Association between gut microbiota community resilience and lupus flares in patients with SLE

Systemic lupus erythematosus (SLE) is an autoimmune disease often associated with inadequate treatment responses and recurrent episodes of disease flares. Despite advancements in therapeutic interventions, around 40% of patients with SLE progress to lupus nephritis (LN). Of these, patients with Class III or IV LN have poor prognoses, with over 40% progressing to end-stage kidney disease within 15 years of diagnosis. The factors contributing to lupus flares remain poorly understood.¹

Compared with healthy gut communities that display remarkable dynamic stability, imbalances within gut commensal communities, termed dysbiosis, are common in patients with various inflammatory and autoimmune diseases. In a previous study by Azzouz et al.,² a direct correlation was reported between SLE Disease Activity Index (SLEDAI) and dysbiotic shifts associated with reduced community diversity and richness. However, the resilience of SLE microbiota communities over time has not yet been examined.¹ Recently, Azzouz et al.¹ published the first longitudinal study investigating the relationship between gut microbiota communities and disease activity in patients with SLE in Annals of the Rheumatic Diseases. Here, we are pleased to summarize the key findings.

Methods¹

This longitudinal observational analysis included 16 patients with lupus and 22 healthy individuals. Two to six fecal samples were collected from all patients over 24–291 weeks. For comparison, 72 sequential samples from 22 healthy adult female control volunteers were evaluated.

Patients with SLE were categorized as

• renal or LN, if they fulfilled American College of Rheumatology criteria for renal involvement at any time (n = 9); or
• non-renal, if they had no documented history or laboratory findings of LN (n = 7).

Renal disease was further defined as active or inactive, based on urinary protein creatinine ratio >0.5 or <0.5, respectively, as per the SLEDAI domain. Patients with an overall SLEDAI score of >8 and <8 were designated as high disease activity and those with <8 as low disease activity.

Time-dependent changes in fecal communities from both patients and healthy controls were assessed through taxonomic analyses, including multivariate analysis of β-diversity. Strains were isolated from gut blooms to analyze genomes and associated glycans.

Key findings¹

Greater community β-diversity associated with increased lupus disease activity

To estimate β-diversity between groups, principal coordinates analysis was used. Commensal communities from patients with SLE displayed considerable heterogeneity, with the majority exhibiting significant distance variance from controls (p = 0.001). In comparison with the control group, variance in β-diversity was highest in the high disease activity (p = 0.001) and active LN subgroup (p = 0.001)

Lupus gut microbiota communities unstable over time

The stability of gut microbiota over time was assessed by comparing community wide multivariate analyses in multiple libraries from each of the 16 patients and nine healthy individuals. There was minimal variation in the overall composition among the sampled communities for each of the controls. On the contrary, there was a significantly higher level of variance observed between communities at different time points within an individual patient (p = 0.03), both in renal (p = 0.02) and non-renal groups (p = 0.03). Interestingly, there were no significant differences in the variance between renal and non-renal groups (p = 0.379), suggesting that SLE is frequently characterized by impaired community composition resilience.

Lupus microbiota communities express dynamic blooms of individual bacterial species

Analyses of the sequential microbiota libraries revealed that, unlike healthy subjects, several individual patients displayed transient and often pronounced blooms of amplicon sequence variants defined taxa within both Veillonella and Fusobacterium genera. However, no blooms of either Fusobacterium or Veillonella occurred concurrently with peak flares of lupus disease activity.

Abundance of Ruminooccus (blautia) gnavus linked with lupus flare-ups

Ruminococcus (blautia) gnavus (RG) abundance in the control libraries (mean, 0.15%) was consistent with previously reported estimates from a large population-based survey. Although there was an overall trend of an increased RG abundance in SLE libraries (p = 0.076), many samples showed undetectable or low abundance of RG. However, patients with SLEDAI scores of ≥8 (p = 0.01) and with active renal disease (p = 0.01) showed a marked increase in RG abundance.

Of the 16 patients with SLE, one non-renal and four renal patients reported a nine-fold increase in RG bloom compared with other time point samples from the same patients. RG bloom samples from these five patients revealed a significantly high correlation between disease activity and RG abundance (p = 0.03). These findings suggest that RG blooms may be a factor in the clinical pattern of relapsing-remitting disease activity occurring in patients with lupus.

Genes from LN RG strains provide better adaptability to an inflamed host

Homology comparisons of 199 genes uniquely associated with RG strains isolated from patients with inflammatory bowel disease (IBD) showed that 34 were present in one or more of the four LN RG strains. Notably, all four LN RG strains included 21 of these genes. Each of these LN RG strains expresses an antioxidant peroxiredoxin gene that protects the host from oxidative stress due to increased production of reactive oxygen species and reactive nitrogen species. The increased levels of serum reactive oxygen species and reactive nitrogen species affect both host cells and gut commensals and differentiate patients with LN from those with non-renal SLE and healthy controls. In addition, RG strains from patients with LN reveal orthologues of a potential RG phosphotransferase system sugar transporter and alpha-L-fucosidase, both of which are associated with catabolism of the intestinal barrier, mucin. Overall, other intestinal bacteria may have a disadvantage compared with LN strains, with the latter having genes that permit better adaptability to an inflammatory host, particularly in patients with LN.

LN strains express the cell-membrane lipoglycan

Immunoblotting assay was carried out to detect non-protein antigen in RG strains. RG isolated from three of the four patients with LN showed serum antibody reactivity with diagnostic oligobands, while these antigenic oligobands remained undetected in strains from IBD, infants, or a healthy control. Furthermore, mass spectrometry analysis was performed to study structural features of LN-associated glycans. Glycan species purified from three independent RG strains displayed acyl-glycerol lipid anchors, called lipoglycans (LGs). These LGs in RG strains also exhibited cross-reactive antigenic determinants. Immunoblots of RG strains from two patients with LN showed presence of immunoglobulin G (IgG) reactive oligobands, while no such reactivity was observed with the RG strains from a healthy donor, patients with IBD, and antibiotic-treated infants.

Serum IgG antibodies to lupus RG strain LGs concordant with RG bloom

To evaluate host immune response to colonization of RG strains, longitudinal sera samples from three patients with lupus was assessed over time, including responses at the time of clinical LN flare. Bead assay, assessing serum IgG-reactivities for binding interactions with RG strain extract versus purified RG2 strain LG, revealed similar reactivity patterns of whole bacterial extracts and the purified LG. These findings confirm that the LG component within RG2 bacterial extract is highly immunogenic. In contrast, the lipopolysaccharide glycan from a Pseudomonas species showed no or minimal IgG reactivity. Collectively, the findings demonstrate high-titre lupus host LG-specific serum antibody reactivities, with peak levels detected at the time of lupus flare and co-occurring with an abundance of RG bloom. Notably, the antibody responses were predominantly of the IgG2 subclass.

Conclusion¹

These findings demonstrate that instability in gut microbiota communities in patients with SLE leads to the abundance of RG blooms, which are associated with disease flares. The study further highlighted the potential pathogenic properties of specific strains isolated from patients with active LN. Future development of an antibody biomarker assay for novel RG strain-associated lipoglycan could aid earlier diagnosis of LN. In addition, therapeutic targeting of pathobiont strains could offer a potential approach to treating LN flares without the need for widespread immunosuppression.