More Questions Than Answers: The Role of the Microbiome in Rheumatic Disease

There are more questions than answers about the role of the human microbiome in rheumatic diseases. Moreover, there are more questions than answers about the microbiome itself, and its role in and on the human body. Experts are beginning to understand how the microbiota interact with the host, how the thousands of species of microbes communicate with each other and how all of these various interactions contribute to both health and disease.

However, the body of research is growing, albeit slowly. For example, it is generally accepted that microbial communities regulate the inflammatory responses that mark diseases like rheumatoid arthritis (RA) or inflammatory bowel disease, but the exact processes of that regulation remain unknown.

“We are starting to figure out that in certain disease states, like RA or psoriatic arthritis, particular kinds of patterns of microbiota are more represented or less represented,” Dan Littman, MD, Helen L. and Martin S. Kimmel Professor of Molecular Immunology and Howard Hughes Medical Institute Investigator, Skirball Institute at New York University School of Medicine told Healio Rheumatology. “But, the big question is whether this over- or under-representation is a cause or an effect. This is not an easy thing to demonstrate. In animal models we are seeing certain triggers, and we can assume that it will be the same in humans, but to prove that is not easy.”

Veena Taneja, PhD, of the Department of Immunology at the Mayo Clinic, has conducted research on the role of the microbiome in RA. “Infections have been suggested as the cause or initiation of autoimmunity, including RA,” she said. “However, no single pathogen has been implicated as a causative factor. With the advent of new technology, it has become possible to determine the trillions of bacteria [that] live in and on us. The possibility that [a] pathogen can be an endogenous commensal is unsuspecting.”

Disturbance in the microbiome as a result of infection, diet, genetics or other environmental factors may lead to a niche for opportunistic microbes to expand, causing inflammation, according to Taneja. “These microbes may produce metabolites that could be involved in inflammatory pathways,” she said.

However, information about the etiology and pathogenesis of rheumatic disease as a function of the microbiome is still speculation. So too, then, is the knowledge base about therapeutic approaches to deal with the various disturbances and inflammatory processes. With this in mind, experts are left to draw conclusions from specific, piecemeal research about specific diseases and disease states to build understanding from there.

Landmark Research in RA

One study that gathered attention was conducted by Scher and colleagues, who investigated a cohort of disease-modified antirheumatic drug [DMARD]- and steroid-naïve patients with new-onset RA. The investigators performed 16S sequencing on 114 stool samples for a cohort of individuals with RA and healthy controls. They also performed shotgun sequencing on a subset of 44 samples, also from RA patients and controls. Results indicated the presence of Prevotella copri strongly correlated with disease in patients who were treatment-naïve and had new-onset RA. As the abundance of Prevotella increased, a reduction in bacteroides was observed, which the researchers suggested amounted to a loss of beneficial microbes in the new-onset patients. Other findings indicated unique Prevotella genes correlated with RA.

“Further, colonization of mice revealed the ability of P. copri to dominate the intestinal microbiota and resulted in an increased sensitivity to chemically induced colitis,” the researchers wrote. “This work identifies a potential role for P. copri in the pathogenesis of RA.”

Scher and colleagues acknowledged the findings are “intriguing” because other data sets have pointed to low Prevotella prevalence in most healthy individuals.

“The reasons for this remain unclear, but it is possible that it relates to dietary factors (as fiber-consuming subjects — and children in particular — have high intestinal levels of Prevotella), geographic variability, or a combination of both,” they wrote. “The relationship between fiber intake and RA incidence, however, is yet to be studied.” Despite a growing body of evidence, Scher and colleagues acknowledged many of the studies show correlation rather than causation.

“This study showed for the first time that in RA patients who were not treated had an abundance of a specific bacteria for patients who were negative for RA-susceptible [human leukocyte antigen] HLA gene,” Taneja said. “This suggests that a subset of patients might have these bacteria as 60% to 70% of patients carry the RA-susceptible HLA genes.”

Recognizing this trigger of RA may be an important piece of information, according to Gregg J. Silverman, MD, professor in the Department of Medicine and Pathology and co-director of the Musculoskeletal Center of Excellence at NYU Langone Medical Center. However, he suggested the study raises as many questions as it answers.

“We can see this microbe was meant to be an immune stimulator,” Silverman said. “It can serve as a protector. However, if you have too many protectors in the gut, too much of these microbes that recognize RA, it can be too much and lead to disease.”

Jordan E. Bisanz, PhD, a post-doctoral scholar at the University of California, San Francisco, encouraged the clinical community to think bigger.

“In the balance between host genetics and environmental factors in disease etiology, it is likely the microbiome is an underappreciated environmental factor which could act as a trigger for disease through its continuous interactions with the immune system,” he said. “Similarly, it may aggravate or suppress symptoms in a community-dependent manner. In future work, it will be important to carry out both studies of individuals with active disease, as well as prospective studies of at-risk individuals as the communities which may act as a trigger for disease may be completely different than the communities which are present once disease occurs.”

Genome and Microbiome

Zhang and colleagues conducted a metagenomics shotgun sequencing and metagenome-wide association study that included a number of microbial samples, including fecal, dental and salivary. The analysis included patients with RA and healthy controls. The investigators reported concordance between the gut and oral microbiomes, which they said may indicate overlap in the abundance and function of microbial species at various sites in the body, according to the results. They found dysbiosis in the gut and oral microbial communities of participants with RA. However, this dysbiosis was resolved following RA therapy.

Other findings indicated participants in the RA group had alterations in the gut, dental or saliva microbiome, while healthy controls did not. These alterations correlated with clinical parameters of the disease and may be useful in stratifying patients with regard to treatment response. A notable depletion in the RA group was Haemophilus spp., which was found in all three body sites and demonstrated a negative correlation with levels of serum antibodies, according to the results. Lactobacillus salivarius was over-represented at all body sites in the RA patient group and was elevated among patients with highly active disease.

“Functionally, the redox environment, transport and metabolism of iron, sulfur, zinc and arginine were altered in the microbiota of individuals with RA,” the researchers wrote, and added that they detected molecular mimicry of human antigens associated with the disease. “Our results establish specific alterations in the gut and oral microbiomes in individuals with RA and suggest potential ways of using microbiome composition for prognosis and diagnosis.”

Bisanz put these findings into context. “Zhang and colleagues took the next step in the microbiome field by moving from simply answering the question of what bacteria are present and how are they different between RA and health, to answering the question of how is the microbiome in functionally different,” he said. “The quantity of information and the depth of DNA sequencing is a daunting feat that will serve to be a useful starting point for many investigations to come.”

The study is also remarkable because specific organisms have been implicated in the pathogenesis of the disease, according to Bisanz. He added the study also demonstrates microbial function in the disease.

“It should however be noted that this is purely an observational study and will need to be replicated and independently confirmed,” he said. “As the authors themselves acknowledge, it still remains to be determined if the microbiota is the chicken or the egg. That is, does the microbiota shape the disease or does the disease shape the microbiota? In reality, the truth likely lies somewhere in between.”

There is no question that all of these diseases have a genetic component to them, according to Littman. “Polymorphisms increase the likelihood of a person to develop a certain type of disease,” he said. “But it is also an interplay between the microbiota and the environment with the genetic factors.”

Silverman suggested that answers may lie in understanding how the human genome and the human microbiome evolved together.

“We have started to recognize certain genes and gene variants predispose you to RA while others may predispose you to lupus,” he said. “We are also seeing how certain microbes appear to express proteins that are recognized by some individuals but not others. Patients with autoimmune disease may have important differences in how they live with and even have immune responses to the commensal bacteria in their intestines.”

He proposed microbes may activate the immune system of certain autoimmune prone individuals because genetic factors that also predisposes the individual to one of these diseases.

“It is sort of like mixing and matching items on a menu, where you can choose an appetizer from column A and an entrée from column B,” he said. “What we are learning is that certain microbes act like category B, activating the immune systems of the individuals who have specific genetic factors from column A in the mixing and matching. If you get a particular combination of those factors turned on, you get lupus or another match you get psoriatic arthritis.”

Our current understanding of the levels of communication or inter-connectedness between microbes and the host’s immune system or among the microbes themselves, remains limited, according to Silverman. “There are so many dynamic interactions to account for,” he said. “But we are certainly making progress in our investigations.”

Taneja took the discussion a step further. “Sequences of certain bugs may have similarities with self-protein,” she said. “When immune cells present the bacterial sequences bound to various ligands, it can generate response to self-protein. It could be a combination of factors, including genetic factors, that lead to onset of disease. An individual with predisposing genetic factors may have a tendency to have abundance of certain types of opportunistic microbes or lower numbers of beneficial bacteria tipping the balance to inflammation.”

For Littman, other complicating factors may also be in play. “There are dietary effects, stresses, environmental factors,” he said. “There is a whole spectrum. Of course, there are some cases where the genetic predisposition is so overwhelming that 100% of people with that genetic burden are going to get sick no matter what. We see that in juvenile Crohn’s disease. But most genetic predispositions are compounded by dozens or even hundreds of other factors, and our understanding of the role of microbiota in this is still limited.”

Proinflammatory Response

Vitetta and colleagues wrote that painkillers may disrupt the integrity of the gastrointestinal barrier and lead to inflammation and hyper-permeability. Probiotics and prebiotics may aid in restoration of this barrier in addition to down-regulating proinflammatory mediators. One way that may occur is the modulation of Clostridia species in the gut. However, they took issue with certain assertions by the clinical community.

“A postulated requirement has been the abrogation of free radical formation by numerous natural antioxidant molecules in order to improve musculoskeletal health outcomes, this notion in our view, is in error,” they wrote.

When reactive oxygen species are produced in environments such as the gastrointestinal tract, and the commensal microbe cohort is sufficiently regulated in its processes, hydrogen peroxide is formed, according to the authors. They suggested hydrogen peroxide is accepted as a messenger for cellular homeostasis and physiological function.

“The [gastrointestinal] commensal profile that tolerates the host does so by regulating proinflammatory and anti-inflammatory [gastrointestinal] mucosal actions through the activity of [reactive oxygen species] signaling thereby controlling the activity of pathogenic bacterial species,” they concluded.

Silverman suggested inflammatory conditions occur when a combination of activation of innate immune processes and adaptive process occur, which may have dominant roles for T-cell activation. “Different components of microbiota can influence innate adaptive processes,” he said. “Microbes can induce different parts of an inflammatory response.”

Under normal conditions, there is a balance between proinflammatory and anti-inflammatory processes, according to Silverman. But when this balance is disturbed, diseases can occur.

“How can particular disturbances in microbiota in the gut can lead to systemic or organ-specific rheumatic disease?” he said. “That is something we do not understand at all. We have seen it only in animal models. There is no direct evidence of causative or initiating factors yet in humans.”

Immune Education

“We need microbes to instruct our immune system at all times, even when we are healthy, but particularly in situations where there is a real threat,” Silverman said. He noted there are 35,000 different kinds of commensal bacteria in humans.

“Understanding this educational process can be daunting. These communities can also shift, with different stressors, intercurrent illnesses or even changes in diet. Abundance of different bacteria types can change. These can proliferate or these can contract in abundance or, maybe most importantly, adapt quicker than us. These shifts in commensal bacteria may provide us with resistance of disease or make better use of a change in our diet.”

Vanaten and colleagues cited the hygiene hypothesis, which suggested that living in a cleaner environment may upset the immune balance, which may unintentionally increase our risk of developing diseases, such as allergies, autoimmune diseases or even cancer.

How does this actually happen, it could be that early antibiotic exposure changes early microbial exposure and colonization that may thereby alter immune maturation.

The researchers followed the development of the microbiome in the gut of 222 infants from Northern Europe. Results indicated Bacteroides species have a low abundance in Russian infants, but dominate in those born in Finland and Estonia. The researchers suggested lipopolysaccharide exposure arose primarily from Bacteroides as opposed to Escherichia coli, which they described as “a potent innate immune activator.” There are structural differences between Bacteroides lipopolysaccharide and E. coli lipopolysaccharide that may inhibit innate signaling in the immune system and tolerance to endotoxins. Lipopolysaccharides from Bacteroides does not reduce the incidence of autoimmune diabetes in non-obese diabetic mice, while lipopolysaccharides from E. coli have that effect.

“Early colonization by immunologically silencing microbiota may thus preclude aspects of immune education,” the researchers concluded.

Silverman suggested geographical location and ethnicity also should be considered. “Lupus in Thailand is not necessarily the same as in Norway or Ireland or the Middle East,” he said. “The background genetics are different, as are the diet and exposures on a day-to-day basis. We need to comb through this information and figure out the exact bacteria that are found in different places before we can prove anything.”

He added, “How exactly are these microbes being recognized by the immune system?” he said. “It is unclear. Sometimes the bacteria try to calm or shut down the immune system. If these bacteria become highly abundant, the immune system may shut down entirely. But in reality, there is still so much we do not know about what is going on with our immune system.”

Therapeutic Approaches

Researchers should always keep an eye on practical implications, according to Bisanz.

“Understanding the role of the microbiota in rheumatic diseases will potentially allow us to have a greater understanding of the underlying causes of the disease, identify new treatments, predict when the disease will occur and, most importantly, identify preventative measures to stop the disease from occurring in the first place,” he said. “For example, if we know that a certain bacterium serves as a trigger for RA, individuals identified as being at risk could undergo targeted therapies to remove, replace or modify the problem organism. This is the one of the main goals of developing technologies in the microbiome and probiotics fields.”

In a review by Yeoh and colleagues, the role of microbiota in rheumatic diseases, including RA, spondyloarthritis and inflammatory bowel disease (IBD) was detailed.

“Mechanisms through which the microbiota may be involved in the pathogenesis of these diseases include altered epithelial and mucosal permeability, loss of immune tolerance to components of the indigenous microbiota, and trafficking of both activated immune cells and antigenic material to the joints,” they wrote. They suggested probiotics and fecal microbial transplant are key strategies for altering or manipulating the microbiome.

“Both approaches are in their infancy with regard to management of rheumatic disease but their potential is worthy of consideration, given the need for novel therapeutic approaches, and the emerging recognition of the importance of microbial interactions with human hosts,” Yeoh and colleagues wrote.

“This is a review which discusses the role of oral and gut microbiome in various diseases,” Taneja said. “There are many studies that support the role of oral and gut microbiome in various diseases. Success of fecal microbial transplant in IBD further underscores the role of the gut microbiome in various diseases.”

Despite the attention it has received in recent years, experts remain cautious about the extent to which fecal microbial transplant could impact the field. “It certainly has the potential to change the paradigm, but we have much more to learn, and the rules that apply in each disease, before we can use it in a targeted way,” Silverman said.

Specific Diseases

Bisanz and colleagues suggested loss of mucosal tolerance in the microbiome may be a contributing factor to spondyloarthritis. Bisanz conducted research at the Canadian Centre for Human Microbiome and Probiotic Research at the University of Western Ontario. The researchers investigated microbiota in the oral cavity of 17 patients with axial spondyloarthritis AxSpA and compared the results to those found in 22 healthy controls. Clinicians conducted full dental examinations of all participants, including an assessment for periodontal disease. They also obtained plaque samples for analysis of bacterial communities.

Patients in the disease group demonstrated active periodontitis. Results also indicated individuals in the AxSpA group had a significantly greater prevalence of periodontitis than controls. However, no differences were reported between the study group and controls with regard to multiple metrics of alpha and beta bacterial community diversity. In addition, they analyzed the individual taxa and higher level taxonomic groupings which did not reveal evidence of correlated organisms in the subgingival plaque that were associated with AxSpA.

“Although 16S rRNA gene sequencing did not identify specific bacterial profiles associated with [axial spondyloarthritis], there remains the potential for the microbiota to exert functional and metabolic influences in the oral cavity which could be involved in the pathogenesis of AxSpA,” the researchers concluded.

“Ankylosing spondylitis (AS) has been associated with the presence of certain bacteria prior to the onset of disease,” Taneja said. “In this study, authors compared oral microbiome of AS patients with healthy individuals. They did not describe any difference in the oral microbiome for specific bacteria, though an increase in periodontitis was observed in patients. Since AS is also associated with IBD, it could be that gut microbiome is involved. Alternatively, treatment of patients may have normalized the oral microbiome. However, the number of patients are too low and a bigger cohort is required to confirm these findings.”

Bisanz agreed. “We performed a pilot investigation into the dental plaque microbiota in AxSpA using a DNA-sequencing approach that focuses on the 16S rRNA gene,” he said. “This allowed us to determine which microbes were present in health and disease. We were unable to find a clear microbial signature of disease — a smoking gun if you will — in our participants suggesting that differences in the microbiota, should they exist, must be due to what the bacteria are doing, rather than what bacteria were there. This requires alternative, and subsequently far more costly and intensive, approaches such as the shotgun sequencing approaches applied by Zhang and colleagues.”

Taneja and colleagues have investigated RA. “We have been able to pinpoint certain bacteria that are expanded in RA patients, although they are generally present with extremely low abundance in healthy individuals,” she said. “Further, we could show that beneficial bacteria belonging to Faecalibacteria were reduced. These bacteria are known to produce butyrate, which is required for gut epithelial layer repair and growth.”

She added, “Our study suggested that a combination of disturbance in certain bacteria may underlie the onset of disease. Using one of the bacteria identified in RA patients, we could show that it increases disease severity in humanized arthritic mice which carry the RA-susceptible gene. Our study points to the use of the gut microbiome in defining them as biomarkers to predict individuals at risk, as well as to target the pathways those bacteria are involved in.”

Scher and colleagues attempted to make some sense of the diverse body of information about what bacteria are involved in which diseases. They recently published a comprehensive history and review.

“Evidence that RA follows the epidemiology of an infectious vector is both intriguing and debated,” they wrote. They suggested some evidence points to periodontitis or disruptions in the oral microbiota as contributing factors. “Another possibility is that the distal airways represent the actual site of citrullination, perhaps as a consequence of environmental insults (ie, smoking) and/or microbial challenges.”

Despite a number of animal studies that show associations and triggers between disruptions in the microbiota and RA, data in humans are scarce, according to Scher and colleagues.

Regarding other diseases investigated by Scher, it was reported that Stebbings and colleagues found that compared to healthy controls, patients with AS had greater abundance of Lachnospiraceae and Prevotellaceae in the terminal ileum, along with a decrease in Ruminococcaceae and Rikenellaceae families. Children with enthesitis-related arthritis carried more Faecalibacterim prausnitzii than healthy children, according to findings from another study cited by Scher and colleagues.

Scher’s group identified decreased levels of Coprococcus in both patients with psoriasis and psoriatic arthritis, while patients with psoriatic arthritis also had reductions in Akkermansia and Ruminococcus. He added that patients with IBD also may have lower levels of bacteria from the Ruminococcaceae family and Akkermansia genus.

Future Research

Littman said presently there is a greater understanding of the microbiome in mouse models than in humans. “By being able to work with animals, we can get better understanding of how individual components of microbiota can influence different branches of host immunity or metabolism,” he said.

However, even the most encouraging data sets, such as the one investigating Prevotella by Littman, Scher and colleagues, have left the clinical community with more questions than answers, according to Littman.

“Few studies have been confirmed in different patient cohorts,” he said. “The Prevotella study has yet to be reproduced. We would like to see findings like these reproduced in other patient cohorts. We need to see the predictive value to doing microbiome analysis and, most importantly, show causality.”

For Taneja, a focus should be kept on finding therapies.

“We are trying to determine if microbiome can be used to determine the drug response in a larger cohort of patients,” she said. “The gut microbiome has a great potential in being used for treatment, as we could show with Prevotella histicola where we showed that arthritic mice treated with the bug did not progress to severe disease and mice treated prophylactically were protected from arthritis.”

There may also be benefit to looking outside of the clinic for answers, according to Silverman.

“We could think about the hygiene hypothesis,” he said. “We evolved to live in a dirty world, with all manner of bacteria living inside us. Now, we have all kinds of Lysol, antiseptics and highly hygienic lifestyles. All of this leads to loss of bacterial diversity. It may be easier for things to get out of balance.”

Silverman added that about half of babies are born by Cesarean section. “The birth canal exposes babies to many important bacteria,” he said. “But with Cesarean birth, it may be that instead of being exposed to 1,500 bacteria, you are only exposed to 1,000. You are in trouble from birth. Every day you go into the street, you are exposed to a range of environmental bacteria, but you may develop the wrong kinds of immune responses if you do not have the right balance among your commensals.”

Silverman noted the research community has a long way to go.

“Koch’s postulates tell us that a particular bacterial species may transmit a disease so that bacteria represents a threat,” he said. “That is a no-brainer. But it is based on the hypothesis centered on the infectious causes of disease. When we consider the commensal communities in our bodies it may be the opposite of how we think now. The microbiome in our bodies is a community of trillions of bacterial cells. Not one individual commensal species is the bad guy, or there may not even be a bad guy at all, just an absence of good guys. Our old way of thinking does not appear to be exactly relevant.”

• References:
• Bisanz JE, et al. Peer J. 2016;doi:10.7717/peerj.2095.
• Chen J, et al. Genome Medicine. 2016;doi: 10.1186/s13073-016-0299-7.
• Marietta E, et al. Arthritis Rheumatol. 2016;doi: 10.1002/art.39785.
• Scher JU, et al. Arthritis & Rheumatology. 2016;doi:10.1002/art.39259.
• Scher JU, et al. eLife. 2013;doi:10.7554/eLife.01202.001.
• Stebbings S, et al. Rheumatology (Oxford). 2002;41:1395-1401.
• Taneja V. FEBS Lett. 2015;doi:10.1016/j.febslet.2014.05.034.
• Vanaten T, et al. Cell. 2016; doi:10.1016/j.cell.2016.04.007.
• Vitetta L, et al. Pathogens. 2013.doi:10.3390/pathogens2040606.
• Yeoh N, et al. Curr Rheumatol Rep. 2013;doi:10.1007/s11926-012-0314-y.
• Zhang X, et al. Nat Med. 2015;doi:10.1038/nm.3914.

• For more information:
• Jordan E. Bisanz, PhD, can be reached at the University of California, San Francisco, HSW1501, 513 Parnassus Ave., San Francisco, CA; email: jordan.bisanz@ucsf.edu.
• Dan R. Littman, MD, can be reached at 540 First Ave., 2nd Floor, Lab 17, New York, NY 10016; email: dan.littman@med.nyu.edu.
• Gregg J. Silverman, MD, can be reached at 450 East 29th St., New York, NY 10016; email: gregg.silverman@nyumc.org.
• Veena Taneja, PhD, can be reached at 200 First St. SW, Rochester, MN 55905; email: taneja.veena@mayo.edu.

Disclosures: Bisanz and Taneja report no relevant financial disclosures. Littman reports he is the founder of the Vedanta Biosciences company. Silverman reports he is a consultant for Celgene, Genentech, Eli Lilly, Pfizer and Roche.