Type 1 diabetes (T1D) results from autoimmune damage to the insulin-producing beta cells in the pancreas, and is caused by a complex combination of genetic and environmental factors. It is estimated that over 75,000 children between 0-14 years develop T1D each year, with up to 30 million people affected worldwide. Finland, Sweden, the USA and Canada have some of the highest incidences of T1D in the world. The exact causes for T1D remain a mystery, but improved hygiene has been associated with a higher frequency of T1D and other autoimmune diseases in human populations.
Researchers have used several different models for studying T1D, and the non-obese diabetic (NOD) mouse remains an excellent model for T1D pathogenesis. NOD mice are genetically susceptible to spontaneous autoimmunity, demonstrate a higher incidence of T1D under clean hygiene status, and present a greater than 2:1 female-to-male sex bias in disease development. The sex bias in T1D development in NOD mice is particularly interesting, as many human autoimmune diseases also exhibit a strong sex bias, presenting 50-80% more frequently in females than in males.
Graduate student Janet Markle, Professor Jayne Danska and their collaborators have recently used the NOD mouse to explore a novel intersection of the gut microbiome and sex hormones (the “sex-specific microbiome”) on T1D etiology. Their results were published in a groundbreaking report in Science in March this year.

In their article, Markle et al. reveal that gut colonization early in life influences the pathogenesis of T1D in a testosterone-dependent manner. Male NOD mice housed in specific pathogen–free (SPF) conditions were protected from disease relative to females, but this sex bias disappeared under germ-free (GF) conditions. In SPF NOD males, normal colonization of gut commensals elevated serum testosterone levels and is associated with T1D protection that develops well after puberty. The investigators transferred adult male gut microbes by oral gavage into weaning age SPF-housed female recipients. Surprisingly, they reported that colonization of young females with gut microbiota from adult males increased serum testosterone levels, resulted in reduced islet inflammation and autoantibody production, and that these effects disappeared under the influence of flutamide, an androgen-signaling antagonist. Compared to splenic T cells from un-manipulated NOD females, splenic T cells from female gavage recipients displayed a much longer latency to T1D development when transplanted into NOD.SCID (severe combined immunodeficient) recipients, and this delayed disease effect also required androgen receptor activity. Thus, in genetically high-risk individuals, commensal communities in the gut regulate autoimmune disease on many levels acting synergistically with sex hormones.
In interviews for IMMpress Magazine, Markle and Danska discussed how the different pieces of the puzzle came together in this work, the resulting impacts on the field, and future directions for studying the sex-specific microbiome.
CONNECTING THE DOTS: Discovering a novel intersection of microbiome and sex in T1D
The Danska lab has long been involved in elucidating genetic elements controlling T1D susceptibility. However, Danska also takes a deep interest in the environmental influences on autoimmune diseases. “The annual incidence [of T1D] has been increasing substantially in the developed world over the last 50 years […],” says Danska, “and the age of onset is getting substantially younger. [So], there clearly are environmental features affecting the very young, developing immune system.”
Thus began a project to create GF NOD mice, initially to explore gene-environment interactions at T1D susceptibility loci. When Markle was introduced to the project, she was immediately interested: “At that point, I had never read a paper about microbiome modification of immune development or responses; it was an emerging field.” Their lab had been collaborating with Kathy McCoy and Andrew Macpherson to re-derive the NOD strain into GF conditions at McMaster University. Strikingly, the sex-bias to T1D susceptibility in NOD mice was eliminated under GF conditions, suggesting that the commensal biota must confer protection in concert with the sex of the animal. Markle went on to show that gut colonization in SPF males elevated serum testosterone levels and altered the levels of multiple long-chain fatty acid metabolites. Furthermore, her time course study of gut microbiomes showed that sex-specific differences in composition arose only after puberty, suggesting a positive feedback loop between the microbiota of male mice and testosterone levels.
Upon joint evidence that the sex bias in T1D requires microbial colonization and had a direct effect on sex hormones, Danska became “very, very excited. This was the intersecting point of what I’d been interested in for so many years.” The next goal was to discern whether the biota of female mice could be changed into a more “male” state that could in turn elicit protection against T1D. However, the GF NOD colony was no longer available, and McCoy and Macpherson had moved to Switzerland. Reflecting on this stage Danska says, “Science, like life, is dependent on all of these moving pieces, and you have to do the best you can with the circumstances.” So rather than giving up, they pursued the transfer of microbiota in SPF mice, at the weanling stage, a time when gut colonization is changing dramatically as the mice transition from breast milk to solid food.
[pullquote]“Science, like life, is dependent on all of these moving pieces, and you have to do the best you can with the circumstances.”[/pullquote]
Markle gavaged female mice with cecal bacteria from either adult NOD males (M→F) or adult NOD females (F→F), quantified hormone levels, and followed pre-diabetes phenotypes up to 35 weeks of age, to capture the time and frequency of spontaneous diabetes onset in NOD mice. She observed that the composition of M→F female recipients’ microbiota was significantly altered compared to controls. These recipients also displayed elevated serum testosterone levels and altered composition of long-chain fatty acid metabolites. Significant changes in microbiome composition were still evident at 14 weeks of age when the M→F female recipients had reduced insulitis severity and insulin autoantibody titers relative to either unmanipulated females or F→F recipients. These changes were significantly blunted in M→F recipients when androgen signaling was antagonized by flutamide treatment demonstrating their dependence on elevated testosterone.
In line with these exciting observations, Markle transferred splenic T cells from six week-old M→F female recipients into lymphocyte-deficient NOD.SCID mice to test the latency of T cell-mediated T1D after microbiome manipulation. The M→F donor T cells had reduced potential to transfer diabetes compared to unmanipulated females, as measured by blood glucose levels. This phenotype was also reliant on an androgen-dependent mechanism. In sum, testosterone activity was crucial to the ability of the male microbiota to protect females from insulitis progression, autoantibody production, and to suppress the diabetogenicity of T cells in NOD mice.
THE LONG AND WINDING GUT: Potential mechanisms in the sex-specific microbiome
Markle et al.’s study is the first to show that early alterations to gut commensals can drive changes in sex hormones, metabolism, and cellular immunity that affect the onset of autoimmune disease in NOD mice. However, an outstanding question remains: what pathways or mechanisms are connecting these diverse elements?
Metabolism in T1D
It has long been understood that the gut microbiome affects host metabolism. In Markle et al.’s study, the transfer of gut bacteria from a male NOD mouse altered serum concentrations of glycerophospholipids and sphingolipids in female recipients. The enzymology responsible for the normal synthesis of long-chain fatty acids is liver-based. Thus to explore how the microbiota may be altering metabolism, Danska’s lab will perform epigenetic and RNA-seq analyses on relevant transcription factors and enzymes in the livers of transplanted female mice.
Of note, the dysregulation of lipid and amino acid metabolism is known to precede seroconversion to produce beta cell autoantibodies in predisposed human children, and it has been suggested that those genetically susceptible to T1D are also more prone to metabolic stress and oxidative damage in pancreatic beta cells. How relevant enzymes are sensitive to testosterone regulation remains an important question in this context.
Testosterone in T1D
In terms of testosterone levels, one can first ask, is it the utilization of testosterone or the generation of testosterone that has been changed by the microbiome in Markle et al.’s studies? “We don’t fully have the answer to that yet,” answers Danska, “but the immediate precursor of testosterone [in M→F female recipients] is also elevated, so it suggests that the production of testosterone is elevated.” Markle adds, “[one of the next steps would be] to individually interrogate the organs that are sources of estrogen and androgen in female mice.” Does the protective phenotype transferred to females stem from changes to the pituitary gland, adrenal cortex, or ovaries? She concludes, “We haven’t honed in on the cellular targets of testosterone signaling that are critical to our T1D phenotype in NOD mice, [but we do know] that testosterone level, metabolic phenotypes, and T cell-driven autoimmunity were all directly affected by microbiome manipulation.”
Cellular immunity in T1D
In Markle et al.’s studies, the diabetogenicity of T cells in M→F female recipients was clearly attenuated. It is unclear whether this systemic change reflects a difference in the repertoire of the T cells, or a difference in their effector differentiation programs, or both. However, Markle examined the frequency of T cells expressing RORgt, Foxp3, and c-myb in gut-associated lymphoid tissues of the control and M→F female recipients, and did not find microbiome-dependent differences in the frequency of T cells at these sites. Experiments investigating the functional potency of T cell subsets will help clarify where the cellular effect is occurring. According to Markle, “the broad metabolic changes we observed still support the notion that microbiome-host crosstalk affects diverse phenotypes well beyond the mucosa.”
Notably, the microbiome has been shown to influence distant immune compartments. For example, bone marrow-derived neutrophils are systemically primed for enhanced killing by the peptidoglycan produced by normal gut biota. In this context, peptidoglycan from gut microbiota is constantly translocated across the gut mucosa into circulation, where it can eventually accumulate in the bone marrow. Could bacterial products specific to the microbiome of males affect T cell development or priming and alter T1D onset in a similar fashion? “Does the neonatal establishment of the microbiome influence thymic T cell output?” muses Markle, “this is an interesting question, and one we have not yet attempted to address […] However, the T cells are clearly altered in some way. It could also be the TCR repertoire, but I didn’t test this.” Danska hopes to explore this concept further by repeating the biota transfers in TCR transgenic NOD mice.
‘TRANSLOCATING’ TO THE CLINIC: Are microbiotic approaches for preventing autoimmunity in reach?
Clearly, changes to commensal structure at an early time can lend protection against T1D in the NOD mouse model. One of the most exciting aspects of this work is that this protection extends long beyond the initial manipulations to the microbiome. This is particularly exciting for Danska because NOD mice carry “a huge genetic liability. 85% of [unmanipulated] NOD females will get this disease. So to be able to block that with one manipulation of a colonized, just-weaned animal, suggests that there is a window of time that you can impose changes on the immune system that are going to be beneficial.”
The Danska lab is collaborating with Emma Allen-Vercoe at Guelph University to create pre-clinical models to validate human microbiome transplant therapies for prevention of T1D in at-risk children. Allen-Vercoe’s group has recently cured two elderly patients suffering from chronic Clostridium difficile infection by colonic transplantation of a lab-grown consortium of healthy donors’ gut microbiota. Poor diversity within the commensal bacterial community provides the opportunity for enteric C. difficile to take over the gut real estate, but re-establishing more beneficial microbial communities cures this infection, in a durable fashion [see separate box]. Allen-Vercoe’s proof of concept study offers promise to advance microbe treatments to the clinic, and in Danska’s eyes, future prevention and treatment approaches to autoimmunity. She hopes to isolate the consortium of bacteria responsible for the T1D protection in male NOD mice, and use it to model a microbiome-based approaches for early intervention in autoimmune disease development in genetically high-risk individuals.
MOVING FORWARD: Impacts on the field and future directions for Markle
Markle and Danska’s work has already gotten the field talking. When asked about the initial responses from the scientific community, Markle describes, “Jayne presented the bulk of the work at a major conference we attended and it created some buzz, so we decided to try for Science […] Post-publication, [we] both experienced a bump in email volume [so] it has been gratifying to see that the work is being communicated to the public.” It certainly will be exciting to see how future work on this project will shed light on the basic biology at work in autoimmune diseases, and ultimately, how progress in translational medicine can be made.
Overall, Markle is grateful for the opportunity to work on such a promising project. In her words, “[I’ve learnt] some important lessons about how to conduct animal-based research. For example, pay attention to sex! […] Few researchers consider this when designing experiments. And you should care about what microbes are colonizing the mice in your studies.” Indeed, Danska is quick to sing praises about Markle’s role in the project. “[Her] work has been fundamental in all of these things we’ve done […] She is incredibly astute in her ability to think about not just what you do, but how you do the science […] She was an outstanding student because of her tenacity.” When asked how she did, in fact, stand strong in the face of adversity, Markle credits IGSA pub nights, long runs along Lake Ontario, and above all, the excellent training environment in Danska’s lab and within the Department. “I live in NYC now, a city that prides itself on the expression, ‘If you can make it here, you can make it anywhere.’ The Danska/Guidos lab is kind of like that,” Markle reflects fondly.
At her new postdoctoral fellowship position in Jean-Laurent Casanova’s lab at Rockefeller University, Markle works on the human genetics of severe extra-pulmonary Mycobacterium tuberculosis infection in children. Her advice for current graduate students still interested in the academic route: “Start contacting potential postdoc supervisors early […] Get the most that you can out of the grad school experience, and then move on.” Regarding her own moving-on experience, Markle explains, “I am really enjoying the Rockefeller University community, where people are incredibly open and supportive [of my research]…I’m a scientist at heart and nothing else appeals to me more than the academic route.” The future certainly looks bright for Markle. “For the moment,” she concludes, “I’ll concentrate on having a productive postdoc, and hope that a robust scientific community still exists in Canada when it comes time to try and join it.” For the benefit of the Canadian scientific community, we certainly hope she returns.
References:
Markle, J.G.M., Frank, D.N., Mortin-Toth, S., Robertson, C.E., Feazel, L.M., Rolle-Kampczyk, U., Bergen, von, M., McCoy, K.D., Macpherson, A.J., and Danska, J.S. (2013). Sex Differences in the Gut Microbiome Drive Hormone-Dependent Regulation of Autoimmunity. Science 339, 1084–1088.

Catherine Schrankel

Latest posts by Catherine Schrankel (see all)
- Issue 1 2016 – Front Cover - June 13, 2016
- Benchmarking IMMpress Magazine – Year 3 - January 15, 2016
- Sciencefreude: New German Words for the Academic Condition - January 15, 2016