THU0239 DYNAMIC TEMPORAL CHANGES IN CLINICAL DISEASE ACTIVITY AND GUT MICROBIOTA REPRESENTATION OF A PATHOBIONT LINKED TO LUPUS NEPHRITIS

Abstract

Listen

Translate article

Background:From a cross-sectional cohort, we have identified a candidate human gut anaerobic pathobiont,Ruminococcus gnavus(RG) of the familyLachnospiraceaethat was linked to active Lupus nephritis (LN)(1). Based on 16S rRNA amplicon analysis, LN patients displayed increased fecal RG abundance, concordant with serum IgG anti-RG antibody responses that appeared intertwined with anti-dsDNA responses implicated in renal pathogenesis. Indeed, monocolonization of germ-free mice is reported to result in generalized inflammation and expansions of Th17 cells. However, RG at low levels are also prevalent in healthy adults, and the temporal dynamics of RG representation within Lupus microbiota ecosystems have not been investigated. Also, genomic sequences of few RG strains have been reported, and these vary greatly in genome structure, gene representation and sequence, which may have broad implications for adaptation to a host with systemic inflammation and/or factors that contribute to immune activation in a susceptible host.Objectives:To investigate the relationships betweenin vivoRG expansions and disease activity that often wax and wane overtime, we initiated longitudinal studies in Lupus patients and controls. As representation of RG strains alone might alter pathogenic potential, we also sought to characterize RG strains from active LN patients.Methods:From our cohort, patients were characterized for demographics, clinical disease activity, and serologies including standard autoantibody and complement levels, and anti-bacterial responses of interest. High throughput 16S rRNA amplicon libraries from fecal samples were analyzed using QIIME 2 and DADA2 (1). Also, individual RG colonies were isolated and subjected to whole genome sequencing. Species and strains were then assigned in part based on multi-locus sequence typing and reference guided genomic assemblies.Results:16S rRNA analysis of 34 samples, at 2-4 timepoints from 14 SLE patients, documented highly conserved patterns of gut community representation overtime in 10/14 patients, based in part on unsupervised hierarchical cluster analysis. Notably, independent of vacillations in clinical disease activity of up to 8 SLEDAI points, conserved microbiome phylogenetic abundance/composition was documented at a family level, and the level of amplicon sequence variants that approximate identification of individual species. In pilot studies, from two active lupus nephritis patients hundreds of fecal bacterial colonies were isolated, with initial assignments by 16S rRNA sequence. From highly redundant whole genome sequence analysis, these Lupus-patient fecal colonies were found to distribute into only four distinct RG strains, which differed from reported strains.Conclusion:Our findings suggest that many Lupus patients have little or no detectable perturbations in representation of theLachnospiraceaefamily or abundance of RG species overtime. Moreover, this seeming microbiota stability was documented even in patients with dramatic changes in disease activity. However, these approaches are inadequate to detect shifts between RG strains. In pilot studies we have isolated and characterized the genomes of four unique RG strains from active LN patients, which include variations in gene content and sequence that may have implications for the host-commensal relationship and immune activation. Broadening of these studies to larger number of SLE patients and healthy subjects, with metagenomic surveys of strain representation in genomic shotgun libraries are currently in progress, in coordination with murine colonization testing for immune modulatory properties of individual strains.