The Big Diabetes Lie

Diabetes Holistic Treatment

Get Instant Access






Note: Caucasian population from Jefferson County, AL and Allegheny County, PA. African-American population from Jefferson County, AL and Allegheny County, PA. Asian population from Hokkaido, Japan and Seoul, Korea.

a Absolute risks are expressed as % through age 30 yr. Source: ref. 23.

Note: Caucasian population from Jefferson County, AL and Allegheny County, PA. African-American population from Jefferson County, AL and Allegheny County, PA. Asian population from Hokkaido, Japan and Seoul, Korea.

a Absolute risks are expressed as % through age 30 yr. Source: ref. 23.

among Chinese (28). The DRB1*15-DQA1*0102-DQB1*0602 is protective and associated with a reduced risk for type 1 diabetes in most populations.

Because of the geographic differences in the incidence of type 1 diabetes, as well as population variation in the frequency of DR-DQ haplotypes, the WHO DiaMond Project began a standardized international case-control molecular epidemiology study of type 1 diabetes in 1990 (29). More than 20 participating centers were involved in this collaboration. Cases were selected from DiaMond incidence registries, and nondiabetic controls were identified from corresponding populations at risk. This approach permitted the estimation of relative risks, absolute risks, and population attributable fractions for individuals with two, one, and zero high-risk HLA-DQ haplotypes for each area. As illustrated in Table 2, relative-risk estimates revealed statistically significant dose-response relationships in all populations studied (25). These data indicate that worldwide, susceptibility for type 1 diabetes is determined, in part, by high-risk HLA-DQ haplotypes (30).

Although these data present useful information in terms of the strength of genetic associations, cumulative incidence rates for specific haplotype combinations can be more meaningful from a clinical and public health perspective (25). These rates reflect the absolute risk of developing type 1 diabetes given two, one, or zero high-risk haplo-types. As illustrated in Table 2, Caucasians from moderate-high-incidence countries, such as Finland, the United States, and New Zealand, with two "susceptible" haplo-types had cumulative type 1 diabetes incidence rates that approached those typically observed for first-degree relatives (i.e., approx 5% through age 30 yr). However, corresponding estimates for the low-incidence populations, such as Korea, Japan, and China, were much less striking (0.1%). Thus, HLA-DQ haplotypes appear to be better predictive markers for type 1 diabetes in Caucasian than Asian populations.

Using these data, population attributable fractions (PAFs) can also be estimated. PAFs provide important information regarding the potential public health implications for disease-prevention strategies. They reflect the proportion of disease incidence that can be attributed to specific risk factors. In this context, they estimate the proportion of cases that could be prevented by modifying the environment for "susceptible" individuals. As shown in Table 2, PAFs were lower for Asian compared to Caucasian or African-American groups (25). This indicates that disease interventions in "susceptible" individuals would likely have greater impact among Caucasians or African-Americans than Asians. However, even among the latter groups, no more than half of the disease would be prevented, assuming a successful intervention in all "susceptible" individuals.

Insulin Gene Polymorphisms

In Caucasians, it has been demonstrated that the insulin gene region (INS), located on chromosome 11p15.5, contains the second major susceptibility locus for type 1 diabetes (i.e., IDDM2) (31,32). Positive associations have been observed with a nontran-scribed minisatellite (variable number of tandem repeats [VNTR]) region in the 5' flanking region. There are two common alleles. The shorter class I allele predisposes to type 1 diabetes, whereas the longer class III allele appears to be protective. The biological plausibility of these associations may relate to the expression of insulin mRNA in the thymus (32). Class III alleles appear to generate higher levels of insulin mRNA than class I alleles. Such differences could contribute to a better immune tolerance for class III positive individuals by increasing the likelihood of negative selection for autoreactive T-cell clones.

The effect of the insulin gene also appears to vary by ethnicity. Undlien et al. found that class I alleles were significantly associated with type 1 diabetes in Caucasians. However, they only reached borderline significance in Tanzanian blacks and were not associated with the disease in Japanese (33). However, other studies found a significant positive association between INS class I alleles and type 1 diabetes in the Japanese (34). Methodological differences, such as heterogeneous case and control groups and variations in allele frequencies in the general populations, may be responsible for the inconsistencies in the literature. There also may be an interaction between the insulin gene and the HLA-DR, DQ loci, which varies by ethnicity (35,36). This may also contribute to the geographic patterns of type 1 diabetes.

Beta Cell Autoantibodies

Evidence that type 1 diabetes is an autoimmune disorder is based on the presence of lymphocytic infiltrates of the pancreas at the onset of the diseases (37), as well as the occurrence of autoantibodies to islet cell antigens (ICAs), tyrosine phosphatase IA-2 (IA-2), glutamic acid decarboxylase (GAD), and insulin autoantibodies (IAA) (38,39). The presence of these autoantibodies indicates that tissue damage has likely been initiated by other etiologic agents. Thus, they represent important preclinical markers rather than risk factors for the disease.

Numerous studies have reported high prevalence rates for ICAs (65-100%) and GAD autoantibodies (60-68%) among Caucasian children at disease onset (38,39). Data from Allegheny County, PA have also revealed that Caucasian children with type 1 diabetes are significantly more likely to have P-cell autoantibodies at disease onset than affected African-American children (45% vs 30%, p < 0.05) (40). This may reflect etiologic differences because some African-American children are able to discontinue daily insulin injections after stabilization of their metabolic control.

Although most type 1 diabetes cases have P-cell autoantibodies at disease onset, they decrease in prevalence over time. In addition, they are rarely observed among first-degree relatives (2-5%) or in the general population (approx 1%) (41-43). Moreover, not all autoantibody positive individuals develop the disease. However, first-degree relatives who are positive for multiple autoantibodies appear to be at very high risk for developing type 1 diabetes. Some clinical studies have estimated the positive predictive values associated with two and three autoantibodies at 65% and >90%, respectively (43,44). It has been suggested that P-cell autoantibodies are as predictive in the general population as they are in high-risk families (45). However, most studies have found them to be better markers among first-degree relatives (41,42). This has important implications regarding potential disease interventions, which will be discussed in greater detail later in this chapter.

Data from our research group have also shown that HLA-DQ haplotypes modify type 1 diabetes risk in the absence of P-cell autoantibodies (46). Among our first-degree relatives who were negative for autoantibodies to GAD, IA-2, and ICA, 32% of those who carried two high-risk HLA-DQ haplotypes developed the disease after 12.5 yr of follow-up. These results emphasize the importance of both genetic and autoantibody markers in estimating type 1 diabetes risk.


Many epidemiologic investigations have supported the involvement of viruses in the etiology of type 1 diabetes (47,48). They are thought to act as initiators, accelerators, or precipitators of the disease. They may function by direct or indirect mechanisms. However, it is not clear whether they are necessary (or sufficient) to cause type 1 diabetes.

The viruses that have received the most attention include the enteroviruses, especially Coxsackie virus B (CVB). Sequence homology between a highly conserved nonstructural CVB4 protein (P2C) and GAD has been reported (49). In addition, several investigations indicated that antibodies to P2C and GAD crossreacted (50,51). However, other studies failed to confirm these results (52) or observed P2C antibodies among healthy GAD-negative controls (53). T-Cell proliferation studies have also been conflicting (54-56). Moreover, two reports (55,57) noted that reactivity was strongest in subjects with DR4, which contrasts with other investigations showing that the CVB associations with type 1 diabetes occurred among individuals with DR3 (58,59). Because CVB infections are frequent during childhood and are known to have systemic effects on the pancreas, they are likely to remain considered as important risk factors for type 1 diabetes.

Other viruses have also been associated with type 1 diabetes. Finnish investigators observed an increase in the incidence of diabetes 2-4 yr after a mumps epidemic (60). Studies of cytomegaloviruses (CMVs) have shown an increased prevalence of integrated viral genome in DNA extracted from lymphocytes when comparing cases with type 1 diabetes to controls without the disease (61). CMV antibody studies, however, have been inconsistent (62,63), suggesting that persistent, rather than acute, CMV infections during childhood may increase type 1 diabetes risk. In addition, rotaviruses, which are common causes of childhood gastroenteritis, contain peptide sequences similar to T-cell epitopes in IA-2 and GAD (64). A recent prospective study showed that high-risk children who developed P-cell autoantibodies or type 1 diabetes were significantly more likely to show serological evidence of rotavirus infections than those who did not develop diabetes autoimmunity (65). These data suggest that infections with rotavirus may also trigger P-cell autoimmunity in genetically susceptible children.

Epidemiologic studies have revealed that the time of exposure to viruses may be important. Approximately 10-20% of children with congenital rubella syndrome (CRS), particularly those who carry high-risk HLA alleles, develop autoimmune type 1 diabetes (66). Recently, T-cell-stimulation studies revealed that rubella virus peptides with binding motifs for HLA-DR3/DR4 were recognized by clones specific to GAD in type 1 diabetics, particularly those with CRS (67). These data provide additional evidence for the importance of early exposures in the etiology of the disease.

Other investigations have shown that enteroviral infections in utero increase the risk of developing the disease (68,69). Most recently, a prospective study of infants at high genetic risk revealed that enterovirus infections were more common among those who developed P-cell autoimmunity than those who remained antibody negative (70). In more than half of the infants, the infections occurred 6 mo prior to the appearance of autoantibodies. These data indicate that enteroviral infections during pregnancy or in early infancy may be related to the development of diabetes autoimmunity in susceptible infants.

Human endogenous retroviruses (HERVs) received considerable attention after a study provided evidence for the involvement of a superantigen in the etiology of type 1 diabetes (71). Superantigens can be products of bacteria, viruses, or endogenous retroviral genes and function by stimulating T-cell families rather than specific clones. One investigator reported that human endogenous retrovirus K (IDDMK1222) RNA was present in 10 other type 1 diabetic cases, but not in controls (72). Other investigators failed to replicate these findings and observed IDDMK1222 sequences and expression among both type 1 diabetic and non diabetic individuals (73,74). However, one group reported HLA-DQA1*0301-DQB1*0302 haplotypes carrying HERV-K long-terminal repeat (LTR3) sequences were preferentially transmitted to type 1 diabetics than those without the LTR3 (75). In contrast, the absence of LTR3 in HLA-DQA1*0201-DQB1*0501 haplotypes conferred increased risk of the disease. The authors hypothesized that these sequences may affect transcription or tissue-specific regulation of DQB1 and, therefore, influence disease risk.

Infant Nutrition

Another hypothesis that has received considerable attention in recent years relates to early exposure to cow's milk protein and the subsequent development of type 1 diabetes (76-79). Experimental studies in rodents revealed that the frequency of type 1 diabetes could be modified by altering the cow's milk protein composition of dietary chows (80). This led investigators to speculate about the role of diet in the etiology of type 1 diabetes in humans.

The first epidemiologic observation of such a relationship was by Borch-Johnsen et al., who found that diabetic children were breast-fed for shorter periods of time than healthy siblings or children from the general population (81). The authors postulated that lack of immunologic protection from insufficient breast-feeding may lead to type 1 diabetes later during childhood. It was also hypothesized that the protective effect of breast feeding may indirectly reflect early exposure to dietary proteins.

Numerous case-control studies subsequently investigated this issue. Results of a meta-analysis revealed a weak positive association between exposure to cow's milk at an early age (< 3 mo) and type 1 diabetes (odds ratio = 1.4; 95% confidence interval =

1.2-1.6) (82). These data were supported by investigations that showed associations with elevated levels of antibodies to cow's milk peptides among cases compared to controls (83,84). However, T-cell proliferation studies in response to cow's milk antigens have been negative or lacked specificity for type 1 diabetes (85,86). Concern has also been raised by short-term natural history studies that showed no association between infant feeding patterns and the development of P-cell autoimmunity (87,88). Moreover, consumption of milk or other dietary proteins later in childhood appears to increase risk of developing the disease (89). Thus, the contribution of the infant diet to the development of type 1 diabetes is far from clear.

Because of recent data showing that GAD-reactive lymphocytes express the gut-specific a4p7 homing receptor (90), investigators have proposed a unifying hypothesis that focuses on the gut immune system (91-93). It has been hypothesized that the protective effect of breast-feeding against the development of type 1 diabetes may be, in part, the result of its role in gut maturation. Breast milk contains growth factors, cytokines, and other substances necessary for the maturation of the intestinal mucosa. Breast-feeding also protects against enteric infections during infancy, and promotes proper colonization of the gut. In contrast, early exposure to dietary antigens, such as cow's milk peptides, may contribute to the loss of tolerance to self-antigens through molecular mimicry. In addition, cow's milk contains bovine insulin, which differs from human insulin by three amino acids (94). Oral exposure to cow's milk formulas appears to induce the formation of antibodies that crossre-acts with human insulin. It also increases cellular responses to bovine insulin in high-risk infants, which is evident at age 3 mo (95). Thus, immune response to dietary bovine insulin may be another early environmental trigger of diabetes autoimmunity. Interesting, enteroviral infections can also interfere with gut immunoregulation, which may explain the epidemiologic associations between viral infections and type 1 diabetes.

The lack of specificity for a particular antigen suggests that type 1 diabetes may involve a general defect in the gut immune system (91-93). Enhanced immune response to early dietary and/or viral exposures may reflect a failure to induce immune tolerance in the gut. Such effects are likely to be mediated by genetic predisposition. Higher levels of antibodies to cow's milk proteins (96) and GAD (97) have been associated with DQB 1*0201, which also increases susceptibility to celiac disease (98). Moreover, the relative risk associated with early introduction of cow's milk (99) and milk consumption in childhood (91) appears to be stronger among individuals with high-risk DQB1 susceptibility alleles than those with low-risk genotypes. These results emphasize the need to match cases and controls by HLA haplotypes in epidemiologic studies of type 1 diabetes.

Was this article helpful?

0 0
New Mothers Guide to Breast Feeding

New Mothers Guide to Breast Feeding

For many years, scientists have been playing out the ingredients that make breast milk the perfect food for babies. They've discovered to day over 200 close compounds to fight infection, help the immune system mature, aid in digestion, and support brain growth - nature made properties that science simply cannot copy. The important long term benefits of breast feeding include reduced risk of asthma, allergies, obesity, and some forms of childhood cancer. The more that scientists continue to learn, the better breast milk looks.

Get My Free Ebook

Post a comment