Autoimmunity and Diabetes
Type 1 diabetes is an autoimmune disease.
Type 1 diabetes does involve a genetic risk, but also generally requires an environmental trigger. Determining the triggers is ongoing (e.g., see these reviews, published in The Lancet: Pociot and Lernmark, 2016; Rewers and Ludvigsson 2016). Various environmental factors may be able to trigger autoimmunity, investigated throughout this website.
The ability of environmental factors to induce autoimmunity is critically dependent on timing. Exposures in utero or early childhood are likely to be most important in type 1 diabetes development. While direct research is so far scarce, it is possible that environmental chemical exposures to the developing fetus or in early childhood may play a role in immune system development, and could potentially contribute to the development of type 1 diabetes later in life.
This page addresses how environmental factors can affect the immune system, focusing on autoimmunity. Studies of twins show that environmental factors are a strong determinant of the immune system throughout life (Brodin et al. 2015).
In an autoimmune response, the immune system attacks the body's own cells. In type 1 diabetes, it is the insulin-producing beta cells that are attacked and destroyed by an autoimmune attack.
Type 1 Diabetes Autoantibodies
T cells are the immune cells that are actually thought to attack the insulin-producing beta cells that reside in the islets (the autoantibodies discussed below are more a marker of the disease and used for diagnosis purposes). These autoreactive T cells, it turns out, are present even in the pancreas of healthy people. In fact they are just as common in those without type 1 as in those with type 1. During type 1 development, they cluster around the islets (Bender et al. 2020). So it seems that maybe there is something about the islets that attract the T cells to them. These factors are discussed more on the Beta Cell Dysfunction page. For more on T cells in type 1 diabetes, see Carré et al. 2021.
Autoantibodies Associated with Type 1 Diabetes
People with type 1 have certain autoantibodies that may appear years before the disease develops, even in utero. These autoantibodies are used as markers of the disease, but do not necessarily lead to or cause beta cell destruction. Some people, in fact, have these autoantibodies but never develop diabetes. The autoantibodies associated with type 1 diabetes include:
islet antigen (ICA) antibodies (present in 90% of people with type 1 at diagnosis);
glutamic acid decarboxylase (GAD) antibodies (present in 73% of patients at diagnosis);
protein tyrosine phosphatase (IA-2) antibodies (present in 75% of patients at diagnosis); and
Zinc transporter 8 (ZnT8A) antibodies (more newly discovered).
There are difficulties testing for islet antibodies, generally present at diagnosis in early-onset type 1 diabetes, but not often present in people with later-onset type 1 diabetes. These antibodies come and go over time. (In people with slow progression to type 1 diabetes, islet antibodies can disappear over time for example (Hanna et al. 2020). GAD antibodies, meanwhile, are present in everyone, with or without diabetes, and a "positive" test means they are higher or lower, depending on an arbitrary cutoff point that is not consistent among studies (Gale 2005).
In Germany, over 90,000 children age 2-5 in the general public were screened for type 1 diabetes-related autoimmunity. Of these, 280 (0.31%) tested positive for 2 or more islet autoantibodies (Ziegler et al. 2020). Rates are much higher among children with a family risk of type 1. Most of the studies below are of people with a family risk.
Autoantibodies over Time
According to a large, international, longitudinal study, children who develop type 1 diabetes generally test positive to two or more autoantibodies. Higher IAA and IA-2 levels, as opposed to GAD antibody levels, appear to especially increase risk of the disease (Steck et al. 2015). The same study shows that autoantibodies at 3-6 months of age were rare, but did occur. Of the 6.5% of children in the study who did develop autoantibodies by age 6, 44% developed IAA only, 38% developed GAD only, 14% developed both IAA and GAD, and smaller percentages had others. IAA peaked in the first year of life and then declined, and GAD peaked in the second year of life and then remained stable. Genetic background also influences the development of these antibodies (Krischer et al. 2015). A German study showed somewhat similar numbers, and that over 50% of the children who developed multiple antibodies developed type 1 diabetes within 10 years (Giannopoulou et al. 2015). German, Swedish, Finnish, and U.S. data show that the more antibodies, the higher the risk of type 1 (Anand et al. 2021). A Finnish study found that IAA antibodies generally appear first, but sometime disappear by the time of diagnosis, whereas IA-2 antibodies are most prevalent at diagnosis (Ilonen et al. 2018).
A Colorado study showed that children who had lower IAA levels and older ages had a slower progression to type 1 diabetes (Steck et al. 2016). These researchers also found that later-onset autoimmunity is more common in African-American or Hispanic children than in Caucasians (Frohnert et al. 2017).
Ethnic differences are also seen in populations with type 1 diabetes from Asia vs Europe; one study found higher GAD antibody frequencies in Europeans, and higher IA antibody frequencies in Asians (Ong et al. 2017). Islet antibodies are linked to type 1 diabetes risk in everyone-- not just those genetically susceptible (Ling et al. 2018).
Interestingly, GAD autoantibodies are linked to a more slow progression of type 1 diabetes development (Ziegler and Bonifacio, 2021).
The ZnT8A antibodies are associated with a more aggressive disease process (Juusola et al. 2016), and are more common in children than adults (Heneberg et al. 2018; Niechciał et al. 2018), and in those under 35 compared to those over 35 (Rogowicz-Frontczak et al. 2018). Some argue that these antibodies could replace the others in the diagnosis of type 1 diabetes (Lounici Boudiaf et al. 2018). Some argue that the ZnT8 proteins (linked to insulin secretion) on the beta cells themselves could be a target of the autoimmune attack. These beta cell surface autoantibodies have been found to appear very early, before other islet antibodies, leading some to propose that these proteins on the beta cell surface might be early targets for autoimmunity (Gu et al. 2021; Jia et al. 2019).
Any second autoantibody increases the risk of diagnosis, and if that 2nd antibody is IA-2A, it confers a greater risk than if the 2nd is GADA or IAA. Additionally, both a younger initial age at seroconversion (testing positive for autoimmunity) and shorter time to the development of the second-appearing autoantibody increased the risk for type 1 development (Vehik et al. 2020).
Autoantibody persistence is also linked to an increased risk of type 1 diabetes. If the antibodies disappear, that reduces risk. Various antibodies tend to appear at different ages in the general population (Pöllänen et al. 2020). Overall, the risk of developing autoantibodies declines with age, even at very young ages (Bonifacio et al. 2021).
The loss of insulin (IAA) antibodies over time is associated with a delayed development of type 1 diabetes (Endesfelder et al. 2016). In general though, once someone has developed multiple autoantibodies, the risk of type 1 remains high even if individual antibodies revert (Vehik et al. 2016). Another study found that people who progressed most rapidly to type 1 diabetes had higher antibody levels, more antibodies, and were younger in age -- or in early puberty (Pöllänen et al. 2017). (see the puberty page for more on type 1 diabetes and puberty). Higher antibody levels have been associated with abnormal glucose levels even before the diagnosis of diabetes (Sanda et al. 2018). In fact, some authors propose calling the time between becoming antibody positive and a type 1 diagnosis as "autoimmune beta cell disorder (ABCD)" (Bonifacio et al. 2017).
Depending on the type 1 antibody, the risk of diagnosis can vary. For example, the risk of developing type 1 was higher in the six months after testing positive for IAA and GAD antibodies (as compared to 3 years after), while for I2A2, the risk did not change over time (Köhler et al. 2017). Among people with 2 autoantibodies, those with GAD antibodies had less risk and those with IA2A had higher risk of type 1 diabetes. Those with IAA and GAD had only a 17% 5 year risk of type 1 diabetes (Jacobsen et al. 2019). Over a longer period of time, 20 years, the risk of testing positive to single to multiple antibodies increases over time, however (Gorus et al. 2017).
According to a TrialNet study, antibody positive identical twins had a 69% risk of diabetes by 3 years compared with 1.5% for autoantibody negative identical twins. In fraternal twins, type 1 diabetes risk by 3 years was 72% in those positive for multiple antibodies, 13% for single antibodies, and 0% for negative antibodies. Siblings had a 3-year type 1 diabetes risk of 47% for for positive multiple antibodies, 12% for single antibodies, and 0.5% for negative antibodies. Younger age, male sex, and genetic predisposition were also significant (Triolo et al. 2018). A further study, of twins where one had type 1 diabetes and one did not, found that in identical twins, by age 20, 14% of the other twin tested positive for islet autoimmunity and 10% for celiac autoimmunity (and 9% and 12% in fraternal twins, respectively). By age 30, 26% of identical and 39% of fraternal twins tested positive for islet autoimmunity. Since the rates of islet autoimmunity are high and similar in both fraternal and identical twins, this suggests a role of possible early environmental factors shared by twins. (Triolo et al. 2020).
The risk of developing autoantibodies drops exponentially with age in children with a first-degree relative with type 1 diabetes, whereas the rate of developing clinical diabetes in children who were islet autoantibody-positive did not decline with age (Hoffmann et al. 2019).
A group of adults without diabetes or autoimmune disease found a percentage (21%) of people tested positive for GAD antibodies. GAD levels were associated with lower beta cell secretion, but not with insulin resistance (Mendivil et al. 2017). GAD antibodies tend to be associated with a slower progression to type 1 diabetes than other antibodies (Ziegler and Bonifacio 2020).
Additional autoantibodies or variants of these autoantibodies are also under investigation as predicting risk of type 1 diabetes (Acevedo-Calado et al. 2019).
The antibodies associated with type 1 can decrease over time in people who already are diagnosed, although tend to be present a year after diagnosis (Cheng et al. 2018). One study found that 30 years after diagnosis, those diagnosed in later childhood tended to still test positive for autoantibodies more than those diagnosed earlier in childhood (Miller et al. 2020). But in general, antibody tests done long after diagnosis may not be positive, even if someone has type 1.
In people who already have type 1, the levels of autoantibodies can correlate with levels of beta cell destruction (Shi et al. 2019; Yamamura et al. 2019). For example, in the TEDDY study, Participants positive for a single autoantibody appeared to have a normal beta cell function. Participants positive for three or more autoantibodies had a lower insulin response as compared to participants with two autoantibodies, supporting the view that their beta cell function had deteriorated (Martinez et al. 2020).
The levels and types of antibodies varies by population and ethnicity. A study of people with diabetes from Ethiopia, for example, did not find any IA-2A antibodies, but did find the others (Siraj et al. 2016). Another Ethiopian study found that most people with type 1 diabetes were antibody positive, especially for GAD antibodies (Balcha et al. 2020). ZnT8A antibodies also vary by ethnic background (e.g., Trisorus et al. 2018). And in people from East Asia, tetraspanin 7 autoantibodies (TSPAN7A) are islet autoantibodies that can be used to identify autoimmune diabetes (Shi et al. 2019). Additional studies also find ethnic differences (e.g., Fawwad et al. 2019 from Pakistan, or Vipin et al. 2020 from India).
A genetic risk score has been developed to identify people of African ancestry who are at increased risk of type 1 diabetes (Onengut-Gumuscu et al. 2019).
Autoantibody Negative Diabetes
Interestingly, a surprisingly high percentage (16%) of children newly diagnosed with type 1 diabetes test negative for the antibodies associated with the disease. The younger the diagnosis, the higher the chance of autoantibodies being positive. Those who tested negative for antibodies had a higher body mass index, and may have a non-immune form of diabetes (perhaps different from either type 1 or type 2) (Wang et al. 2011).
Because his mother and younger brother both have type 1 diabetes, my older son was screened for autoantibodies through TrialNet. These antibodies are a sign of potential diabetes development. So far he has tested negative, thankfully.
Additional Markers of Type 1 Diabetes Development
Other markers are also being investigated to see if they can help predict the development of type 1 diabetes and/or autoimmunity. For example, electrochemiluminescence assays can help predict the risk of both. Those who test positive on this assay are at increased risk of developing multiple autoantibodies, and progressing to type 1 diabetes (Sosenko et al. 2017). Another new biomarker is "IA-2ec autoantibodies," which are "autoantibodies directed to the extracellular domain of the neuroendocrine autoantigen IA-2." (I do not know what any of these markers really mean...). Anyhow, they have been detected in people with type 1 diabetes (and even a small percentage of people with type 2) (Acevedo-Calado et al. 2017). Also, various aspects of the antibodies (aside from just their number) can be used to predict type 1 (So et al. 2021).
Other researchers have found T cells and B cells that show up before autoantibodies as well (Habib et al. 2019; Heninger et al. 2017). Levels of unmethylated DNA, an epigenetic marker, are another potential marker of beta cell death (Zhang et al. 2017). Additional epigenetic biomarkers may also help predict type 1 diabetes and beta cell damage (Bertoccini et al. 2019; Liu et al. 2018). And, other markers (intermediate monocytes) are also linked to the level of residual beta cell function after type 1 diagnosis (Ren et al. 2017). Another biomarker, 12-hydroxyeicosatetraenoic acid (12-HETE), has been found in people newly diagnosed with type 1, but not in people with established type 1 (Hennessy et al. 2017). Antibodies to "post-translationally modified insulin" is another one (Strollo et al. 2017). As are various changes in gene expression (Mehdi et al. 2018; Xhonneux et al. 2021). Other markers are being investigated as well (Carry et al. 2021; Ekman et al. 2019; Kallionpää et al. 2019; Lamichhane et al. 2018; Li et al. 2020; Mitchell et al. 2021; Oras et al. 2019; Salami et al. 2021; Salami et al. 2018; Scherm et al. 2019; Sen et al. 2020; Shapiro et al. 2019; Simmons et al. 2019; Strollo et al. 2019; Vaitaitis et al. 2019; Vecchio et al. 2018; Vecchione et al. 2019; Xu G et al. 2019; Xu X et al. 2019; Yeo et al. 2019; Ziegler et al. 2019).
Some markers are even found in umbilical cord blood, implying that the process leading to type 1 can start even before birth (Lamichhane et al. 2019).
A combination of factors may really help predict type 1 diabetes development; the TEDDY study (The Environmental Determinants of Diabetes in the Young) is trying to identify the combinations that are most accurate (Ferrat et al. 2020; Jacobsen et al. 2019; Krischer et al. 2019). They have identified a variety of metabolic, inflammatory, and autoimmune markers that appear about a year before people test positive for autoimmunity (Balzano-Nogueira et al. 2021).
The innate immune system also seems to be involved in type 1 diabetes development (autoantibodies are part of the adaptive immune system), and this knowledge could be used to help design prevention trials (Petrelli et al. 2021).
All of this research has been fundamental in giving us the possibility of preventing type 1 (Bruggeman and Schatz, 2019). Now that we can predict it so well, we just need to figure out how to prevent it! (See preventT1D.org for more on that!).
Other Autoimmune Diseases Associated with Type 1 Diabetes
About 20% of adults with type 1 diabetes also have another autoimmune disease (at least from one of the largest studies on this topic, from Finland) (Mäkimattila et al. 2020). In the UK, almost one quarter of people diagnosed with type 1 diabetes under 21 years have at least one other organ specific autoantibody (Kozhakhmetova et al. 2018). People with type 1 diabetes are at an increased risk of other autoimmune diseases, thyroid disease, celiac disease, autoimmune gastritis, and Addison's disease (Krzewska and Ben-Skowronek, 2016; Nederstigt et al. 2019). The risk of Addison's disease is about 10 times higher in people with type 1 diabetes (at least in some populations), and it develops at a younger age (Chantzichristos et al. 2018). Interestingly, a U.S. study found that being diagnosed with type 1 at an older age, as well as being female and older age, increases the risk of other autoimmune diseases (Hughes et al. 2018). In Swedish children with islet autoimmunity, thyroid autoimmunity was more common in girls, while in boys, there was a strong association between levels of thyroid and islet autoantibodies (Jonsdottir et al. 2018).
People, especially adults, diagnosed with celiac disease often already have antibodies for type 1 diabetes (and other autoimmune diseases as well) (Tiberti et al. 2020). A free full text review of celiac disease and type 1 diabetes discusses how celiac and type 1 may share common causes. Environmental factors are involved in the development of both diseases because of identical twin studies (a significant percentage of twins do not get the disease even when their twin does), and incidence rates of both diseases have been rising for decades. Also, long-term prospective studies in genetically predisposed infants show that the autoantibody tests can turn positive to both diseases beginning in the first years of life (Goodwin, 2019). Note that celiac disease can be present without symptoms in people with type 1 diabetes, especially in adults (Mahmud et al. 2020). Mass screening has identified a high prevalence of undiagnosed celiac disease autoimmunity in children as well, which is not associated with symptoms (Stahl et al. 2020).
In Greece, over half of people with type 1 or LADA (Latent Autoimmune Diabetes in Adults) had other autoimmune disorders, and these were more common in adults with LADA than in adults with type 1 (Gougourelas et al. 2021).
In Poland, the incidence of additional autoimmune diseases is increasing in children diagnosed with type 1 diabetes (Głowińska-Olszewska et al. 2020).
Additional autoimmune conditions are also associated with type 1 diabetes, including cardiac autoimmunity (Sousa et al. 2019).
Meanwhile, the antibodies associated with type 1 diabetes are linked to a lower risk of eczema, allergies, and asthma in relatives of people with type 1 diabetes (Krischer et al. 2020).
Some people also think that autoimmunity may play a role in type 2 diabetes (Koufakis et al. 2021).
Autoimmune diseases are split into organ-specific and systemic varieties. Type 1 diabetes in an organ-specific disease that affects the pancreas. There are as many as 120 distinct autoimmune diseases, which may affect 50 million Americans.
Natural History of Type 1 Diabetes
Atkinson and Gale (2003) point out that the autoimmune processes that leads to type 1 diabetes can begin very early in life, and that to explore causation, we should consider the developing immune system, genetics, and the timing, duration, and combination of various environmental exposures.
Purcell et al. (2019) summarize recent developments from a NIDDK workshop on autoimmunity in type 1 diabetes, what we know and still don't know.
After Diabetes Develops: The Honeymoon Period
Partial remission, also known as the "honeymoon" period, often-- but not always-- occurs after people are diagnosed with type 1 and start taking insulin (Fonolleda et al. 2017; Zhong et al. 2019). Essentially the beta cells produce some insulin, which tends to make blood sugar levels easier to control. Those diagnosed as adults tend to have longer remissions than those diagnosed at young ages, although every individual is different. One study even found two honeymoon periods in two people diagnosed with type 1 (age 17 and 27)-- these individuals both were diagnosed, started taking insulin, stopped, started again, and stopped again (Li et al. 2017). No one knows why this occurred. A Polish study found that 34% of people who had had type 1 diabetes for an average of 5.6 years still had some level of beta cell function (Kalinowska et al. 2017).
There are a few case studies on this webpage that describe cases of partial or full remission in people who did develop type 1 diabetes, e.g., see the vitamin D page and the wheat and dairy page. Exercise can also extend the honeymoon period (Chetan et al. 2018).
There are also cases of people who can go on and off insulin, e.g., due to drugs or other diseases (van Megen et al. 2017), which is interesting.
Researchers are looking into various markers that are associated with partial remission and early stage type 1 diabetes (Villalba et al. 2019).
The Immune System
The immune system protects the body against pathogens. To accomplish this, immune cells must be able to recognize and fight these pathogens, but also be able to identify which molecules are pathogens, and which are harmless. The immune system is a complex system that controls the balance between defense and tolerance. If the system is altered, the result can be chronic inflammation, including autoimmunity (Lehmann 2017).
A number of environmental factors can affect the body's immune system. Some are thought to be supportive to the development and proper functioning of the immune system, such as vitamin D, breastmilk, and omega-3 fatty acids. These factors have sometimes been shown to be protective against type 1 diabetes development as well, perhaps due to their immune system effects.
Other factors can have detrimental effects on the immune system. Viruses, for example, fall into this category (although they may sometimes show protective effects as well). A number of chemicals can have toxic effects on the immune system. Some of these chemicals may influence the progression of autoimmune disease. The ability for chemical exposures to affect autoimmunity appears to depend on genetic susceptibility, the duration of exposure, and the timing of the exposure (Inadera 2006), which appears to be the case for viruses as well (van der Werf et al. 2007).
NIEHS Report on Autoimmunity
The National Institute of Environmental Health Sciences held a workshop to evaluate scientific evidence of environmental factors in autoimmune disease. Their report is available online: Epidemiology of environmental exposures and human autoimmune diseases: findings from a National Institute of Environmental Health Sciences Expert Panel Workshop (Miller et al. 2012).
The NIEHS found that: "There has been virtually no epidemiologic research on risks associated with relatively widespread synthetic chemical exposures, such as plasticizers (e.g., phthalates and bisphenol A). Some of these chemicals can act as endocrine or immune disruptors, and increased risks of some immune-mediated diseases, including asthma and eczema, in relation to exposure levels, have been reported in children. More research is needed to determine the role of plasticizers and other industrial chemicals in consumer products in the development of autoimmune disease."
The NIEHS states that: "The growing number of genome-wide association studies and the largely incomplete concordance for autoimmune diseases in monozygotic twins support the role of the environment (including infectious agents and chemicals) in the breakdown of tolerance leading to autoimmunity via numerous mechanisms." These potential mechanisms are described in the report, Mechanisms of environmental influence on human autoimmunity: a National Institute of Environmental Health Sciences expert panel workshop (Selmi et al. 2012).
Environmental pollutants can alter the immune system by affecting the function of immune cells in such a way that they might react against the body's own tissues. These are a known effect of pesticides, heavy metals, wood preservatives, and volatile organic compounds (Lehmann 2017). In fact, many types of chemicals can enhance the immune system, depending on the levels, timing, and route of exposure. New research is focusing on the impacts of chemical exposure on the immune system during development in early life (Kreitinger et al. 2016). Exposures to various environmental chemicals in the womb and during early life may promote autoimmunity, perhaps via epigenetic mechanisms (Blossom and Gilbert, 2018). A variety of mechanisms may be involved in the appearance of autoimmune disease after toxic chemical exposures (Kharrazian 2021).
There is some evidence that environmental exposures to adults may result in other autoimmune diseases (besides type 1 diabetes). Studies of occupational exposures have associated various autoimmune diseases to silica, solvents, pesticides, and ultraviolet radiation, although data are limited (Cooper et al. 2002). Thyroid autoimmunity is linked to environmental chemical exposures for example (Benvenga et al. 2020). A variety of environmental chemicals are linked to changes in the immune system, including autoimmune disease (Nowak et al. 2019; Popescu et al. 2021), and exposure to these chemicals during development can promote autoimmunity as well (Rychlik et al. 2019). While the mechanisms by which environmental chemicals can trigger autoimmunity are still being worked out, it probably involves things like inflammation or damage to specific organs (Moloudizargari et al. 2019; Pollard et al. 2018). Endocrine disruption and hormones may also play a role in the development of autoimmune diseases relating to environmental chemical exposures (Merrheim et al. 2020).
Scientists are now developing screening tools to help identify which environmental chemicals are toxic to the immune system. (Chemicals can affect the immune system in other ways, aside from autoimmunity, for example by suppressing the immune system) (Boverhof et al. 2014).
Sex differences can affect the ability of chemicals to influence autoimmunity, which may help explain why more women than men develop autoimmune diseases (Edwards et al. 2018).
Exposures During Development
Developmental immunotoxicology is a subfield of immunotoxicology that looks at the effects of exposure to various environmental factors (including toxic chemicals, drugs, viruses, etc.) on the developing immune system (beginning in the womb through childhood). These effects may occur during childhood or in adulthood, and include autoimmunity, immunosuppression, allergic responses, and inflammation (Dietert 2009).
A disturbance in the early development of the immune system may allow an autoimmune reaction that would have been suppressed to flourish instead, and contribute to the accelerated progression of type 1 diabetes (Gale 2005).
In a fetus, the immune system develops in particular stages during particular times. An environmental exposure during one of those critical times can disrupt the development of the immune system, and lead to lifelong effects. Dietert and Piepenbrink (2006) point out a number of important factors to consider with regard to exposures during gestation or in childhood:
The doses required to produce an effect during development of the immune system are significantly lower than those required to affect an adult.
The effects of the exposures depend on timing of the dose, if given during a critical period of immune system development.
The effects of exposures on a fetus, or during early childhood, can be different from those seen in adults. The perinatal period (the months before and after the time of birth), is a sensitive time for the developing immune system. Adult exposure, therefore, does not necessarily predict the effects of exposure to a fetus, during the perinatal period, or in early childhood.
The effects of early life exposures often last long after the exposure, sometimes having lifelong effects. And, early exposures may be "latent" and invisible, until stressed later in life (for example, by a virus), causing abnormal immune responses.
Response can differ by gender; there are many cases where males and females have differing immune responses to early environmental exposures.
A number of environmental chemicals can be toxic to the developing immune system, including various persistent organic pollutants such as chlordane, PCBs and dioxin; heavy metals such as mercury and cadmium, benzo[a]pyrene, diazinon, bisphenol A, trichloroethylene, polycyclic aromatic hydrocarbons (PAHs), diesel exhaust, and some pesticides (e.g., atrazine) (Holladay 1999; Dietert and Dietert 2007). Some chemicals have been shown to increase the risk of autoimmunity in particular, including mercury, lead, gold, trichloroethylene, hexachlorobenzene (HCB), and dioxin (TCDD) (Dietert et al. 2010), as well as PCBs, bisphenol A, and phthalates.
The Timing Makes the Poison
For centuries, toxicology has been based on the idea that "the dose makes the poison." In other words, something might not be toxic at low levels of exposure even if it is toxic at high levels. But for exposures during early development, another paradigm has emerged, that the "the timing makes the poison." A number of animal studies show that events that occur during immune system development may be responsible for disease later in life. Some of these events are seen with environmental chemicals at levels similar to those humans are exposed to in the environment (Grandjean et al. 2008). For example, exposing mice during development to a mixture of 23 chemicals found in fracking operations exacerbated autoimmunity later in life (Boulé et al. 2018).
Exposure during early development is linked to later-life development of autoimmunity. Other periods of life are also times of vulnerability to autoimmunity in women-- puberty, pregnancy, and menopause, for example (Desai and Brinton, 2019). All of these periods correspond to major changes in the body's hormone levels, implying hormone levels can interact with the immune system to promote autoimmunity.
Timing may be critical for the effects of immunosuppressive substances. Immunosuppression is often considered to be the opposite of autoimmunity, which involves a hyperactive immune system. Yet both might be the result of a disruption in the balance of the immune system. Some environmental exposures, in fact, have been associated with both increased autoimmunity and decreased immune competence (Hertz-Picciotto et al. 2008). For example, exposure to mercury was found to first lead to immunosuppression, and then to autoimmunity in mice (Havarinasab and Hultman 2005).
The adverse outcomes of developmental and adult immunotoxicity include autoimmunity. In the past, most studies have focused on immunosuppression. The detection of immune suppression is useful, but ignores other health outcomes. Toxicity testing currently only focuses on immunosuppression, not autoimmunity.
Source: Dietert et al. 2010, EHP.
Regulatory T-Cells and the Thymus
How might early exposure to environmental factors lead to autoimmune diseases such as type 1 diabetes later in life? There are a number of possible mechanisms, many of which are too complicated to describe here (e.g., see Dietert and Piepenbrink 2006).
One possible mechanism might involve "regulatory T cells," or T regs. T regs are one type of immune system cell that are involved in destroying autoreactive cells (the cells that are attacking the beta cells, for example). During gestation, T regs are produced in the thymus, and learn how to function. Disruption of the biological processes that lead to the production and activation of T regs could be a significant factor in later autoimmune disease. The thymus is critical for the development of the immune system in utero, and plays a major role through childhood, although it plays a more minor role later in life (Dietert and Piepenbrink 2006). In fact, removing the thymus of newborn mice can lead to autoimmune disease (Sakaguchi 2004). Geenen (2021) reviews the role of the thymus in autoimmunity. Researchers are still trying to figure out how exactly T regs play a role in autoimmune disease (Brown et al. 2020).
A number of environmental chemicals can alter the fetal thymus, by shrinking it or by affecting the development of thymocytes. Affecting the maturation of thymocytes may have effects later in life, and possibly lead to autoimmunity (Holladay 1999).
Researchers are beginning to study the potential role of T regs in the development of type 1 diabetes. One small study, for example, has found that some children with type 1 diabetes have lower percentages of T reg cells as compared to children without diabetes (Luczyński et al. 2009). More recent studies, however, found T regs were higher in children with type 1 diabetes, and correlated with disease onset. However these changes were not found in those who were autoantibody positive, implying that they are unrelated to the onset of autoimmunity itself (Viisanen et al. 2019). Others have found defects in the T regs of people with type 1 diabetes (Pellegrino et al. 2019). Treatments that involve inducing T regs are now being tested for use in people with type 1 diabetes (Orban et al. 2010). Further research into how chemicals can affect T regs and immune system development and their potential relation to type 1 diabetes are clearly in order.
To see articles on the role of autoimmunity in diabetes, as well as articles about various environmental factors that can initiate or influence the progression of autoimmunity, see my PubMed collection Autoimmunity.