Microbiome research in Autoimmune and Immune-Mediated Inflammatory Diseases: Lessons, Advances and Unmet Needs

Abstract

The increasing prevalence of autoimmune and immune-mediated diseases (AIMDs) underscores the need to understand environmental factors that contribute to their pathogenesis, with the microbiome emerging as a key player. Despite significant advancements in understanding how the microbiome influences physiological and inflammatory responses, translating these findings into clinical practice remains challenging. This viewpoint reviews the progress and obstacles in microbiome research related to AIMDs, examining molecular techniques that enhance our understanding of microbial contributions to disease. We discuss significant discoveries linking specific taxa and metabolites to diseases such as rheumatoid arthritis (RA), systemic lupus erythematosus (SLE), and spondyloarthritis (SpA), highlighting the role of gut dysbiosis and host-microbiome interactions. Furthermore, we explore the potential of microbiome-based therapeutics, including fecal microbiota transplantation and pharmacomicrobiomics, while addressing the challenges of identifying robust microbial targets. We advocate for integrative, trans-disease studies and emphasize the need for diverse cohort research to generalize findings across populations. Understanding the microbiome’s role in AIMDs will pave the way for personalized medicine and innovative therapeutic strategies.

Introduction

Despite the known genetic risk factors of autoimmune and immune-mediated diseases (AIMDs), the steady increase in prevalence of these conditions suggests that their pathogenesis is partially mediated by environmental factors¹. Among those, the microbiome has emerged as a central player². While fundamental discoveries have revealed how the microbiome modulates physiology and inflammatory responses, advances in the development of microbiome-based diagnostics and therapeutics in autoimmunity have yet to reach daily clinical practice. In this viewpoint, we review progress and challenges to date, discuss approaches to improve our understanding of microbiome mechanisms of pathogenesis in AIMDs, and describe strategies to translate findings into actionable knowledge for patient care.

The microscopic elephant: a combination of molecular techniques shape our understanding of the microbiome

As in the parable of the blind men and the elephant, multiple molecular approaches are required to understand how microbes contribute to AIMDs. The use of 16S rRNA gene sequencing has been a common approach to identify and quantify bacterial taxa, given its sensitivity, moderate cost, and quick turnaround for analysis. However, this method is limited in its ability to resolve taxonomic differences at the species and strain level and does not offer direct functional information. In contrast, shotgun metagenomics improves resolution and identifies microbial taxa, including bacteria, viruses and fungi, as well as genomic variations. This is particularly relevant, as recent work has demonstrated that genomic variations in specific taxa are relevant in AIMDs³. Metagenomic sequencing can also identify genes and pathways encoded in microbial genomes, which has proven to be highly relevant in inflammatory bowel disease (IBD) and other AIMDs⁴, although it cannot determine if those functions are expressed. Metatranscriptomics can resolve both microbial composition and activity, providing an additional layer of biologically relevant information⁵. Combining sequencing with flow cytometry refines microbiota characterization by identifying viable (vs. dead) microbes and those tagged by immunoglobulin A (IgA-Seq). While these techniques primarily target bacteria, sequencing commensal fungi and viruses also provides insights into microbial-host interactions in autoimmune conditions. Thus, multiple “omic” technologies shape our knowledge of microbial contributions to AIMDs.

Microbiome in Autoimmune Diseases

Animal models and patient studies have shaped our understanding of microbial contributions to AIMDs. In systemic lupus erythematosus (SLE), notable findings have reported the translocation of a gut pathobiont as driver of disease in both mice and humans⁶, as well as the identification of gut commensals with orthologs of human Ro60 that can lead to cross-reactive immune cells targeting the human version and initiating lupus-like disease⁷.

Since our original publication more than a decade ago⁸, the rheumatoid arthritis (RA) microbiome field has witnessed remarkable translational advances. Larger cohorts have expanded our ability to identify significant effects and integrated the use of metagenomics⁹, while work incorporating host genetics and metabolomics has provided additional mechanistic insights. In a large UK Twins study, Prevotella spp was associated with RA genotype in the absence of clinical disease, including in individuals at high risk of developing RA¹⁰. This suggests that host genotype (HLA-DR risk alleles) is associated with microbiota before and immediately after disease onset, an effect that we and others have observed in multiple inflammatory arthritides^8,11,12. Given its relevance to the field, we encourage the integration of microbiome data with HLA and non-HLA risk alleles in clinical studies to characterize their synergistic effects in AIMDs pathogenesis.

Intriguingly, and likely mediated by gut microbial dysbiosis, metabolomic perturbations such as lower levels of short-chain fatty acids, including butyrate, are already observed in individuals at high risk that ultimately transition to arthritis^13,14. Similarly, new onset RA patients have high abundance of butyrate-consuming microbes in the gut, while dietary butyrate supplementation conferred anti-inflammatory benefits in murine models of arthritis^15–17. This is of high clinical importance, as metabolites have the potential to be used for diagnostics and/or manipulated as therapeutics.

Arguably, microbiome studies in spondylarthritis (SpA) began four decades ago when Taurog and colleagues observed that, under germfree conditions, HLA-B27 transgenic rats do not develop gut and joint inflammation¹⁸, which was later confirmed in other SpA models. In patients with axial and peripheral SpA, several studies have reported microbiota correlations with disease activity as well as concomitant subclinical gut inflammation^19,20. Similarly, the use of IgA-Seq has identified novel bacteria associated with IBD and disease progression²¹, and has demonstrated that patients with Crohn’s disease-associated SpA (CD-SpA) have enrichment in IgA-coated Escherichia coli compared to patients with CD alone. Further, colonizing K/BxN mice with these CD-SpA-derived E. coli strains led to more severe inflammatory arthritis. In patients with psoriatic arthritis (PsA), we and others have shown a gut dysbiotic process at disease onset⁹ and the interplay of microbiota and different biologics during treatment and progression to comorbid intestinal inflammation²². Taken together, these studies show that microbes and/or their products likely contribute to a spectrum of AIMDs.

Microbial Therapeutics: From FMT to Pharmacomicrobiomics

The malleability of the microbiota in response to interventions makes it a promising therapeutic avenue. Indeed, previous studies have shown encouraging results in other conditions when modifying microbial communities through the use of antibiotics²³ or diet²⁴. In AIMDs, however, there is still a lack of robust, well-validated microbial targets, which could explain the sparsity of microbiome interventions in AIMDs to date.

Microbial therapeutics, particularly through the use of fecal microbiota transplants (FMTs), has been pioneered in refractory Clostridiodes difficile infections (rCDI) and ulcerative colitis (UC). In rCDI, FMTs have been widely successful. In UC, a multifactorial disease with genetic and environmental risk factors, results have been positive with FMTs but less definitive with microbial products based on defined communities¹². In rheumatic diseases, studies to manipulate the gut microbiota are emerging. FMT from preclinical RA individuals into CIA mice leads to intestinal barrier dysfunction, alterations in mucosal immunity and development of arthritis¹³. While anecdotal evidence has shown improvement in PsA activity post-FMT for CDI, the first double-blind placebo-controlled randomized study (FLORA) found worse response in the FMT group than in the placebo group²⁵. Despite the limitations of this proof-of-principle trial (small sample size, single gastroscopic-FMT, lack of microbiome analysis to measure actual engraftment), subsequent work suggests that increased intestinal permeability and absence of gut metabolites correlate with failure to FMT²⁶. The challenges associated with whole community FMTs, including concerns about safety and the complexity of accurately dosing when using fecal material²⁷, imply that their use in clinical practice might be limited, and defined communities and/or metabolite interventions are more likely to be implemented in the future.

Another important advance in the field has been the development of pharmacomicrobiomics, the study of how microbiome variation modulates drug disposition, action, and toxicity. The cancer field has pioneered clinically-relevant results, demonstrating that the gut microbiome prior to treatment can predict response to immunotherapies, and that altering microbial composition can lead to enhanced response to immunotherapy^28–30. It is plausible that the microbiota composition only serves as a surrogate for active inflammation, rather than being a marker of lack of response to therapy. To overcome this challenge and using metagenome sequencing in new-onset RA patients, our group reported for the first time that the pretreatment microbiome predicts response to oral methotrexate (MTX). Importantly, we further demonstrated experimentally that MTX is depleted by human gut bacteria ex vivo and that, when incubated with fecal samples, MTX levels measured in the supernatant correlate with future clinical response. A recent follow-on longitudinal observational study in DMARD-naïve, newly diagnosed RA patients during introduction of treatment³¹ has further demonstrated that pre-treatment microbiome differentiates response to DMARD. These findings suggest that the gut microbiome can serve as a potential biomarker of clinical response to DMARDs, an observation that has also been reported in other AIMDs treated with biologics^32,33.

A Global Understanding of the Microbiome in Autoimmune Diseases

Microbiome work to date has generally focused on AIMDs in isolation, using specific populations in studies based on single centers. We argue that studies across AIMDs and in different cohorts are required to generalize mechanisms of pathogenesis and protection.

Although AIMDs have distinct phenotypic presentation, the overlap in response to common anti-inflammatory treatments suggests that underlying molecular pathways might be partially shared. Trans-disease studies are currently underway in the US through the NIH/FNIH-sponsored AMP AIM Consortium, as well as in Europe through the Innovative Medicines Initiative (IMI)-sponsored Taxonomy, Treatment, Targets, Remission (3TR) consortium to study SLE, RA, multiple sclerosis and IBD among other AIMDs (Table 1). AMP AIM investigates the immune and microbial factors of early RA, SLE, psoriasis, PsA and Sjögren’s disease. Within this consortium, the Microbiome Technology and Analytic Center Hub (Micro-TeACH) has been created with the goals of: 1) identifying microbes and microbial pathways as potential determinants of AIMD initiation and perpetuation, 2) interrogating host-microbe spatial interactions in affected tissues, and 3) discovering microbiome-based precision medicine approaches. A critical strength of the AMP AIM consortium is the ability to integrate microbiome results with other technologies, such as single-cell RNA-Seq, spatial transcriptomics or metabolomics, providing a comprehensive landscape of pathogenesis that incorporates both microbial and host factors. Importantly, the Human Ocular Microbiome initiative has also been launched to characterize the microbiome on the ocular surface and examines how microbes residing in other parts of the body might affect the eye. This initiative is integrated with the AMP AIM network and promises to enhance our knowledge of AIMDs that affect the eye, including Sjogren’s disease and uveitis.

TABLE 1.

Ongoing Multicenter, Multi-Stakeholder Research in Microbiomics Related to AIMDs.

Microbiome Research Efforts in AIMDs	Funding Source/s	Geographic Region	Specific Center	Scope	Website
AMP AIM Consortium	FNIH, NIH, Industry	USA, UK, Australia	Microbiome Technology and Analytic Center Hub (Micro-TeACH)	RA, SLE, Psoriatic Disease, Sjogren’s Disease	https://www.niams.nih.gov/grants-funding/niams-supported-research-programs/accelerating-medicinespartnership-amp
Human Ocular Microbiome initiative	NIH	USA	N/A	Health, Uveitis, Sjogren’s Disease	https://grants.nih.gov/grants/guide/rfa-files/RFA-EY-22-001.html
Taxonomy, Treatment, Targets, Remission (3TR) consortium	Innovative Medicines Initiative (IMI), Industry	Europe	3TR Chronic Autoimmune Diseases and IBD Groups	RA, SLE, MS, IBD	https://3trimi.eu/research/diseases
NIH Integrative Human Microbiome Project (iHMP)	NIH	USA	Inflammatory Bowel Disease Multi-omics Database (IBDMDB)	IBD	https://ibdmdb.org

Another important unaddressed question is whether findings from these studies will be shared in different populations. Most AIMDs affect women disproportionately, while race and ethnicity are known to be significantly associated with disease risk³⁴. Human populations exhibit significant differences in their microbiome^35,36, and bacterial species and strains are shared within family members but not in unrelated people³⁷. Integrating novel datasets between European and North American initiatives as well as expanding the generated knowledge of the microbiome in AIMDs to cohorts with a broader ethnic/racial representation, i.e. non-Western populations, is necessary to distinguish shared mechanisms of pathogenesis from those unique to specific groups³⁸.

Prospectus

While these initial results are promising, results from well-powered, multi-center studies will provide a unique opportunity to understand how the microbiome is associated with pathogenesis, disease progression and response to treatments in AIMDs. Expanding these findings to diverse populations will be fundamental for health equity and to provide truly personalized medicine in an effective manner. Ongoing efforts to close the knowledge gap include the use baseline microbes to predict treatment response at the point-of-care, manipulating available treatments using microbes or metabolites as co-adjuvants, improving methods that accurately capture microbial populations in low biomass samples (e.g. ocular and skin microbiome) and utilizing functional platforms coupled with host genetics/epigenetics and immune response to help advance our mechanistic understanding of autoimmune diseases.