodies against human serum albumin should also be produced. This complicates the theory linking BSA and T1D, and raises several questions, such as the level of similarity required in the sequence of p69 and BSA to trigger destruction of β-cells.  Anti-BSA antibodies have also been found in individuals presenting other autoimmune diseases, such as rheumatoid arthritis, which suggests that BSA may simply trigger a general immune response and that the presence of anti-BSA antibodies in patients with T1D may not be a β-cell specific immunity, but merely the reflection of a flawed process of immunologic tolerance associated with a predisposition to autoimmunity (Atkinson et al., 1993).   Cross Reactivity Between Bovine and Human Insulin    Insulin is believed to be the primary autoantigen occurring in T1D (Vaarala, 2002) and to play a major role in β-cell destruction as it is the sole β-cell specific autoantigen in T1D (Harrison & Honeyman, 1999; Vaarala, 2002). Thus, the occurrence of autoantibodies to insulin (IAA) and islet cell antibodies (ICAs) is often used as a risk marker for T1D, as these autoantibodies are related to the impaired secretion of insulin by β-cells and usually precede the development of T1D. Studies have shown that the frequency of IAA varies from 20 to 60% in individuals with newly diagnosed T1D and from 1 to 10% in healthy subjects (Vaarala et al., 1998). A prospective cohort study on offspring of parents with T1D has shown that IAA appeared earlier and were detected more frequently than other autoantibodies (Ziegler et al., 1999), suggesting that immunization to insulin may be one of the earliest occurrences in the development of an autoimmune process against β-cells in individuals.    Thus, an alternative view of the role of cow’s milk in T1D, other than mechanisms attributed to -casein A1 variant and bovine serum albumin, is the possibility that bovine  21 insulin may create cross-reactive immunity to human insulin (Vaarala et al., 1998). Bovine insulin differs from human insulin by only three amino acid residues, the difference being observed at position 8 and 10 in the A-chain and position 30 in the B-chain (Vaarala et al., 1998), and is reported to be immunogenic in humans (Reeves & Kelly, 1982). Vaarala et al. (1998) have demonstrated orally-induced immunization to bovine insulin. They measured IgG antibodies to human and bovine insulin in children fed with formula containing whole cow’s milk protein, formula containing hydrolyzed casein or breastfed during the first months of their life. They observed that, at 6 months of age, the infants who received cow’s milk formula presented higher levels of IgG antibodies to bovine insulin compared to those who received hydrolysed casein peptides formula or who were exclusively breastfed. Throughout the follow-up, a decrease of insulin-binding antibodies and T cell responses seemed to occur in most of the infants, reflecting tolerance to the dietary antigens. On the other hand, bovine insulin-binding antibodies identified by enzyme immunoassay were shown to increase in children who developed diabetes-associated autoantibodies during a follow-up at 18 months from birth. These observations suggest that the insulin-specific immune response caused by dietary insulin in early infancy may not be regulated in infants who are susceptible to developing autoimmunity (Vaarala et al., 1998). Thus, the high incidence of insulin-binding antibodies in infants with T1D may be the result of oral immunization to bovine insulin contained in cow’s milk formula.  Implication of the Gut Immune System in the Development of T1D – Are Dairy Proteins the Only Triggering Agents?    There is increasing evidence, in studies on rodents as well as on human beings, that the gut-associated immune system is a keystone in the development of autoimmune diseases,  22 most likely due to disturbed oral tolerance mechanisms. Oral tolerance, which depends on the normal maturation and immunological homeostasis of the gut, may be influenced by several factors, such as gut flora, infection, growth factors and cytokines provided by breast milk, as well as diet (Figure 3). All these factors may interfere with gut immune function and lead to immunization by oral antigens instead of tolerance (Harrison & Honeyman, 1999; Kolb & Pozzilli, 1999; Wasmuth & Kolb, 2000; Vaarala, 2002).   Epidemiological studies, such as the Finnish birth cohort study, have suggested that short-term breastfeeding and the early introduction of cow’s milk-based infant formula was associated with the risk of developing β-cell autoimmunity in young children presenting a genetic predisposition to T1D (Kimpimäki et al., 2001). These observations are consistent with the hypothesis presented more than two decades earlier by Borch-Johnsen et al. (1984), from their investigation of the relation between breast-feeding and incidence rates of IDDM of children in three Scandinavian countries, that it is not the introduction of cow’s milk into the diet per se, but the lack of breast milk that may have detrimental consequences for at-risk individuals. Breast milk is known to contain a variety of growth factors and cytokines, which are believed to play a role in the maturation of the gut-associated lymphoid tissue (Srivastava et al., 1996), whereas these compounds are almost nonexistent in cow’s milk-based infant formulae (Kolb & Pozzilli, 1999). It is hypothesized that a lack of these factors may hinder the development of mucosa-mediated tolerance to islet autoantigens, including insulin present in breast milk or cross-reactive epitopes (Harrison & Honeyman, 1999). Moreover, a lack of passively-transferred maternal immunity is believed to predispose infants to enteric infections. It is possible that enteroviral infections, by increasing intestinal permeability or by disturbing mucosal immunity, may contribute to an increase in the exposure to dietary antigens and a higher incidence of autoimmune reactions (Harrison & Honeyman, 1999; Wasmuth & Kolb, 2000).   23 Newborn babies, especially preterm infants, present an increased intestinal permeability, and this can result in unwanted large molecules and infectious agents to enter the bloodstream and trigger the development of inflammation, infection and systemic hypersensitivity (van Elburg et al., 2003; Colomé et al., 2007). Several factors, such as gestational and postnatal age, as well as the feeding method have been proposed to have an impact on the maturation of the small intestine mucosa, leading to a decrease in gut permeability (van Elburg et al., 2003; Colomé et al., 2007). Maturation of the intestinal mucosa is required to promote normal immunity and tolerance (Insoft et al., 1996). An increased intestinal permeability has been suggested to play a role in the development of several autoimmune diseases such as T1D by enhancing immune activation and thus triggering autoimmunity (Vaarala, 2008; Visser et al., 2009).    Cow’s milk proteins are usually the first dietary antigen encountered by the body, but other dietary proteins may also lead to an increased immunity in individuals predisposed to T1D (Harrison & Honeyman, 1999; Wasmuth & Kolb, 2000). Animal studies in which NOD mice and diabetes-prone BioBreeding (BBdp) rats were fed with different dietary proteins have shown that the source of protein has a major impact on the development of T1D and that not only milk, but also soy and wheat may act as promoting agents in the development of T1D (Scott, 1996; Lefebvre et al., 2006). In fact, a blind, multi-centre (New Zealand, Canada and United Kingdom) study showed that a wheat-based, milk free-diet resulted in the highest diabetes incidence in NOD mice and BBdp rats (Beales et al., 2002). This finding supports previous studies reporting a higher occurrence of spontaneous diabetes in animals fed with plant-based diets presenting a high content of wheat (Scott, 1996; MacFarlane et al., 2003).   The role of wheat proteins in the pathogenesis of T1D has also been supported by several prospective studies. Moreover, the timing of initial exposure to dietary proteins during infancy may play a role on islet autoimmunity risk in predisposed children. The BABYDIAB  24 study, a cohort study conducted from 1989 to 2003 following 1610 newborn children of parents with T1D, showed that the early introduction (before 3 months of age) of gluten into the diet was associated with an increased risk of development of islet autoantibodies. However, no increased risk of islet autoantibodies was observed when gluten was introduced after the age of 6 months (Ziegler et al., 2003). On the other hand, the DAISY birth cohort study, in which 1183 children at increased risk of T1D were followed for 9 months to 9 years, showed that infants exposed to cereals between 0 and 3 months or after the age of 7 months presented an increased risk of developing islet autoantibodies (Norris et al., 2003). The consumption of soy milk formula was also found to be associated with an increased risk of T1D in Chinese infants in a prospective study by Strotmeyer et al. (2004). In fact, the exposure to soy milk formula from 4 to 6 months of age was shown to result in a doubling of the risk of T1D whereas the consumption of soy milk formula from 7 to 12 months of age was associated with a 50% higher risk (OR = 2.0, 95% CI 1.1-3.4; P 0.02 and OR = 1.5, 95% CI 1.0-2.1; P 0.05, respectively).   These results reinforce the belief that individuals predisposed to T1D may have an impaired mucosa-mediated tolerance and thus present an enhanced autoimmunity to dietary proteins in general and not specifically to dairy proteins.    DAIRY FOOD INTAKE AND TYPE 2 DIABETES: POSSIBLE ASSOCIATIONS   It is known that diet plays an important role in the prevention and management of metabolic syndrome, which is a combination of various metabolic disturbances, including abdominal obesity, insulin resistance, hypertension, impaired endothelial and haemostatic  25 function and atherogenic dyslipidaemia, that strongly increase the risk of developing T2D and cardiovascular disease (Mensink, 2006; van Meijl et al., 2008). Over the past decade, several cohort studies have investigated the relationship between dairy food consumption and the risk of the metabolic syndrome and/or T2D, by analyzing either the dairy food intake per se or dietary patterns that included dairy products and the incidence of metabolic syndrome and/or T2D in the study population. Overall, as presented in the following sections, these studies have indicated that the consumption of dairy foods is associated with a reduced risk of T2D. The mechanisms underlying this association are still unclear, but various dairy components, including the protein and fat fractions as well as the content of vitamin D, calcium and other minerals, have been suggested to contribute to the perceived beneficial effects of dairy foods on reducing the risk of developing T2D and on blood glucose regulation.   DATA FROM OBSERVATIONAL STUDIES - ASSOCIATION BETWEEN DAIRY FOOD INTAKE AND TYPE 2 DIABETES   A growing body of evidence from observational studies has suggested, with a few exceptions, an inverse association between the consumption of dairy food and the metabolic syndrome (Elwood et al., 2008; Tremblay & Gilbert, 2009). The type of dairy foods and their fat and nutrient contents are some of the main factors suggested to have an impact on the strength of the association. On the other hand, only a few studies have looked into the relationship between dairy consumption and the incidence of T2D (Table 5); thus, this relationship is not as clearly defined as it is for the metabolic syndrome. Elwood et al. (2007) examined the association between dairy or milk intake and the metabolic syndrome in 2,375 men aged 45-59 years. In this study, the clinical importance of the syndrome on incident ischemic stroke, heart disease and diabetes was investigated by a  26 follow-up examination of the men at 5-year intervals over about 20 years. Food consumption at baseline, including specifically milk and dairy foods, was assessed for all the men in the study using a semi-quantitative food frequency questionnaire (FFQ). In addition, a representative sample of 640 men kept a 7-day weighed dietary intake (WDI) record.  In this sample, 89 men had the metabolic syndrome at baseline, and 41 developed diabetes during the follow-up period. The incidence of diabetes was significantly associated with criteria related to the metabolic syndrome, a relative risk (RR) of 4.09 (95% CI 2.92-5.74) being observed in men presenting two or more criteria.  Inverse relationships were observed for the consumption of both milk and dairy products and the metabolic syndrome at baseline (Elwood et al., 2007). Compared to men who affirmed drinking little or no milk as reported on the FFQ, men who consumed one pint or more of milk per day had an adjusted odds ratio for the metabolic syndrome of 0.38 (95% CI 0.18- 0.78; Ptrend 0.002). Similar findings were obtained for the association between the metabolic syndrome and milk intake estimated from the WDI. Moreover, when considering the total dairy intake (defined as milk, cheese, yogurt, butter and cream), an odds ratio for the syndrome of 0.44 (95% CI 0.21-0.91; Ptrend 0.023) was found in the quarter of men with the highest dairy food intake based on WDI. The data of the WDI also suggested a lower incidence of T2D in men with the highest quarter of milk intake. However, this association was not significant and, as pointed out by the authors, it was not possible to draw a clear conclusion on the association between milk intake and T2D due to the limited number of men in the WDI sample.   Data from two prospective cohort studies that have reported the incidence of T2D in relation to dairy consumption have shown a decrease in T2D incidence with each serving-per-day increase in dairy intake (Choi et al., 2005 & Liu et al., 2006). The first study by Choi et al. (2005) prospectively examined the relationship between dairy consumption and incidence  27 of T2D in 41,254 males who were aged 40-75 years when first recruited for the Health Professionals Follow-up Study. Using the data accumulated over 12 years, the relative risk of T2D for men in the top quintile of total dairy intake (i.e. 4.1 servings/day) was 0.77 (95% CI 0.62-0.95; Ptrend 0.003) in comparison with those in the lowest quintile (i.e. 0.5 serving/day). The RR of T2D determined for low-fat dairy consumption was 0.73 (95% CI 0.59-0.89; Ptrend 0.001) and for high-fat dairy foods was 0.97 (95% CI 0.78-1.21; Ptrend 0.78). Each serving-per-day increase in total dairy intake was associated with a 9% lower risk for T2D in the men in this study.  Similar trends were reported by Liu et al. (2006) for 37,183 middle aged and older women from the Women’s Health Study who were followed for about 10 years. After adjusting for potential confounding variables, they reported a RR for T2D of 0.79 (95% CI 0.67-0.94; Ptrend 0.007) in the highest quintile (i.e. >2.9 servings/day) compared with women in the lowest quintile (i.e. <0.85 serving/day). A 4% lower risk of T2D was associated with each serving-per-day increase in dairy intake. The relationship between dairy consumption and the incidence of T2D did not change after adjustment for dietary calcium, vitamin D, glycemic load, fat and fibre as well as magnesium intake (RR = 0.68; 95% CI 0.52-0.89; Ptrend 0.006). The inverse association was shown to be essentially due to the consumption of low-fat dairy foods, the RRs comparing the highest to the lowest quintiles being 0.79 (95% CI 0.67-0.93; Ptrend 0.002) and 1.05 (95% CI 0.88-1.24; Ptrend 0.44) for low-fat and high-fat foods,  respectively.   In these two studies of male and female health professionals (Choi et al., 2005 & Liu et al., 2006, respectively), the negative relationship between dairy food intake and the incidence of T2D seemed to be mainly resulting from low-fat dairy products, with a lower relative risk of developing T2D being associated with intake of either total or low-fat dairy foods (skim/low-fat milk, yogurt, cottage/ricotta cheese and/or sherbet), but not high-fat dairy  28 foods (whole milk, cream, sour cream, butter, ice cream, cream cheese and/or other cheese). Moreover, in both studies the relationship was not significantly affected by BMI (i.e. <25 vs ≥25 kg/m2). The reason for the weak or lack of association with high-fat dairy foods was not explained in either study; however, the presence of saturated fat in dairy foods was speculated by Liu et al. (2006) to be a potential factor in negating the benefits provided by the insulinotropic properties of milk proteins.   Findings from some human studies have suggested that higher intake of vitamin D (Isaia et al., 2001; Pittas et al., 2006), calcium (Pittas et al., 2006; van Dam et al., 2006), magnesium (Lopez-Ridaura et al., 2004; van Dam et al., 2006) and their major food sources, such as dairy products, may help to lower the risk of T2D. These findings propose that altered magnesium, calcium and vitamin D homeostasis may have an impact on the development of T2D. van Dam et al. (2006) conducted an 8-year Black Women’s Health Study on the association between the incidence of T2D and the magnesium and calcium intakes of 41,186 African-American women aged 21 to 69 years. They found a hazard ratio (HR) of T2D of 0.69 (95% CI 0.59-0.81; Ptrend <0.0001) for dietary magnesium and 0.86 (95% CI 0.74-1.00; Ptrend 0.01) for dietary calcium for the highest (calcium: 661 mg/day; magnesium: 244 mg/day) compared with the lowest quintile (calcium: 219 mg/day; magnesium: 115 mg/day) intakes. When calcium and magnesium intakes were mutually adjusted, the relation for calcium was no longer observed (HR = 1.04; 95% CI 0.88-1.24; Ptrend 0.88), but the association remained for magnesium. A daily consumption of low-fat dairy products was associated with a lower risk of T2D in comparison with the consumption of less than one serving per week (HR = 0.87; 95% CI 0.76-1.00; Ptrend 0.04). Pittas et al. (2006) have also studied the association between vitamin D and calcium intakes and the risk of T2D. They followed 83,779 women (aged 30 to 55 years) from the Nurses’ Health Study for 20 years. After adjustment for the various potential confounders, a 33% lower risk of T2D was  29 observed for women with combined daily intakes of  >1,200 mg calcium and >800 IU vitamin D with RR of 0.67 (95% CI 0.49-0.90), compared to those presenting combined intakes of ≤600 mg of calcium and ≤400 IU of vitamin D. Since dairy foods represent the major dietary source of vitamin D and calcium in the United States where the study was carried out, the authors also investigated, in a separate analysis, the relation between dairy intake and T2D. They found that women who consumed three or more servings per day, compared to those who consumed less than one serving per day of dairy products, had 11% lower risk of T2D (RR = 0.89; 95% CI 0.81-0.99; Ptrend 0.008). However, the strength of this association was weakened (RR = 0.94; 95% CI 0.83-1.06; Ptrend 0.12) after adjustment for total vitamin D and calcium intake.  In a recent prospective study by Kirii et al. (2009), in which 59,796 middle-aged and older women and men of the Japan Public Health Center-based Prospective Study were followed for a period of 5 years, calcium and vitamin D alone were not significantly associated with T2D in either men or women. However, among subjects who consumed a greater amount of vitamin D, intake of calcium was inversely associated with the incidence of diabetes (OR = 0.62; 95% CI 0.41-0.94; Ptrend 0.050 and OR = 0.59; 95% CI 0.38-0.91; Ptrend 0.043, for the highest versus the lowest quartiles of calcium intake in men and women, respectively). This finding suggests that these two nutrients may act in conjunction, and not independently, in reducing the risk of T2D. Moreover, a significant inverse relationship between total dairy food intake and the risk of T2D was found in women (OR = 0.65; 95% CI 0.49-0.88; Ptrend 0.007, for the highest versus the lowest intake). However, the strength of this association was weakened after adjustment for several confounding factors including body mass index, total energy and family history of diabetes (OR = 0.71; 95% CI 0.51-0.98; Ptrend 0.054, for the highest versus the lowest intake). No association between dairy consumption and the incidence of diabetes was observed in men.    30  DATA FROM OBSERVATIONAL STUDIES - ASSOCIATION BETWEEN DIETARY PATTERNS THAT INCLUDE DAIRY PRODUCTS AND TYPE 2 DIABETES   Although cohort studies that investigate the relationship between the intake of dairy products and the incidence of T2D usually make adjustments for different variables such as the age of the participants, total energy intake or BMI, they fail to take into account the effect of multiple, concurrent dietary exposures (Hoffman et al., 2004; Nettleton et al., 2008). A possible way to consider the combined effect of food intakes on the development of a disease is the use of dietary patterns. Table 6 summarizes several of the recent prospective cohort studies that have investigated the relationship between dietary patterns and the risk of T2D. Even if clear conclusions on the specific effects of dairy foods per se on the incidence of diabetes are harder to draw with dietary pattern based studies, overall an inverse relationship has been observed between the incidence of T2D and dietary patterns characterized by high intake in low-fat dairy products, whole grain, fruit and vegetables (Nettleton et al., 2008).  In a recent cohort study, Erber et al. (2010) investigated the effects of three dietary patterns, “Fat and Meat”, “Vegetables”, and “Fruit and Milk” on the risk of diabetes in 75,512 Caucasians, Japanese American and native Hawaiians of the Multiethnic Cohort (MEC). The “Fruit and milk” pattern, including food groups such as milk, yogurt, cheese, citrus fruits, melons and berries, was found to lower the risk of diabetes in women (HR = 0.85; 95% CI 0.76-0.96; Ptrend 0.005), whereas the beneficial effect was less pronounced in men (HR = 0.92; 95% CI 0.83-1.02; Ptrend 0.04). Similarly, Villegas et al. (2010), who prospectively examined the association between dietary patterns and the incidence of T2D in 64,191 middle-aged women (40-70 years) from the Shanghai Women’s Health Study, found that the participants presenting the highest intake of dairy milk and the lowest intake of staple foods had a lower  31 risk of T2D, with a RR of 0.78 (95% CI 0.71-0.86) compared to participants presenting the highest intake in staples and the lowest intake in dairy milk.  Moreover, Nettleton et al. (2008) reported that dietary patterns characterized by high intake of whole grains, fruit, nuts/seeds, green leafy vegetables and low-fat dairy foods (a pattern derived empirically by principal component analysis and referred to as “whole grains and fruit”) was associated with a 15% reduction of the risk of T2D (HR = 0.85; 95% CI 0.76-0.95; Ptrend 0.005) in 5,011 participants from the Multi-Ethnic Study of Atherosclerosis (MESA) including white, black, Hispanic and Chinese adults aged 45 to 84 years. An inverse association between the incidence of diabetes and a dietary pattern defined a priori by the authors as “low-risk food pattern”, based on the intake of foods previously reported to be associated with T2D, was also observed (HR = 0.87; 95% CI 0.81-0.99; Ptrend 0.04). On the other hand, dietary patterns presenting a high intake of refined grains, red meat, tomatoes, beans and high-fat dairy foods was linked to a 18% higher risk of T2D (HR = 1.18; 95% CI 01.06-1.32; Ptrend 0.004).  A “healthy” eating pattern, characterized by a high intake of fruit, vegetables, whole-meal bread, low-fat dairy and little alcohol, was reported to reduce the risk of diabetes during 15 years of follow-up in the Whitehall II study (“healthy” vs “unhealthy” patterns, HR = 0.74; 95% CI 0.58-0.94; Ptrend 0.016) (Brunner et al., 2008). The impact of the adherence to the Dietary Approaches to Stop Hypertension (DASH), a diet rich in fruits, vegetables and low-fat dairy products, on the risk of T2D has also been investigated by Liese et al. (2009) in 862 black, Hispanic and non-Hispanic white participants of the Insulin Resistance Atherosclerosis Study (IRAS). An inverse correlation between the adherence to the DASH diet and the incidence of T2D in whites was observed (tertile 3 vs tertile 1: OR = 0.31; 95% CI 0.13-0.75; Ptrend 0.03), but not in Hispanics or blacks (tertile 3 vs tertile 1: OR = 1.34; 95% CI 0.70-2.58;  32 Ptrend 0.68). The lack of association in blacks and Hispanics, as pointed out by the authors, could be due to the limited number of participants.    Elwood et al. (2010) performed a meta-analysis of prospective cohort studies investigating the relationship between dairy food consumption and the incidence of T2D. Of the five studies included in the meta-analysis, one study looked at dietary patterns while the other four looked specifically at the intake of dairy products. Individuals with the highest consumption of dairy food, compared to those with the lowest intake, were found to present a lower risk of diabetes (RR = 0.85; 95% CI 0.75-0.96; P for homogeneity 0.122). This finding is in accordance with most of the results from prospective studies that seem to show a consistent inverse relationship between the incidence of T2D and the consumption of dairy foods, particularly low-fat dairy products.      DAIRY CONSUMPTION AND TYPE 2 DIABETES: POSSIBLE MECHANISMS   While the incidence of T2D is greatly influenced by genetic predisposition and lifestyle, dietary composition may play an important role in its development and its complications (Risérus et al., 2009). As presented in the following sections, proteins as well as several non-protein constituents of milk have been suggested to contribute to the beneficial effect of dairy food consumption on the incidence of T2D.     33 Effects of Non-Protein Constituents in Dairy Products on the Risk of Type 2 Diabetes and Glucose Regulation    Lactose is the second major milk constituent after water (Lindmark Månsson, 2008). However, despite its high content in cow’s milk, the current knowledge on the possible roles of dairy components in the incidence of T2D does not seem to suggest lactose as a dominant factor in the inverse association between dairy food consumption and diabetes. In fact, lactose intake was not found to be significantly related to T2D (RR = 0.99; 95% CI 0.80-1.22; Ptrend 0.33) in a prospective study by Janket et al. (2003), in which the intake of total and specific types of sugar, including lactose, was investigated in 38,480 women of The Women’s Heath Study. Lactose, dairy protein and fat have been suggested to enhance satiety and lower the risk of overweight and obesity, known to be major risk factors for T2D, compared with other high-carbohydrate foods and beverages (Pereira et al., 2002). However, in a study by Pereira et al. (2002), adjusting for these nutrients had no significant effects on the inverse relationship they observed between dairy food intake and the insulin resistance syndrome Cow’s milk is known to be a major food source of calcium, magnesium as well as vitamin D. Vitamin D deficiency has been shown, in animal and human models, to increase peripheral tissue insulin resistance and decrease insulin secretion (Isaia et al., 2001; Moreira & Hamadeh, 2010). In fact, a positive correlation between 25-hydroxyvitamin D (25(OH)D) concentration and insulin sensitivity was observed in glucose-tolerant individuals, whereas hypovitaminosis D was found to have negative effect on pancreatic -cell function (Chiu et al., 2004). Magnesium deficiency has also been found to have adverse effects on insulin action (Lopez-Ridaura et al., 2004; Martini et al., 2010). Hypomagnesemia may alter tyrosine-kinase activity at the insulin receptor level and may lead to increased intracellular calcium concentration, and these two effects can result in impaired insulin action (Martini et  34 al., 2010). On the other hand, calcium is needed for insulin-mediated intracellular processes and changes in calcium concentration in insulin target tissues may affect the insulin signal transduction and cause peripheral insulin resistance (Pittas et al., 2007).  Although vitamin D, magnesium and calcium may play a role in glucose regulation, the findings of their effects on the risk of T2D are conflicting and the significance of their possible contribution to the inverse association observed between dairy product intake and diabetes remains unclear. Total vitamin D and total calcium intake were significantly associated with the risk of T2D in a prospective study by Pittas et al. (2006). These associations, however, were found when these nutrients were obtained from supplement, rather than from dietary sources (Pittas et al., 2006). On the other hand, calcium and vitamin D intake alone were not linked to the risk of T2D in a further study by Kirii et al. (2009), and a positive association between calcium and T2D was observed only in individuals with a higher vitamin D intake. Yet, calcium plus vitamin D supplementation was not shown to reduce the risk of diabetes in a randomized placebo-controlled trial (de Boer et al., 2008). Calcium and vitamin D intake were not found either to explain the inverse association between dairy consumption and the insulin resistance syndrome observed in overweight adults of The Coronary Artery Risk Development in Young Adults (CARDIA) study (Pereira et al., 2002). Vitamin D is mainly produced endogenously by the skin during exposure to the sunlight, and diet only accounts for approximately 30% of the vitamin D obtained (Moreira and Hamadeh, 2010). Thus, dietary intake of this vitamin alone may not provide an accurate indication of an individual’s overall vitamin D status.         The impact of dietary fat on the incidence of T2D has been of particular interest due to the role that fatty acids play on glucose regulation by affecting cell membrane function, insulin signaling, genetic expression as well as enzymatic activity (Risérus et al., 2009). Several factors may affect bovine milk fatty acid composition, including the breed, the type of  35 feeding and the season (Pešek et al., 2005; Lindmark Månsson, 2008), but in general bovine milk contains approximately 4.2% of fat, of which almost 70% is saturated whereas mono-unsaturated and poly-unsaturated fatty acids account for about 25% and 2.3%, respectively (Lindmark Månsson, 2008). Moreover, approximately 7-9% of milk fat is composed of medium-chain fatty acids (MCFAs) (Pešek et al., 2005), which present a chain length of 8 to 12 carbons, and about 3-7% of short-chain fatty acids, which present a chain length of 4 to 6 carbons (Pešek et al., 2005; Lindmark Månsson, 2008). A diet characterized by a high intake of saturated fat disturbed insulin sensitivity (Pfeuffer & Schrezenmeir, 2006; Wein et al., 2009). On the other hand, MCFAs have been reported to improve insulin sensitivity in some animal and human studies (Pfeuffer & Schrezenmeir, 2002; Pfeuffer & Schrezenmeir, 2006; Wein et al., 2009). Unlike long chain fatty acids, MCFAs do not require binding to fatty acid-binding proteins for their transport within cells and across membranes and this distinction impacts numerous regulatory pathways. However, the effect of MCFAs on insulin level or plasma glucose observed in most human or animal studies is not so clear-cut (Pfeuffer & Schrezenmeir, 2006).  Another type of fatty acid naturally found in cow’s milk, trans-palmitoleic acids, was recently suggested to have a preventive effect on the risk of diabetes (Mozaffarian et al., 2010). Mozaffarian et al. (2010) examined the association between blood trans-palmitoleate level and the incidence of T2D in 3,736 subjects of the Cardiovascular Health Study. Higher circulating levels of trans-palmitoleate were associated with significantly lower risk of T2D (HR 0.41, 95% CI 0.27-0.64 [quintiles 4 versus 1] and HR 0.38, 95% CI 0.24-0.62 [quintiles 5 versus 1]). The authors speculated that trans-palmitoleate could partly mimic the biological function of adipose-produced cis-palmitoleate which have been found in animal experiments to improve insulin resistance and related metabolic abnormalities in hepatic and skeletal muscle in addition to help in suppressing hepatic fat production. Increased hepatic fat  36 production has been suggested to contribute to nonalcoholic steatohepatitis, a condition associated with insulin resistance (Mozaffarian et al., 2010).  In order to assess the role of specific fatty acids in the pathogenesis of T2D, Krachler et al. (2008) investigated the erythrocyte membrane fatty acid (EMFA) composition, which provides an estimation of the fatty acid intake, of 159 individuals with T2D and 291 referents. Higher levels of pentadecanoic (15:0) and heptadecanoic (17:0) in erythrocyte membrane, which were found to be positively correlated with the intake of high-fat dairy products such as butter and 3% fat milk, were associated with a lower risk of T2D (OR = 0.65; 95% CI 0.50-0.85; P 0.002 and OR = 0.47; 95% CI 0.35-0.63; P <0.001, respectively) (Krachler et al., 2008). Kröger et al. (2011) also reported a strong link between the incidence of T2D and the fatty acid profile of erythrocyte membrane phospholipids and the activity of desaturase enzymes in 2724 subjects, including 673 diabetes cases, of The European Prospective Investigation into Cancer and Nutrition (EPIC)-Potsdam Study. However, dietary fatty acid profile was found to present only a modest or low correlation with EMFA composition and to not be significantly related to the risk of diabetes (Kröger et al., 2011).  Although milk fat may have some effects on the incidence of T2D, it seems unlikely that it is the primary mechanism for the inverse association between diabetes and dairy product consumption observed in epidemiological studies, as the protective effect of dairy has been more strongly linked to the intake of low-fat dairy foods. From the mixed results yielded by the studies that have assessed the effects of vitamin D, calcium and magnesium on the incidence of T2D, it also seems unlikely that these nutrients are the main contributors to the beneficial effect of dairy product consumption on the risk of T2D reported in observational studies. Moreover, even if there are evidences of possible associations between dairy fat, vitamin D, calcium, magnesium and the incidence of T2D, assessing whether the reported effects are specific to the risk of developing diabetes is complex because these components  37 have also been suggested to have anti-obesity properties and/or help in lowering the risk of the metabolic syndrome, which are both in turn associated with the risk of developing T2D (Pfeuffer & Schrezenmeir, 2006).        Effects of Milk Casein and Whey Proteins on Glucose Regulation   The possible effect of dairy foods on the incidence of T2D has also been suggested to be attributable to the protein fraction of milk. Dietary proteins are found to cause the release of insulin in both healthy and diabetic individuals when ingested with or without carbohydrate. Moreover, dietary proteins are also known to induce glucagon secretion (Claessens et al., 2008).  Milk protein seems to induce an increase in postprandial insulin response with a reduction in postprandial blood glucose levels (Liljeberg Elmståhl & Björck, 2001; Östman et al., 2001). When compared with casein or other animal or vegetable proteins, whey proteins were found to be particularly insulinotrophic (Nilsson et al., 2004). This finding was suggested to be due to the rapid digestibility of whey proteins (Akhavan et al., 2010) and its high content of branched-chain amino acids (particularly leucine), which are known for their high capacity to increase insulin response (Peterson et al., 2009). Several amino acids can potentially act as direct insulin secretagogues (Nilsson et al., 2007), and increased levels of amino acids in postprandial plasma are believed to be associated with enhanced insulin response (Calbet & MacLean, 2002). Leucine, isoleucine, valine, lysine and threonine were among the essential amino acids showing a pronounced postprandial increase in plasma following the ingestion of whey drink and were also shown to present the strongest correlation with insulin response (Nilsson et al., 2004).    It has also been suggested that the glycemic regulatory properties of milk protein may be related to its effect on the release of incretin hormones from the gut. In fact, proteins, such  38 as whey protein, trigger the release of cholecystokinin, glucagon-like peptide-1 (GLP-1), glucose-dependent insulinotropic polypeptide (GIP) and peptide tyrosine tyrosine (PYY) from the intestinal enteroendocrine cells (Akhavan et al. 2010). These peptides play an important role in stomach emptying and glycemic control. A protein-stimulated insulin response in healthy individuals or individuals with T2D that did not mirror the increase in plasma amino acid levels was observed in some studies, supporting the role of incretin hormones in protein-stimulated insulin release (Nilsson et al., 2004).  A recent study on the effects of consumption of whey protein (WP) and whey protein hydrolysate (WPH) before a meal on satiety and on blood glucose and insulin concentrations in young adults (Akhavan et al. 2010) showed that only WP contributed to blood glucose regulation by insulin-dependent and insulin-independent mechanisms. This observation suggests that noninsulintropic mechanisms require stimulation arising from the digestion of intact proteins. Despite the fact that the mechanism of action by which premeal WP leads to postmeal glucose regulation is not well defined, it seems probable that the insulin-independent actions of premeal consumption of WP is partly due to the effect of protein on gastric emptying caused by the release of incretin hormones. A mixture of branched-chain amino acids was shown to produce the same effect as intact whey protein on insulin, but not on gut hormones (Nilsson et al., 2004). This observation, as well as the possible existence of a threshold for amino acid-induced GIP release (Nilsson et al., 2007), may explain the lack of effect of WPH on blood glucose.  Manders et al. (2006) investigated the effects on blood glucose concentration of the consumption of a beverage containing casein protein hydrolysate/leucine mixture after each main meal. They observed a reduction in the prevalence of hyperglycemia, with a significant reduction of the average 24-h blood glucose level, in T2D patients receiving the protein hydrolysate/leucine mixture compared to those receiving a placebo mixture. Similarly, the  39 addition to a glucose drink of 5-20 grams of protein in the form of a protein supplement known as “glycemic index lowering peptide” (GILP), which is composed of a blend of whey peptides and intact whey protein with high content of branched-chain amino acids, was shown to cause a dose-dependent decrease in post-prandial glycemia in 10 healthy subjects (Petersen et al., 2009). Although the mechanism by which GILP causes a decrease in the postprandial glycemia is not known, several hypotheses have been raised, such as the effect of branched chain amino acids in stimulating insulin response, the effect of whey protein on incretin hormone secretion and the effect of the whey peptide fraction as a source of inhibitors of dipeptidyl peptidase-IV (DPP-IV), a member of the prolyl endopeptidase family known to hydrolyse incretin hormones such as GLP-1 (Petersen et al., 2009). Interestingly, in animal studies showing the effect of whey protein on the reduction of postprandial glycemia, higher levels of intact GIP and GLP-1 were observed, and this despite the fact that only the secretion of GLP-1 was increased (Gunnarsson et al., 2006). It is suggested that the digestion of whey protein might generate protein fragments (di- or tripeptides) that may act as endogenous inhibitors of DPP-IV, and thus lead to decreased degradation of the incretin hormones and to higher levels of intact GIP and GLP-1 (Drucker, 2006; Gunnarsson et al., 2006).    Effect of Protein Intake on the Incidence of T2D – Is the Effect Specific to Dairy Proteins?    Amino acids and dietary proteins are known to play an important role in the modulation of glucose metabolism and insulin sensitivity (Tremblay et al., 2007). In addition to dairy foods, the consumption of several other food products, such as fish (Ouellet et al., 2007; Tremblay et al., 2007; Patel et al., 2009), soy (Lavigne et al., 2000), and nut (Jiang et al., 2002), has also been reported to have an impact on glycemia regulation.   40  It was observed in a rat study that animals fed with soy proteins and cod proteins present an improved fasting glucose tolerance and insulin sensitivity when compared to rats fed with casein proteins (Lavigne et al., 2000). However, the beneficial effect of soy protein on glycemic control has not always been observed in human studies (Gobert et al., 2010) and differences in postprandial glucose, as well as hormonal responses following the ingestion of dairy and soy proteins, have also not always been observed (Bos et al., 2003). It is well known that the incidence of T2D in individuals from regions such as Alaska and Greenland, where the consumption of fish is important, is particularly low (Tremblay et al., 2007). The beneficial effect of fish consumption was initially believed to be attributable to the fish oil, but recent studies on the consumption of lean fish, such as cod, have also shown an inverse correlation with the incidence of T2D (Tremblay et al., 2007), suggesting that fish protein might actually be the constituent responsible for the beneficial effect of lean fish consumption. In studies in which animals were administered a high-fat diet, cod protein-fed rats were shown to be protected against the development of insulin resistance in comparison with rats fed soy or casein protein (Lavigne et al., 2001). The protective effect of cod on insulin resistance is believed to be attributable to an enhanced insulin-stimulated glucose uptake in skeletal muscles (Lavigne et al., 2001).   These data suggest that dairy is not unique, and that it might be the general intake of proteins and amino acids that modulate insulin action and glucose metabolism, and thus have beneficial effects in the prevention of T2D.   CONCLUDING REMARKS   Epidemiological and observational studies have suggested a positive correlation between dairy food intake and the risk of development of T1D in genetically predisposed  41 individuals. However, it should be noted that the correlations or associations shown by ecological or epidemiological studies between cow’s milk and T1D do not necessarily indicate a cause and effect relationship. Although several hypotheses have been suggested to explain this relationship, it is not possible to conclude from the actual data available that cow’s milk per se is pathogenic in T1D. Several lines of evidence show that the gut-associated immune system is a keystone in the development of autoimmune diseases and that individuals predisposed to T1D may simply present an enhanced autoimmunity to dietary proteins in general and not specifically to dairy protein.  On the other hand, observational studies have also suggested that the consumption of dairy products may reduce the incidence of T2D. While several milk constituents may contribute to the beneficial effect of dairy food intake on lowering the risk of developing T2D, the protein fraction of milk may play a dominant role in the observed effects. However, the exact mechanism by which dairy proteins participate in insulin/glucose regulation is still unclear. The consumption of other dietary proteins, such as fish proteins, has also been shown to be inversely correlated to the development of metabolic syndrome and T2D, suggesting that the effect of protein consumption on T2D might not be uniquely associated with dairy proteins.  More research is required to fully understand the physiological effects of the consumption of dairy products, as well as the role and mechanisms of specific constituents of dairy foods that may be involved in influencing the development of diabetes. However, designing studies that could provide absolute certainty whether dairy product consumption is associated with diabetes is challenging. Intervention studies may be the best approach to obtain clearer answers on the relation between cow’s milk consumption and T1D, but breastfeeding remains an important confounding factor that is impossible to completely eliminate as it cannot ethically be limited. On the other hand, nutritional interventions are  42 unlikely to be feasible for adults in which diet is less easy to control and would have to be monitored for decades in order to evaluate the effect of dairy food consumption on the risk of T2D.  Finally, although diet can play a role in the incidence of diabetes, it should be kept in mind that it is only one among many other environmental factors, including lifestyle, that may contribute to the development of the disease.  43 REFERENCES  Åkerblom, H.K., Vaarala, O., Hyöty, H., Ilonen, J., and Knip, M. (2002). Environmental factors in the etiology of type 1 diabetes. Am. J. Med. Genet. 115: 18-29.  Akhavan, T., Luhovyy, B.L., Brown, P.H., Cho, C.E., and Anderson, G.H. (2010). Effect of premeal consumption of whey protein and its hydrolysate on food intake and postmeal glycemia and insulin responses in young adults. Am. J. Clin. Nutr. 91: 966-975.  Alting, A.C., Meijer, R.J.G.M., and van Beresteijn, E.C.H. (1999). Cow’s milk proteins and insulin-dependent diabetes mellitus. Something to worry about? Nahrung 43: 209-210.  Atkinson, M.A., Bowman, M.A., Kao, K.-J., Campbell, L., Dush, P.J., Shah, S.C., Simell, O.,  and Maclaren, N.K. (1993). Lack of immune responsiveness to bovine serum albumin in insulin-dependent diabetes. New Engl. J. Med. 329: 1853-1858.  Beales, P.E., Elliott, R.B., Flohé, S., Hill, J.P., Kolb, H., Pozzilli, P., Wang, G.S., Wasmut, H., and Scott, F.W. (2002). A multi-centre, blinded international trial of the effect of A(1) and A(2) beta-casein variants on diabetes incidence in two rodent models of spontaneous Type 1 diabetes. Diabetologia 45: 1240-1246.  Bell, S.J., Grochoski, G.T., and Clarke, A.J. (2006). Health implication of milk containing β-casein with A2 genetic variant. Crit. Rev. Food Sci. Nutr. 46: 93-100.   44 Borch-Johnsen, K., Joner, G., Mandrup-Poulsen, T., Christy, M., Zachau-Christiansen, B., Kastrup, K. and Nerup, J. (1984). Relation between breast feeding and incidence rates of insulin-dependent diabetes mellitus. A Hypothesis. Lancet 324: 1083-1086.  Bos, C., Metges, C.C., Gaudichon, C., Petzke, K.J., Pueyo, M.E., Morens, C., Everwand, J., Benamouzig, R., and Tomé, D. (2003). Postprandial kinetics of dietary amino acids are the main determinant of their metabolism after soy or milk protein ingestion in humans. J. Nutr. 133: 1308-1315.  Brunner, E.J., Mosdøl, A. Witte, D.R., Martikainen, P., Stafford, M., Shipley, M.J., and Marmot, M.G. (2008). Dietary patterns and 15-y risks of major coronary events, diabetes, and mortality. Am. J. Clin. Nutr. 87: 1414-1421.  Calbet, J.A.L., and MacLean, D.A. (2002). Plasma glucagon and insulin responses depend on the rate of appearance of amino acids after ingestion of different protein solutions in humans. J. Nutr. 132: 2174-2182.  Canadian Dairy Information Center. (2010). Canada’s dairy industry at a glance. http://www.dairyinfo.gc.ca/pdf/At_a_glance_e_2009.pdf. Accessed June 25, 2010.  Canadian Diabetes Association. (2011) Prevalence and costs of diabetes. http://www.diabetes.ca/diabetes-and-you/what/facts/. Accessed January 6, 2011.  Chiu, K.C., Chu, A., Go, V.L.W., and Saad, M.F. (2004). Hypovitaminosis D is associated with insulin resistance and  cell dysfunction. Am. J. Clin. Nutr. 79: 820-825.   45 Choi, H.K., Willett, W.C., Stampfer, M.J., Rimm, E., and Hu, F.B. (2005). Dairy consumption and the risk of type 2 diabetes mellitus in men. Arch. Intern. Med. 165: 997-1003.  Chruścińska, E., Olczar, J., Zabrocki, J., Dyba, M., Micera, G., Sanna, D., and Kozłowski, H. (1998). Specific interactions of bovine and human -casomorphin-7 with Cu(II) ions. J. Inorg. Biochem. 69: 91-95.  Claessens, M., Saris, W.H.M., and van Baak, M.A. (2008). Glucagon and insulin responses after ingestion of different amounts of intact and hydrolysed proteins. Brit. J. Nutr. 100: 61-69.  Colomé, G., Sierra, C., Blasco, J., García, M.V., Valverde, E., and Sánchez, E. (2007). Intestinal permeability in different feedings in infancy. Acta Paediatrica 96: 69-72.  Couper, J.J., Steele, C., Beresford, S., Powell, T., McCaul, K., Pollard, A., Gellert, S., Tait, B., Harrison, L.C., and Colman, P.G. (1999). Lack of association between duration of breast-feeding or introduction of cow’s milk and the development of islet autoimmunity. Diabetes 48: 2145-2149.  Dahl-Jørgensen, K., Joner, G., and Hanssen, K.F. (1991). Relationship between cows’ milk consumption and incidence of IDDM in childhood. Diabetes Care 14: 1081-1083.     46 de Boer, I.H., Tinker, L.F., Connelly, S., Curb, J.D., Howard, B.V., Kestenbaum, B., Larson, J.C., Manson, J.E., Margolis, K.L., Siscovick, D.S., and Weiss, N.S. (2008). Calcium plus vitamin D supplementation and the risk of incident diabetes in the women’s health initiative. Diabetes Cares 31: 701-707.  De Noni, I., and Cattaneo, S. (2010). Occurrence of β-casomorphins 5 and 7 in commercial dairy products and their digests following in vitro simulated gastro-intestinal digestion. Food Chem. 119: 560-566.  De Noni, I., FitzGerald, R.J., Korhonen, H.J.T., Le Roux, Y., Livesey, C.T., Thorsdottir, I., Tomé, D., and Witkamp, R. (2009). Review of the potential health impact of β-casomorphins and related peptides. EFSA Science Report 231: 1-107.  Drucker, D.J. (2006). Enhancing the action of incretin hormones: A new whey forward? Endocrinology 147: 3171-3172.  Elliott, R.B., Harris, D.P., Hill, J.P., Bibby, N.J., and Wasmuth, H.E. (1999). Type 1 (insulin-dependent) diabetes mellitus and cow milk: casein variant consumption. Diabetologia 42: 292-296.  Elwood, P.C., Givens, D.I., Beswick, A.D., Fehily, A.M., Pickering, J.E., and Gallacher, J. (2008). The survival advantage of milk and dairy consumption: an overview of evidence from cohort studies of vascular diseases, diabetes and cancer. J. Am. Coll. Nutr. 27: 723S-734S.   47 Elwood, P.C., Pickering, J.E., and Fehily, A.M. (2007). Milk and dairy consumption, diabetes and the metabolic syndrome: the Caerphilly prospective study. J. Epidemiol. Commun. Health 61: 695-698.  Elwood, P.C., Pickering, J.E., Givens, D.I., and Gallacher, J.E. (2010). The consumption of milk and dairy foods and the incidence of vascular disease and diabetes: An overview of the evidence. Lipids 45: 925-939.   Erber, E., Hopping, B.N., Grandinetti, A., Park, S.-Y., Kolonel, L.N., and Maskarinec, G. (2010). Dietary patterns and risk for diabetes: The Multiethnic Cohort. Diabetes Care 33: 532-538.  Esfarjani, E., Razzahy Azar, M., and Gafarpour, M. (2001). IDDM and early exposure of infant to cow’s milk and solid food. Indian J. Pediatr. 68: 107-110.  Farrell Jr., H.M., Jimenez-Flores, R., Bleck, G.T., Brown, E.M., Butler, J.E., Creamer, L.K., Hicks, C.L., Hollar, C.M., Ng-Kwai-Hang, K.F., and Swaisgood, H.E. (2004). Nomenclature of proteins of cow’s milk-sixth revision. J. Dairy Sci. 87: 1641-1674.   Fava, D., Leslie, R.D.G., and Pozzilli, P. (1994). Relationship between dairy product consumption and incidence of IDDM in childhood in Italy. Diabetes Care 17: 1488-1490.  Gerstein, H.C. (1994). Cow's milk exposure and type I diabetes mellitus. A critical overview of the clinical literature. Diabetes Care 17: 13-19.    48 Gobert, C.P., Pipe, E.A., Capes, S.E., Darlington, G.A., Lampe, J.W., and Duncan, A.M. (2010). Soya protein does not affect glycaemic control in adults with type 2 diabetes. Brit. J. Nutr. 103: 412-421.    Gunnarsson, P.T., Winzell, M.S., Deacon, C.F., Larsen, M.O., Jelic, K., Carr, R.D., and Ahrén, B. (2006). Glucose-induced incretin hormone release and inactivation are differently modulated by oral fat and protein in mice. Endocrinology 147: 3173-3180.       Harrison, L.C., and Honeyman, M.C. (1999). Cow’s milk and type 1 diabetes: the real debate is about mucosal immune function. Diabetes 48: 1501-1507.  Hilger, C., Grigioni, F., De Beaufort, C., Michel, G., Freilinger, J., and Hentges, F. (2001). Differential binding of IgG and IgA antibodies to antigenic determinants of bovine serum albumin. Clin. Exp. Immunol. 123: 387-394.  Hirayama, K., Akashi, S., Furuya, M., and Fukuhara, K.-i. (1990). Rapid confirmation and revision of the primary structure of bovine serum albumin by ESIMS and frit-FAB LC/MS. Biochem. Biophys. Res. Commun. 173: 639-646.   Hoffmann, K., Schulze, M.B., Schienkiewitz, A., Nöthlings, U., and Boeing, H. (2004). Application of a new statistical method to derive dietary patterns in nutritional epidemiology. Am. J. Epidemiol. 159: 935-944.   49 Holmberg, H., Wahlberg, J., Vaarala, O., and Ludvigsson, J. (2007). Short duration of breast-feeding as a risk-factor for -cell autoantibodies in 5-year-old children from the general population. Brit. J. Nutr. 97: 111-116.  Insoft, R.M., Sanderson, I.R., and Walker, W.A. (1996). Development of immune function in the intestine and its role in neonatal diseases. Pediatr. Clin. North Am. 43: 551-571.  Isaia, G., Giorgino, R, and Adami, S. (2001).High prevalence of hypovitaminosis D in female type 2 diabetes population. Diabetes Care 24: 1496.  Janket, S.-J., Manson, J.E., Sesso, H., Buring, J.E., and Liu, S. (2003). A prospective study of sugar intake and risk of type 2 diabetes in women. Diabetes Care 26: 1008-1015.  Jarmolowska, B., Bielikowicz, K., Iwan, M., Sidor, K., Kostyra, E., and Kaczmarski, M. (2007). Serum activity of dipeptidyl peptidase IV (DPPIV; EC 3.4.14.5) in breast-fed infants with symptoms of allergy. Peptide 28: 678-682.   Jiang, R., Manson, J.E., Stampfer, M.J., Liu, S., Willett, W.C., and Hu, F.B. (2002). Nut and peanut butter consumption and risk of type 2 diabetes in women. J. Am. Med. Assoc. 288: 2554-2560.  Kamiński, S., Cieślińska, A., and Kostyra, E. (2007). Polymorphism of bovine beta-casein and its potential effect on human health. J. Appl. Genet. 48: 189-198.   50 Karjalainen, J., Martin, J.M., Knip, M., Ilonen, J., Robinson, B.H., Savilahti, E., Akerblom, H.K., and Dosch, H.-M. (1992). A bovine albumin peptide as a possible trigger of insulin-dependent diabetes mellitus. New Engl. J. Med. 327: 302-307.  Kimpimäki, T., Erkkola, M., Korhonen, S., Kupila, A., Virtanen, S.M., Ilonen, J., Simell, O., and Knip, M. (2001).Short-term exclusive breastfeeding predisposes young children with increased genetic risk of Type 1 diabetes to progressive beta-cell autoimmunity. Diabetologia 44: 63-69.  King, J.C. (2005) The milk debate. Arch. Intern. Med. 165: 975-976.  Kirii, K., Mizoue, T., Iso, I., Takahashi, Y., Kato, M., Inoue, M., Noda, M., and Tsugane, S. (2009). Calcium, vitamin D and dairy intake in relation to type 2 diabetes in Japanese cohort. Diabetologia 52: 2542-2550.  Knip, M., Virtanen, M., Seppä, K., Ilonen, J., Savilahti, E., Vaarala, O., Reunanen, A., Teramo, K., Hämäläinen, A.-M., Paronen, Dosch, H.-M., Hakulinen, T., and Åkerblom, H.K. (2010). Dietary intervention in infancy and later signs of beta-cell autoimmunity. N. Engl. J. Med. 363: 1900-1908.  Koch, G., Wiedemann, K., and Teschemacher, H. (1985). Opioid activities of human β-casomorphins. Naunyn-Schmiedeberg’s Arch. Pharmacol. 331: 351-354.  Kolb, H., and Pozzilli, P. (1999). Cow’s milk and type 1 diabetes: the gut immune system deserves attention. Immunol. Today 20: 108-110.  51 Krachler, B., Norberg, M., Eriksson, J.W., Hallmans, G., Johansson, I., Vessby, B., Weinehall, L., and Lindahl, B. (2008). Fatty acid profile of the erythrocyte membrane preceding development of type 2 diabetes. Nutr. Metab. Cardiovasc. Dis. 18: 503-510.   Kröger, J., Zietemann, V., Enzenbach, C., Weikert, C., Jansen, E.H.J.M., Döring, F., Joost, H.-G., Boeing, H., and Schulze, M.B. (2011). Erythrocyte membrane phospholipids fatty acids, desaturase activity, and dietary fatty acids in relation to risk of type 2 diabetes in the European Prospective Investigation into Cancer and Nutrition (EPIC)-Potsdam Study. Am. J. Clin. Nutr. 93: 127-142.  Lamb, M.M., Myers, M.A., Barriga, K., Zimmet, P.Z., Rewers, M., and Norris, J.M. (2008). Maternal diet during pregnancy and islet autoimmunity in offspring. Pediatric Diabetes 9: 135-141. DOI: 10.1111/j.1399-5448.2010.00668.x  Laugesen, M., and Elliott, R. (2003). Ischaemic heart disease, Type 1 diabetes, and cow milk A1 β-casein. N. Z. Med. J. 116: U295.     Lavigne, C., Marette, A., and Jacques, H. (2000). Cod and soy proteins compared with casein improve glucose tolerance and insulin sensitivity in rats. Am. J. Physiol. Endocrinol. Metab. 278: E491-E500.  Lavigne, C., Tremblay, F., Asselin, G., Jacques, H., and Marette, A. (2001). Prevention of skeletal muscle insulin resistance by dietary cod protein in high fat-fed rats. Am. J. Physiol. Endocrinol. Metab. 281: E62-E71.  52 Lefebvre, D.E., Powell, K.L., Strom, A., and Scott, F.W. (2006). Dietary proteins as environmental modifiers of type 1 diabetes mellitus. Ann. Rev. Nutr. 26: 175-202.  Liese, A.D., Nichols, M., Sun, X., D’Agostino, R.B. Jr., and Hafner, S.M. (2009). Adherence to the DASH diet is inversely associated with incidence of type 2 diabetes: The insulin resistance atherosclerosis study. Diabetes Care 32: 1434-1436.   Liljeberg Elmståhl, H., and Björck, I. (2001). Milk as a supplement to mixed meals may elevate postprandial insulinaemia. Eur. J. Clin. Nutr. 55: 994-999.  Lindmark Månsson, H. (2008). Fatty acids in bovine milk fat. Food & Nutrition Research 52. DOI:10.3402/fnr.v52i0.1821. Retrieved February 21, 2011, from http://www.foodandnutritionresearch.net/index.php/fnr/article/view/1821/1798  Liu, S., Choi, H.K., Ford, E., Song, Y., Klevak, A., Buring, J.E., and Manson, J.E. (2006). A prospective study of dairy intake and the risk of type 2 diabetes in women. Diabetes Care 29: 1579-1584.   Lopez-Ridaura, R., Willett, W.C., Rimm, E.B., Liu, S., Stampfer, M.J., Manson, J.E., and Hu, F.B. (2004). Magnesium intake and risk of type 2 diabetes in men and women. Diabetes Care 27: 134-140.  Ludvigsson, J. (2003). Cow-milk-free diet during the last trimester of pregnancy does not influence diabetes-related autoantibodies in nondiabetic children. Ann. N.Y. Acad. Sci. 1005: 275-278. DOI:10.1196/annals.1288.042.   53 MacFarlane, A.J., Burghardt, K.M., Kelly, J., Simell, T., Simell, O., Altosaar, I., and Scott, F.W. (2003). A type 1 diabetes-related protein from wheat (triticum aestivum) cDNA clone of a wheat storage globulin, GIb1, linked to islet damage. J. Biol. Chem. 278: 54-63.  Manders, R.J.F., Praet, S.F.E., Meex, R.C.R., Koopman, R., de Roos, A.L., Wagenmakers, A.J.M., Saris, W.H.M., and van Loon, L.J.C. (2006). Protein hydrolysate/leucine co-ingestion reduces the prevalence of hyperglycemia in type 2 diabetic patients. Diabetes Care 29: 2721-2722.  Martini, L.A., Catania, A.S., and Ferreira, S.R.G. (2010). Role of vitamins and minerals in prevention and management of type 2 diabetes mellitus. Nutr. Rev. 68: 341-354.   Mathieu, C., and Badenhoop, K. (2005). Vitamin D and type 1 diabetes mellitus: state of the art. Trends Endocrin. Met. 16: 261-266.  Mensink, R.P. (2006). Dairy products and the risk to develop type 2 diabetes or cardiovascular disease. Int. Dairy J. 16: 1001-1004.  Merriman, T.R. (2009). Type 1 diabetes, the A1 milk hypothesis and vitamin D deficiency. Diabetes Res. Clin. Pr. 83: 149-156.  Miyazaki, I., Gaedigk, R., Hui, M.F., Cheung, R.K., Morkowski, J., Rajotte, R.V., and Dosch, H.-M. (1994). Cloning of human and rat p69 cDNA, a candidate autoimmune target in type 1 diabetes. Biochim. Biophys. Acta 1227: 101-104.  54 Monetini, L., Barone, F., Stefanini, L., Petrone, A., Walk, T., Jung, G., Thorpe, R., Pozzilli, P., and Cavallo, M.G. (2003). Establishment of T cell line to bovine β-casein and β-casein-derived epitopes in patients with type 1 diabetes. J. Endocrinol. 176: 143-150.  Moreira, T.S., and Hamadeh, M.J. (2010). The role of vitamin D deficiency in the pathogenesis of type 2 diabetes mellitus. e-SPEN the Eur. e-j Clin. Nutr. Metab. 5: e155-e165.  Mozaffarian, D., Cao, H., King, I.B., Lemaitre, R.N., Song, X., Siscovick, D.S., and Hotamisligil, G.S. (2010). Trans-Palmitoleic acid, metabolic risk factors, and new-onset diabetes in U.S. adults. Ann. Intern. Med. 153:790-799.  Muntoni, S., Cocco, P., Aru, G., Cucca, F., and Muntoni, S. (2000). Nutritional factors and worldwide incidence of childhood type 1 diabetes. Am. J. Clin. Nutr. 71: 1525-1529.  Nettleton, J.A., Steffen, L.M., Ni, H., Liu, K., and Jacobs Jr., D.R. (2008). Dietary patterns and risk of incident type 2 diabetes in the multi-ethnic study of atherosclerosis (MESA). Diabetes Care 31: 1777-1781.  Nilsson, M., Holst, J.J., and Björck, I.M.E. (2007). Metabolic effects of amino acid mixtures and whey protein in healthy subjects: studies using glucose-equivalent drinks. Am. J. Clin. Nutr. 85: 996-1004.   55 Nilsson, M., Stenberg, M., Frid, A.H., Holst, J.J., and Björck, I.M.E. (2004). Glycemia and insulinemia in healthy subjects after lactose-equivalent meals of milk and other food proteins: the role of plasma amino acids and incretins. Am. J. Clin. Nutr. 80: 1246-1253.   Norris, J.M., Barriga, K., Klingensmith, G., Hoffman, M., Eisenbarth, G.S., Erlich, H.A., and Rewers, M. (2003). Timing of initial cereal exposure in infancy and risk of islet autoimmunity. J. Am. Med. Assoc. 290: 1713-1720.  Norris, J.M., Beaty, B., Klingensmith, G., Yu, L., Hoffman, M., Chase, P., Erlich, H.A., Hamman, R.F., Eisenbarth, G.S., and Rewers, M. (1996). Lack of association between early exposure to cow’s milk protein and -cell autoimmunity. J. Am. Med. Assoc. 276: 609-614.    Norris, J.M., and Scott, F.W. (1996). A meta-analysis of infant diet and insulin-dependent diabetes mellitus: Do biases play a role? Epidemiology 7: 87-92.  Östman, E.M., Liljeberg Elmståhl, H.G.M., and Björck, I.M.E. (2001). Inconsistency between glycemic and insulinemic responses to regular and fermented milk products. Am. J. Clin. Nutr. 74: 96-100.  Ouellet, V., Marois, J., Weisnagel, S.J., and Jacques, H. (2007). Dietary cod protein improves insulin sensitivity in insulin-resistant men and women. Diabetes Care 30: 2816-2821.   56 Patel, P.S., Sharp, J.S., Luben, R.N., Khaw, K.-T., Bingham, S.A., Wareham, N.J., and Forouhi, N.G. (2009). Association between type of dietary fish and seafood intake and the risk of incident type 2 diabetes. Diabetes Care 32: 1857-1863.   Pereira, M.A., Jacobs, D.R., Van Horn, L., Slattery, M.L., Kartashov, A.I., and Ludwig, D.S. (2002). Dairy consumption, obesity, and the insulin resistance syndrome in young adults The CARDIA Study. J. Am. Med. Assoc. 287: 2081-2089.  Persaud, D.R., and Barranco-Mendoza, A. (2004). Bovine serum albumin and insulin-dependent diabetes mellitus: is cow’s milk still a possible toxicological causative agent of diabetes? Food Chem. Toxicol. 42: 707-714.  Pešek, M., Špička, J., and Samková, E. (2005). Comparison of fatty acid composition in milk fat of Czech Pied cattle and Holstein cattle. Czech J. Anim. Sci. 50: 122-128.  Petersen, B.L., Ward, L.S., Bastian, E.D., Jenkins, A.L., Campbell, J., and Vuksan, V. (2009). A whey protein supplement decreases post-prandial glycemia. Nutr. J. 8: 47. DOI: 10.1186/1475-2891-8-47. Retrieved June 25, 2010, from http://www.nutritionj.com/content/8/1/47  Pfeuffer, M., and Schrezenmeir, J. (2002). Milk lipids in diet and health – medium chain fatty acids (MCFA). IDF Bulletin 377: 32-42.  Pfeuffer, M., and Schrezenmeir, J. (2006). Milk and the metabolic syndrome. Obes. Rev. 8: 109-118.  57 Pietropaolo, M., Castaño, L., Babu, S., Buelow, R., Kuo, Y.-L.S., Martin, S., Martin, A., Powers, A.C., Prochazka, M., Naggert, J., Leiter, E.H., and Eisenbarth, G.S. (1993). Islet cell autoantigen 69 KD (ICA69). Molecular cloning and characterization of a novel diabetes-associated autoantigen. J. Clin. Invest. 92: 359-371.  Pittas, A.G., Bawson-Hughes, B., Li, T., van Dam, R.M., Willett, W.C., Manson, J.E., and Hu, F.B. (2006). Vitamin D and calcium intake in relation to type 2 diabetes in women. Diabetes Care 29: 650-656.  Pittas, A.G., Lau, J., Hu, F.B., and Dawson-Hughes, B. (2007). Review: The role of vitamin D and calcium in type 2 diabetes. A systematic review and meta-analysis. J. Clin. Endocrinol. Metab. 92: 2017-2029.  Reeves, W.G., and Kelly, U. (1982). Insulin antibodies induced by bovine insulin therapy. Clin. Exp. Immunol. 50: 163-170.  Risérus, U., Willett, W.C., and Hu, F.B. (2009). Dietary fats and prevention of type 2 diabetes. Prog. Lipid Res. 48: 44-51.  Robinson, B.H., Dosch, H.-M., Martin, J.M., Åkerblom, H.K., Savilahti, E., Knip, M., and Ilonen, J. (1993). A model for the involvement of MHC class II proteins in the development of Type 1 (insulin-dependent) diabetes mellitus in response to bovine serum albumin peptides. Diabetologia 36: 364-368.   58 Rosenbauer, J., Herzing, P., and Giani, G. (2007). Early nutrition and risk of type 1 diabetes mellitus – A nationwide case-control study in preschool children. Exp. Clin. Endocrinol. Diabetes 115: 502-508.  Scott, F.W. (1990). Cow milk and insulin-dependent diabetes mellitus: is there a relationship? Am. J. Clin. Nutr. 51: 489-491.   Scott, F.W. (1996). Food-induced type 1 diabetes in the BB rat. Diabetes Metab Rev. 12: 341-359.  Sebokova, E., Christ, A.D., Boehringer, M., and Mizrahi, J. (2007). Dipeptidyl peptidase IV inhibitors: the next generation of new promising therapies for the management of type 2 diabetes. Curr. Top. Med. Chem. 7: 547-555.  Srivastava, M.D., Srivastava, A., Brouhard, B., Saneto, R., Groh-Wargo, S., and Kubit, J. (1996). Cytokines in human milk. Res. Commun. Mol. Pathol. Pharmacol. 93: 259-262.  Strotmeyer, E.S., Yang, Z., LaPorte, R.E., Chang, Y.-F., Steenkiste, A.R., Pietropaolo, M., Nucci, A.M., Shen, S., Wang, L., Wang, B., and Dorman, J.S. (2004). Infant diet and type 1 diabetes in China. Diabetes Res. Clin. Pract. 65: 283-292.  TEDDY Study Group. (2008). The Environmental Determinants of Diabetes in the Young (TEDDY) Study. Annals of the New York Academy of Sciences 1150: 1–13.   59 The DIAMOND Project Group. (2006). Incidence and trends of childhood type 1 diabetes worldwide 1990-1999. Diabetic Med. 23: 857-866.  Thorsdottir, I., and Ramel, A. (2003). Dietary intake of 10- to 16-year-old children and adolescents in central and northern Europe and association with the incidence of type 1 diabetes. Ann. Nutr. Metab. 47: 267-275.  Tremblay, A., and Gilbert, J.-A. (2009). Milk products, insulin resistance syndrome and type 2 diabetes. J. Am. Coll. Nutr. 28: 91S-102S.  Tremblay, F., Lavigne, C., Jacques, H., and Marette, A. (2007). Role of dietary proteins and amino acids in the pathogenesis of insulin resistance. Annu. Rev. Nutr. 27: 293-310.  TRIGR Trial to Reduce IDDM in the Genetically at Risk. (2011). About TRIGR. http://trigr.epi.usf.edu/about.html. Accessed February 18, 2011.  Truswell, A.S. (2006). Reply: the A2 milk case: a critical review. Eur. J. Clin. Nutr. 60: 924-925.  Vaarala, O. (2002). The gut immune system and type 1 diabetes. Ann. NY Acad. Sci. 958: 39-46.   Vaarala, O. (2008). Leaking gut in type 1 diabetes. Curr. Opin. Gastroenterol. 24: 701-706.    60 Vaarala, O., Paronen, J., Otonkoski, T., and Åkerblom, H.K. (1998). Cow milk feeding induces antibodies to insulin in children - a link between cow milk and insulin-dependent diabetes mellitus? Scand. J. Immunol. 47: 131-135.  van Dam, R.M., Hu, F.B., Rosenberg, L., Krishnan, S., and Palmer, J.R. (2006). Dietary calcium and magnesium, major food sources, and risk of type 2 diabetes in U.S. black women. Diabetes Care 29: 2238-2243.  van Elburg, R.M., Fetter, W.P.F., Bunkers, C.M., and Heymans, H.S.A. (2003). Intestinal permeability in relation to birth weight and gestational and postnatal age. Arch. Dis. Child Fetal Neonatal Ed. 88: F52-F55.   van Meijl, L.E.C., Vrolix, R., and Mensink, R.P. (2008). Dairy product consumption and the metabolic syndrome. Nutr. Res. Rev. 21: 148-157.  Villegas, R., Yang, G., Gao, Y.-T., Cai, H., Li, H., Zheng, W., and Shu, X.O. (2010). Dietary patterns are associated with lower incidence of type 2 diabetes in middle-aged women: the Shanghai Women’s Health Study. Int. J. Epidemiol. 39: 889-899.  Virtanen, S.M., and Knip, M. (2003). Nutritional risk predictor of  cell autoimmunity and type 1 diabetes at young age. Am. J. Clin. Nutr. 78: 1053-1067.      61 Virtanen, S.M., Räsänen, L., Ylönen, K., Aro, A., Clayton, D., Langholz, G., Pitkäniemi, J., Savilahti, E., Lounamaa, R., and Tuomilehto, J.  (1993). Early introduction of dairy products associated with increased risk of IDDM in Finnish children. The Childhood in Diabetes in Finland Study Group. Diabetes 42: 1786-1790.  Virtanen, S.M., Uusitalo, L, Kenward, M.G., Nevalainen, J., Uusitalo, U., Kronberg-Kippilä, C., Ovaskainen, M.-L., Arkkola, T., Niinistö, S., Hakulinen, T., Ahonen, S., Simell, O., Ilonen, J., Veijola, R., and Knip, M. (2011). Maternal food consumption during pregnancy and risk of advanced -cell autoimmunity in the offspring. Pediatric Diabetes 12: 95-99. DOI:10.1111lj.1399-5448.2010.00668.x  Visser, J., Rozing, J., Sapone, A., Lammers, K., and Fasano, A. (2009). Tight junctions, intestinal permeability, and autoimmunity celiac disease and type 1 diabetes paradigms. Molecular Structure and Function of the Tight Junction: Ann. N.Y. Acad. Sci. 1165: 195-205. DOI: 10.1111/j.1749-6632.2009.04037.x  Wasmuth, H.E., and Kolb, H. (2000). Cow’s milk immune-mediated diabetes. Proc. Nutr. Soc. 59: 573-579.  Wein, S., Wolffram, S., Schrezenmeir, J., Gasperiková, D., Klimes, I., and Seböková, E. (2009). Medium-chain fatty acids ameliorate insulin resistance caused by high-fat diets in rats. Diabetes Metab. Res. Rev. 25: 185-194.   62 Wild, S., Roglic, G., Greeen, A., Sicree, R., and King, H. (2004). Global prevalence of diabetes: estimates for the year 2000 and projections for 2030. Diabetes Care 27: 1047-1053.   Ziegler, A.-G., Hummel, M., Schenker, M., and Bonifacio, E. (1999). Autoantibody appearance and risk for development of childhood diabetes in offspring of parents with type 1 diabetes. The 2-year analysis of German BABYDIAB study. Diabetes 48: 460-468.  Ziegler, A.-G., Schmid, S., Huber, D., Hummel, M., and Bonifacio, E. (2003). Early infant feeding and risk of developing type 1 diabetes-associated autoantibodies. J. Am. Med. Assoc. 290: 1721-1728.  TABLES  Table 1 Ecological and Epidemiological Studies on the Association between Dairy Product Consumption and the Incidence of Type 1 Diabetes  Study Countries investigated Dietary variables investigated Association between  dairy consumption and T1D1  Scott, 1990  Japan, France, German Democratic Republic, Austria, England, Canada, New-Zealand, Netherlands, Denmark, USA, Norway, Sweden, Finland  Unfermented milk products  Unfermented milk products (+): r = 0.86, p < 0.01 Dahl-Jorgensen et al., 1991 Finland, Sweden, Norway, Great Britain, Denmark, United States, New Zealand, Netherlands, Canada, France, Israel, Japan  Fluid milk  Fluid milk (+) : r = 0.96 Fava et al., 1994 Italy Fluid milk & cheese Fluid milk (+) : r = 0.829, P <0.004 Cheese: No association Muntoni et al., 2000 Algeria, Sudan, Tanzania, Canada, United States, Cuba, Mexico, Puerto Rico, Brazil, Chile, Peru, Israel, Japan, Kuwait, Republic of Korea, Australia, New Zealand, Austria, Belgium, Bulgaria, Denmark, Finland, France, Germany, Greece, Hungary, Iceland, Italy, Latvia, Luxembourg, Malta, Netherlands, Norway, Poland, Portugal, Romania, Russia, Spain, Sweden, United Kingdom  Dietary energy from major food groups (meat, cereals, dairy products, fruit & vegetable) Dairy products (+) : r = 0.80, P <0.0001 Thorsdottir & Ramel, 20032 Austria, Belgium, Denmark, Finland, Germany, Iceland, Ireland, Northern Ireland, Norway, Sweden, Switzerland Dietary intake Milk intake (+) : r = 0.829, p 0.042 Animal products3 (+): r = 0.999, p 0.001  1 Plus (+) signs refer to a positive association 2 Result of analysis excluding Icelandic data 3 Animal products include milk, meat and eggs  Table 2 Case-Control Studies on Infant-Feeding and the Incidence of Type 1 Diabetes  Study Location Age of cases & controls Number of cases & controls Selection of controls Elements of the diet assessment questionnaire Results1  Virtanen et al., 1993  Finland  < 15 yrs of age  690 case-control pairs  Birth-date-matched & sex-matched from the Finnish national population registry  Duration (overall & exclusive) of breast-feeding, age at introduction of supplementary milk feeding and solid foods, type of supplementary milk    Dairy exposure < 2 mo of age (+) Esfarjani et al., 2001 Tehran 1.5 to 14 yrs of age 52 case-control pairs Sex, age, social status & geographical location-matched from children at the outpatient pediatric clinic   Duration (overall & exclusive) of breast-feeding, age at introduction of dietary products including formula, cow’s milk and solid foods  No relationship between early exposure to cow's milk and T1D Strotmeyer et al., 2004 China < 15 yrs of age 247 cases 443 controls Sex, age ( 5 yrs), ethnic group, area of residence-matched from hospital administration for injuries/minor surgery and locally resident populations  Intake of milk (breast milk & breast milk substitutes) and food (rice, eggs, meat, fruit, vegetables, steamed bread, fish, noodles, and others)  No relationship between cow's milk consumption < 6 mo of age and T1D  Cow’s milk consumption between 7-12 mo of age (-): OR = 0.60, 95% CI 0.43-0.85; P 0.004 Rosenbauer et al., 2007 Germany < 5 yrs of age 749 cases 624 controls Children from the acquaintance of case families Duration (overall) of breast-feeding, age at introduction of breast milk substitutes and solid food, type of breast milk substitutes, current level of customary cow’s milk intake  Exposure to formula/cow's milk < 5 mo of age (+): OR = 1.34, 95% CI 1.03-1.74; P 0.030  1 Plus (+) signs refer to a positive association with T1D; minus (-) sign refers to a negative association withT1D Table 3 Cohort Studies on Infant-Feeding and the Development of Diabetes-Associated Autoantibodies  Study Location Design of the cohort (follow-up, mo or yrs) Number of children Diabetes-associated autoantibodies investigated1 Association between age of first exposure to cow's milk and positivity for diabetes-associated autoantibodies2 External Factors/Adjustments  Norris et al., 1996  United States  Diabetes Autoimmunity Study in the Young DAISY3    18 cases & 153 controls  IAA, GADA, IA-2A  <3 mo of age = no association <6 mo of age = no association  Maternal age at birth and maternal diabetes status Couper et al., 1999 Australia The Australian Baby DIAB study (median 29 mo) 317 IAA, GADA, IA-2A No association NA4 Kimpimäki et al., 2001 Finland The Finnish Type I Diabetes Prediction and Prevalence (DIPP) Study (4 yrs) 65 cases & 390 controls ICA, IAA, GADA, IA-2A <2 mo of age (+): OR = 9.46, 95% CI 1.50-59.8; p 0.017 for all four autoantibodies 2-3.9 mo of age (+): OR =17.9, 95% CI 1.87-172; p 0.012 for all four autoantibodies  Mother's age & duration of education, relative height & weight at 12 mo of age Holmberg et al., 2007 Sweden All Babies in Southeast Sweden (ABIS) Study (5-6 yrs) 3776 IAA, GADA, IA-2A 1-3 mo of age (+): OR = 1.84, 95% CI 1.01-3.37; P 0.047 for IAA, GADA and/or IA-2A Mother's age & duration education, infections during pregnancy, delivery mode, low birth weight (<2500g), early age of gestation (≤37 weeks), first-degree relative with T1D, coeliac disease or T2D, gastroenteritis    1 IAA = insulin autoantibodies, GADA = glutamic acid decarboxylase antibodies, IA-2A = tyrosine phosphatase-like protein IA-2 antibodies, ICA = islet cell antibodies 2 Plus (+) signs refer to a positive association with development of  cell autoimmunity 3 No follow-up period in the retrospective study by Norris et al. (1996) 4 No adjustments were mentioned by Couper et al. (1999)    Table 4  Differences in amino acid sequences of β-casein variants. (Adapted from Kamińsky et al., 2007)                  β-casein variant amino acid at sequence position 18 25 35 36 37 67 72 88 93 106 117 122 137 138 A1 Ser-P Arg Ser-P Glu Glu His Glu Leu Gln His Gln Ser Leu Pro A2      Pro         A3      Pro    Gln     B            Arg   C   Ser  Lys          D Lys     Pro         E    Lys  Pro         F              Leu G            Leu   H1  Cys    Pro  Ile       H2      Pro Glu  Leu     Glu I      Pro   Leu      Table 5 Observational Studies on the Association between Dairy Intake and Type 2 Diabetes  RR = relative risk, HR = hazard ratio, BMI = body mass index, Ca = calcium, vit. D = vitamin D, Mg = magnesium 1 Dairy intake shown as servings/day, except for the Elwood et al. (2007) and the Kirii et al. (2009) studies in which intakes were reported as quartiles and g/day, respectively  2 Results with multivariate model 3 adjustments 3 Data from a food frequency questionnaire was available for 2375 men, but only 640 men kept a 7-day weighed dietary intake record Study Number and gender of subjects Location Design of the Cohort  (follow-up, yrs) Dairy intake (servings/day)1 (lowest and highest amount) Results Adjustments Choi et al., 2005 41,254 men United States Health Professionals Follow-up cohort HPFS (12 yrs) Total dairy: <0.9 and ≥2.9 Total dairy: RR = 0.77, 95% CI 0.62-0.95; Ptrend 0.003 Age, BMI, total energy, smoking, high cholesterol, hypertension, diabetes history, alcohol, physical activity, follow-up time, certain dietary variables, fiber, glycemic load  Low-fat: <0.14 and >1.58  Low-fat: RR = 0.73, 95% CI 0.59-0.89; Ptrend 0.001 High-fat: <0.38 and >1.72  High-fat: RR = 0.97, 95% CI 0.78-1.21; Ptrend 0.78 Liu et al., 20062 37,183 women United States Women's Health Study cohort WHS (10 yrs) Total dairy: <0.85 and >2.9 Total dairy: RR = 0.68, 95% CI 0.52-0.89; Ptrend 0.006 Age, BMI, total energy, smoking, high cholesterol. hypertension, diabetes history, alcohol, physical activity, hormone therapy, total fat, fiber, glycemic load, vit. D, Ca, Mg Low-fat: ≤0.27 and >2.00  Low-fat: RR = 0.69, 95% CI 0.52-0.91; Ptrend 0.007 High-fat: <0.20 and >1.329  High-fat: RR = 0.99, 95% CI 0.82-1.20; Ptrend 0.90 Pittas et al., 2006 83,779 women United States Nurses' Health Study cohort NHS (20 yrs) Total dairy: <1.0 and >3.0 Total dairy: RR = 0.89, 95% CI 0.81-0.99; Ptrend 0.008 Age, BMI, total energy, smoking, hypertension, diabetes history, alcohol, physical activity, type of fat, glycemic load, retinol, caffeine, fiber, Ca, vit. D  van Dam et al., 2006 41,186 black women United States Black Women's Health Study cohort BWHS (8 yrs) Total dairy: 0.07 and 2.53 Total dairy: HR = 0.93, 95% CI 0.75-1.15; Ptrend 0.31 Age, BMI, total energy, smoking, physical activity, alcohol, diabetes history, education level, coffee, sugar-sweetened soft drink, meats Low-fat: 0.00 and 1.22 Low-fat: HR =0.87, 95% CI 0.76-1.00; Ptrend 0.04 High-fat: 0.07 and 1.33 High-fat: HR = 1.03, 95% CI 0.88-1.20; Ptrend 0.94 Elwood et al., 2007  640 men3 United Kingdom The Caerphilly cohort (20 yrs) Total milk: Highest 1/4 Total milk: RR = 0.57, 95% CI 0.20-1.63; Ptrend 0.247 Age, BMI, smoking, social class Kirii et al., 2009 25,877 men & 33,919 women  Japan The Japan Public Health Center-based Prospective Study JPHC(5 yrs) Total dairy: <50 and ≥300 Total dairy: OR = 0.71, 95% CI 0.51-0.98; Ptrend 0.054  Age, BMI, total energy, smoking, hypertension, diabetes history, alcohol, physical activity, coffee, energy-adjusted magnesium, area Milk: <50 and ≥200 Milk: OR = 0.87, 95% CI 0.70-1.09; Ptrend 0.16  Cheese: 0 and ≥5 Cheese: OR = 1.12, 95% CI 0.80-1.57; Ptrend 0.56 Yogurt: 0 and ≥60 Yogurt: OR = 0.77, 95% CI 0.58-1.01; Ptrend 0.13 Table 6 Observational Studies on the Association Between Dietary Patterns that Include Dairy Products and Type 2 Diabetes RR = relative risk, OR = odds ratio, HR = hazard ratio, BMI = body mass index, WHR = waist-to-hip ratio 1 Dietary patterns are characterized by greater loading of the listed food groups  2 “Fats & Processed Meats” (greater loading for the food groups “fats & oils”, “high-fat & processed meats”, “fried potatoes”, “salty snacks”, “desserts”);   “Vegetables & Fish” (greater loading for various vegetable groups [i.e. “dark-yellow”, “cruciferous”, “other vegetables”], “fish”, “soups”);   “Beans, Tomatoes & Refined Grains” (greater loadings for food groups “legumes”, “tomatoes”, “refined bread, rice, pasta”, “high-fat cheeses & cheese & cream sauces”, “avocados & guacamole”);   “Whole Grains & Fruit” (greater loading for food groups “whole-grain bread, rice, pasta”, “fruit”, “seeds, nuts, peanut butter”, “green leafy vegetables”, “low-fat milk”)    3 Results for tertile 3 versus tertile 1  Study Number, nationality and gender of subjects Location Design of the Cohort  (follow-up, yrs)             Dietary patterns1 Results Adjustments Brunner et al., 2008 7,731 men & women London Whitehall II study (15 yrs) "Unhealthy" (white bread, processed meat, fries, full-cream milk), "Sweet" (white bread, biscuits, cakes, processed meat, high-fat dairy products), "Mediterranean-like" (fruit, vegetables, rice, pasta, wine), "Healthy" (fruit, vegetables, whole-meal bread, low-fat dairy, little alcohol) "Healthy" vs "Unhealthy" pattern: HR = 0.74, 95% CI 0.58-0.94; Ptrend 0.016 Age, sex, total energy, smoking, physical activity, social class, ethnicity Nettleton et al., 20082 2,634 men & 2,377 women (2,177 whites, 1,205 blacks, 1,016 Hispanics & 613 Chinese) Six study centers in the United States  Multi-Ethnic Study of Atherosclerosis  MESA (~5 yrs) 4 PCA-derived dietary patterns ("Fats & Processed Meats", "Vegetables & Fish", "Beans, Tomatoes, and Refined Grains", "Whole Grains & Fruit"), Priori-defined low-risk food pattern score based on the consumption of 10 food groups (i.e. Whole Grains, Nuts/Seeds, Vegetables, Low-Fat Dairy, Coffee, High-Fat Dairy, Red Meat, Processed Meat, White Potatoes, Regular Soda) "Whole grains & Fruit" pattern:  HR = 0.85, 95% CI 0.76-0.95; Ptrend 0.005  A priori low-risk food pattern: HR = 0.87, 95% CI 0.81-0.99; Ptrend 0.04  Age, sex, total energy, smoking, physical activity, education level, race/ethnicity, study center, weekly supplement Liese et al., 20093 862 men & women (346 whites, 516 blacks/Hispanics) Four Study centers in the United States Insulin Resistance Atherosclerosis Study  IRAS (5 yrs) 8 food groups from the “DASH” diet (Grains, Vegetables, Fruits, Dairy, Meat, Nuts/Seeds/Legumes, Fats/Oils, Sweets) Whites: OR = 0.31, 95% CI 0.13-0.75; Ptrend 0.03 Blacks/Hispanics: OR = 1.34, 95% CI 0.70-2.58; Ptrend0.68 Age, BMI, sex, total energy, smoking, diabetes history, education level, race/ethnicity/clinic, glucose tolerance status  Erber et al., 2010 36,256 men & 39,256 women (29,759 Caucasians, 35,244 Japanese Americans &  10,509 Native Hawaiians) Hawaii & California The Multiethnic Cohort (14 yrs) "Fat & Meat" (discretionary fat, meat, eggs, cheese), "Vegetables" (high amounts of vegetables & low loading of fruits), "Fruit and Milk" (high loadings on milk, yogurt, cheese, fruits) "Fruit and milk" pattern:  Women:  HR = 0.85, 95% CI 0.76-0.96; Ptrend 0.005 Men:  HR = 0.92, 95% CI  0.83-1.02; Ptrend 0.04   BMI, total energy, smoking, high blood pressure, alcohol, physical activity, education level, marital status, and stratified by age at cohort entry Villegas et al., 2010 64,191 Chinese Women Shanghai Shanghai Women's Health Study  SWHS (6.9 yrs) "Cluster 1" (highest intake of staples, i.e. rice, noodles, steamed bread, bread), "Cluster 2" (highest intake of dairy milk), "Cluster 3" (highest intake of poultry, pork, red meat, fish & shellfish, eggs, fruits & vegetables and the highest energy intake) "Cluster 2" versus "Cluster 1": RR = 0.78, 95% CI 0.71-0.86  Age, BMI, total energy, smoking, hypertension, alcohol, physical activity, education level, income level, occupation, WHR  36 FIGURE CAPTIONS   Figure 1  Primary amino acid sequence of BSA. (Adapted from Farrell et al., 2004. The entry name and file number of the protein in the UniProt Knowledgebase of ExPASy Proteomics Server are ALBU-BOVIN and P02769). The ABBOS fragment sequences as reported by Robinson et al. (1993) (region 128-145, dotted line) and by Karjalainen et al. (1992) (region 152-169, solid line, incorporating the missing Tyr in the sequence updated by Hirayama et al., 1990) are also represented.  Figure 2  Regions of similarity between human p69 and BSA sequences. (Adapted from Persaud & Barranco-Mendoza, 2004; the numbering of the amino acid fragments of BSA has been adapted to be in accordance with the BSA primary amino acid sequence shown in Figure 1).  Figure 3  Factors playing roles in gut maturation and the possible implication of the gut immune system in the development of T1D. (Adapted from Wasmuth & Kolb, 2000).  37   1Asp-Thr-His-Lys-Ser-Glu-Ile-Ala-His-Arg10-Phe-Lys-Asp-Leu-Gly-Glu-Glu-His-Phe-Lys20-Gly-Leu-Val-Leu-Ile-Ala-Phe-Ser-Gln-Tyr30-Leu-Gln-Gln-Cys-Pro-Phe-Asp-Glu-His-Val40-Lys-Leu-Val-Asn-Glu-Leu-Thr-Glu-Phe-Ala50-Lys-Thr-Cys-Val-Ala-Asp-Glu-Ser-His-Ala60-Gly-Cys-Glu-Lys-Ser-Leu-His-Thr-Leu-Phe70-Gly-Asp-Glu-Leu-Cys-Lys-Val-Ala-Ser-Leu80-Arg-Glu-Thr-Tyr-Gly-Asp-Met-Ala-Asp-Cys90-Cys-Glu-Lys-Gln-Glu-Pro-Glu-Arg-Asn-Glu100-Cys-Phe-Leu-Ser-His-Lys-Asp-Asp-Ser-Pro110-Asp-Leu-Pro-Lys-Leu-Lys-Pro-Asp-Pro-Ans120-Thr-Leu-Cys-Asp-Glu-Phe-Lys-Ala-Asp-Glu130-Lys-Lys-Phe-Trp-Gly-Lys-Tyr-Leu-Tyr-Glu140-Ile-Ala-Arg-Arg-His-Pro-Tyr-Phe-Tyr-Ala150-Pro- Glu-Leu-Leu-Tyr-Tyr-Ala-Asn-Lys-Tyr160-Asn-Gly-Val-Phe-Gln-Glu-Cys-Cys-Gln-Ala170-Glu-Asp-Lys-Gly-Ala-Cys-Leu-Leu-Pro-Lys180-Ile-Glu-Thr-Met-Arg-Glu-Lys-Val-Leu-Ala190-Ser-Ser-Ala-Arg-Gln-Arg-Leu-Arg-Cys-Ala200-Ser-Ile-Gln-Lys-Phe-Gly-Glu-Arg-Ala-Leu210-Lys-Ala-Trp-Ser-Val-Ala-Arg-Leu-Ser-Gln220-Lys-Phe-Pro-Lys-Ala-Glu-Phe-Val-Glu-Val230-Thr-Lys-Leu-Val-Thr-Asp-Leu-Thr-Lys-Val240-His-Lys-Glu-Cys-Cys-His-Gly-Asp-Leu-Leu250-Glu-Cys-Ala-Asp-Asp-Arg-Ala-Asp-Leu-Ala260-Lys-Try-Ile-Cys-Asp-Asn-Gly-Asp-Thr-Ile270-Ser-Ser-Lys-Leu-Lys-Glu-Cys-Cys-Asp-Lys280-Pro-Leu-Leu-Glu-Lys-Ser-His-Cys-Ile-Ala290-Glu-Val-Glu-Lys-Asp-Ala-Ile-Pro-Glu-Asn300-Leu-Pro-Pro-Leu-Thr-Ala-Asp-Phe-Ala-Glu310-Asp-Lys-Asp-Val-Cys-Lys-Asn-Tyr-Gln-Glu320-Ala-Lys-Asp-Ala-Phe-Leu-Gly-Ser-Phe-Leu330-Tyr-Glu-Tyr-Ser-Arg-Arg-His-Pro-Glu-Tyr340-Ala-Val-Ser-Val-Leu-Leu-Arg-Leu-Ala-Lys350-Glu-Tyr-Glu-Ala-Thr-Leu-Glu-Glu-Cys-Cys360-Ala-Lys-Asp-Asp-Pro-His-Ala-Cys-Tyr-Ser370-Thr-Val-Phe-Asp-Lys-Leu-Lys-His-Leu-Val380-Asp-Glu-Pro-Gln-Asn-Leu-Ile-Lys-Gln-Asn390-Cys-Asp-Gln-Phe-Glu-Lys-Leu-Gly-Glu-Tyr400-Gly-Phe-Gln-Asn-Ala-Leu-Ile-Val-Arg-Tyr410-Thr-Arg-Lys-Val-Pro-Gln-Val-Ser-Thr-Pro420-Thr-Leu-Val-Glu-Val-Ser-Arg-Ser-Leu-Gly430-Lys-Val-Gly-Thr-Arg-Cys-Cys-Thr-Lys-Pro440-Glu-Ser-Glu-Arg-Met-Pro-Cys-Thr-Glu-Asp450-Tyr-Leu-Ser-Leu-Ile-Leu-Asn-Arg-Leu-Cys460-Val-Leu-His-Glu-Lys-Thr-Pro-Val-Ser-Glu470-Lys-Val-Thr-Lys-Cys-Cys-Thr-Glu-Ser-Leu480-Val-Asn-Arg-Arg-Pro-Cys-Phe-Ser-Ala-Leu490-Thr-Pro-Asp-Glu-Thr-Tyr-Val-Pro-Lys-Ala500-phe-Asp-Glu-Lys-Leu-Phe-Thr-Phe-His-Ala510-Asp-Ile-Cys-Thr-Leu-Pro-Asp-Thr-Glu-Lys520-Gln-Ile-Lys-Lys-Gln-Thr-Ala-Leu-Val-Glu530-Leu-Leu-Lys-His-Lys-Pro-Lys-Ala-Thr-Glu540-Glu-Gln-Leu-Lys-Thr-Val-Met-Glu-Asn-Phe550-Val-Ala-Phe-Val-Asp-Lys-Cys-Cys-Ala-Ala560-Asp-Asp-Lys-Glu-Ala-Cys-Phe-Ala-Val-Glu570-Gly-Pro-Lys-Leu-Val-Val-Ser-Thr-Gln-Thr580-Ala-Leu-Ala583    Figure 1   38     p69  =  39Lys-Ala-Thr-Gly-Lys-Lys44                        BSA  = 127Lys-Ala-Asp-Glu-Lys-Lys132     p69  = 205Phe-Asp-Lys-Leu-Lys-Met-Asp-Val212    BSA  = 373Phe-Asp-Lys-Leu-Lys-His-Leu-Val380     p69  = 353Gly-Ala-Cys-Leu-Gly-Pro358    BSA  = 174Gly-Ala-Cys-Leu-Leu-Pro179    Figure 2 1st region :  2nd region :  3rd region :   39          Figure 3                       Gut mucosa and gut-associated lymphoid tissue Growth hormones and cytokines provided by breast milk Intestinal flora Dietary antigens (e.g. bovine insulin, milk proteins, soy, gluten) Viral infection Genetic predisposition Deficient oral tolerance Increased gut permeability ß-cell  autoimmunity 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.52383.1-0343572/manifest

Comment

Related Items