Faculty Bibliography

Background/Purpose: Apolipoprotein L1 (APOL1) risk variants (RV), G1 and G2, associate with CKD in African Americans (AA) and are evolutionarily preserved due to improved infectious resistance. Interferons (IFN) in SLE, have been shown to increase APOL1 expression and RV toxicity in endothelial cells and podocytes. Though RV homozygotes with SLE nephritis demonstrate advanced renal progression, associations with renal histopathologies have not been validated in SLE.
Method(s): Herein, this cohort study tested the hypothesis that RV homozygosity (RV/RV) associates with specific clinical and biopsy features compared to reference allele (G0) homozygosity (G0/G0) or RV heterozygosity (RV/G0). Whole blood DNA for genotyping, kidney biopsy slides, and clinical reports from 77 AA SLE patients with biopsy-proven nephritis reviewed for: biopsy class, activity index (AI), chronicity index (CI) and clinical features across APOL1 genotype. RVattributed mitochondrial morphology, was assessed on electron microscopy (EM) images. Analysis was confirmed by two blinded pathologists. As proof of concept, primary endothelial cells across genotype were given IFN to over-express APOL1, and features on EM were compared.
Result(s): The G0/G0, RV/G0, and RV/RV groups comprised 35%, 52%, and 12% of the cohort. There were no genotype differences in SLE history or demographics. Compared to G0/G0, and RV/G0 groups, the RV/RV had higher urine protein to creatinine ratios (uPCR) and creatinine (Cr) at biopsy (mean uPCR: 2.5; 2.7; 4.3 mg/L p=0.06 and Cr: 1.3; 1.03; 2.3 mg/dL p=0.01 respectively). Adjusting for dsDNA, AI, CI, and percent sclerotic glomeruli, the RVs independently associated with proteinuria at biopsy (OR=2.1, p=0.05). Paradoxically, the G0/G0 and RV/G0 vs the RV/RV group displayed higher AI and CI with a trend toward higher dsDNAs at biopsy (G0/G0 or RV/G0: AI: 5.3/24; CI: 2.6/12 dsDNA: 447 vs RV/RV: AI: 1.2/24; CI: 1.3/12; dsDNA: 69, p=AI: 0.004; CI: 0.01; dsDNA: 0.1). In 30% of the RV/RV cases, the reading pathologist commented that clinical severity was out of proportion to the biopsy lesion. The RVassociated with ESRD in 7.9%, 3.9%, and 20% of the G0/ G0, RV/G0, and RV/RV cases (OR: 5.7; p=0.03). On EM, RVs associated with mitochondrial condensation (mitochondrial area: G0/G0: 0.17; RV/G0: 0.09; RV/RV: 0.06 mum²; p<0.01); this result was recapitulated in our cell culture model. IFNtreated endothelial cells increased APOL1 expression 18 fold across genotypes (p<0.01). Compared to G0/G0 and RV/G0 cells, RV/RV cells had mitochondrial areas: G0/G0: 0.08; RV/G0: 0.07; RV/RV: 0.04 mum²; (p<0.01).
Conclusion(s): In this SLE cohort, APOL1 RVs associated with poorer prognosticators, initial proteinuria and creatinine, and ultimately progressive nephritis. These features were paradoxically out of proportion to the SLE lesion on biopsy. The literature supports RV-conferred podocyte and endothelial cell mitochondrial defects owing to a mitophagy deficiency. As evidenced by EM images from both SLE patient biopsies and primary cell cultures, these genes may conferrer intrinsic renal pathology. Consequently, traditional scoring of histopathologic severity may underestimate injury that associates with RValleles

Background/Purpose: Systemiclupus erythematosus (SLE) disproportionally affects women of childbearing age. Low disease activity for 6 months prior to conception leads to the best outcomes; however, there is little prospective data describing the relative frequency and predictors of flares during and after pregnancy under such conditions.
Method(s): Analyses used data from the PROMISSE study, a multicenter, prospective observational study (2003-2014) of 384 pregnant women meeting >=4 ACR SLE criteria. Subjects were enrolled <12 wks gestation and samples collected throughout pregnancy and post-partum. Exclusion criteria were multi-fetal pregnancy, active disease (prednisone >20 mg/ d), or renal disease (proteinuria >1 gm/24h, and creatinine > 1.2 mg/dL). Mild/moderate and severe flares were defined using the SELENA-SLEDAI Flare Index. Flares during pregnancy were assessed in all 384 patients, and post-partum flares in those with study visits 2-6 months post-partum. Logistic regression models were fit to the data to identify independent predictors of experiencing any type of flare during pregnancy and post-partum.
Result(s): Rates of Flare: 105 flares were recorded during pregnancy (3.8% 1st Trimester, 53.3% 2nd, 42.9% 3rd). Counting one flare per person, 100 of 384 patients (26%) flared at any point during pregnancy; 20.8% of patients had mild/ moderate flares and 6.25% had severe. 57 of 234 patients (24.4%) with a study visit 2-6 months post-partum flared; 22.7% had mild/moderate flares and 1.7% severe. Post-partum flares were mild, and 19 of 57 (33.3%) were treated; 13 with an increase in prednisone, 6 with NSAID or hydroxychloroquine, and 1 with mycophenolate mofetil. The rate of any type of flare was 0.39/person-year at any point during pregnancy and 0.84/person-year post-partum. The proportion of subjects who had any flares during and after pregnancy was similar, but the post-partum flares occurred over a shorter duration of follow-up. Correlates of Flare: Among baseline variables considered (Table) only age, ethnicity/race, low complement, and PGA were independently predictive of having any flare during pregnancy. Clinical features associated with adverse pregnancy outcome (platelet count, antihypertensive use, and LAC) were not predictive of flare. The mean time from the last visit during pregnancy to the post-partum visit was 20 weeks and ranged from 9 to 34 weeks. Neither baseline clinical variables nor clinical variables of the last visit during pregnancy were associated with occurrence of any post-partum flare.
Conclusion(s): Flares during pregnancy are correlated with clinical and serological activity during the first trimester. Flares during and after pregnancy are typically mild, infrequently require treatment, and occur at similar rates. (Table Presented)

Background/Purpose: Towards understanding the molecular mechanisms that link maternal anti-Ro antibodies to the development of conduction system disease in a second trimester fetus, single cell (scRNA-seq) and bulk RNA-seq were applied to a fetal heart dying with complete congenital heart block (CHB) and a gestational age-matched healthy heart from an elective termination.
Method(s): The CHB heart was obtained from a 20-week fetus identified to have complete block at 19 weeks. The mother (35 y/o Asian with SS on no hydroxychloroquine) declined dexamethasone or IVIG and elected to terminate, thus no exposure to maternal medications confounded interpretation of findings. Both hearts were obtained under identical conditions. Freshly collected single-cell suspensions were generated using a Langendorff preparation with cannulation and perfusion of the aorta with collagenase and trypsin enzymes. Two approaches were taken to mine the transcriptome in the resulting cell suspensions: agnostic evaluation applying 10X Genomics platform-based scRNA-seq and low input RNA-seq of flow sorted cells upon leukocytes (DAPI negative, CD45+) and fibroblasts (DAPI negative, CD45-, podoplanin-positive).
Result(s): For scRNA-seq, we obtained 2,693 and 5,408 high-quality scRNA-seq profiles from the control and CHB hearts, respectively. We applied a graph-based clustering method and identified 13 and 14 major clusters of cells from the control and CHB hearts, respectively, as visualized by t-distributed stochastic neighbor embedding (t-SNE). Differential gene expression analysis guided by established lineage markers revealed four cardiomyocyte clusters (CM1-CM4), three fibroblast clusters (FB1-FB3), endothelial cells (EC), erythroblasts (EB), macrophages (MAC), dendritic cells (DC), Tcells (TC) and B cells (BC). Ranked by abundance, the control heart exhibited CM>FB>EC>MAC>DC>EB, BC, TC; the CHB heart exhibited CM>FB>EC, MAC>TC, BC, EB. The CHB heart also contained natural killer cells (NK) and mast cells (MC, lowest abundance). Given the high abundance of MACs among the immune cells (control:108;CHB:606) and the consistent identification of MACs on histologic analysis of CHB hearts, differential expression analysis demonstrated overexpression of interferon-induced genes (4-fold or greater, i.e. log2(CHB-control)>2) in CHB MACs. In CHB, most cell types expressed high levels of ISG1, IFITM1 and IFITM3, whereas in the control only IFITM3 showed widespread expression. For SIGLEC1, expression was restricted to MACs and was expressed by 18% of CHB MACs and only 6% of control MACs. While the transcriptome using low input RNA-seq of anti-CD45 flow-sorted CHB leukocytes did not allow granular analysis of leukocyte subpopulations, expression of SIGLEC1 and interferon-related genes were increased in CHB versus control. Applying 10X Genomics, proliferating fibroblasts expressed MKI67 and TOP2A in CHB but not control fibroblasts.
Conclusion(s): This unprecedented opportunity to obtain CHB tissue absent any exposure to maternal medications support scRNA-seq's utility to survey landscape and heterogeneity not possible with low input RNA-seq of flow-sorted cells. IFNand SIGLEC1-positive macrophages may contribute to fibrosis

Background/Purpose: Two Apolipoprotein L1 (APOL1) risk variants (RV), G1 and G2, are enriched in ancestrally African populations due to a conferred superior resistance to Trypanosoma brucei. This improved infectious evolutionary fitness comes at the cost of propensity toward progressive renal disease by multiple causes including SLE. In Ghana's Ashanti people, Allelic frequencies for G0, G1, and G2 have been reported at 0.46, 0.49, and 0.13 respectively. Despite their high frequencies, and the established role in SLE nephritis, no study has examined outcomes in a Ghanaian SLE cohort. Accordingly, this prospective study evaluated APOL1 risk traits including renal, damage accrual, and mortality in 101 Ghanaian patients followed at Korle bu Teaching Hospital in Accra, Ghana.
Method(s): From 05/2015-04/2018, 101 Ghanaian patients meeting at least 4 ACR criteria for SLE were followed prospectively with data evaluated every six months. DNA was extracted from saliva and patients were stratified by APOL1 genotype as follows: reference allele (G0/G0), RV heterozygote (RV/G0), and RV homozygotes (RV/RV). Sera were shipped to the NYU clinical lab for confirmation of anti-dsDNA. Clinical endpoints included demographics, ACR criteria, SLEDAI score, SLICC damage index, mortality, vital signs, and laboratory values as available.
Result(s): The frequencies of the G1, and G2 alleles were 0.24, and 0.12 respectively- lower than would be expected given the reported regional frequencies. Subjects were 100% female, with an average age of 32.1 years and disease duration of 2.9 years. There were no differences in demographics across the genotypes. The RV associated with higher BP: 108/71, 108/ 73, and 120/82 in the G0/G0, RV/G0, and RV/RV groups respectively (p=0.04 systolic; 0.009 diastolic). Among those with nephritis, RV/RV carriers presented with higher dipstick proteinuria than G0/G0 and RV/G0 carriers (2.6 vs 1.2 and 1.3 respectively; p=0.05; F=3.1). While proteinuria responded most robustly to therapy in RV/RVs, it remained higher at each time point compared to the other genotypes (p=0.002; F=5.4). SLEDAI scores were comparable across the genotypes, however RV/RV carriers had lower dsDNA titers (10.7 IU/mL) than G0/G0 (57.1 IU/mL) and RV/G0 (95.6 IU/mL) carriers (p: 0.03). RV homozygosity associated with elevated SLICC damage index: G0/G0 or RV/G0: 0.95 vs RV/RV: 1.7; driven by renal, CVD, and neurologic manifestations G0/G0: 0.46, RV/G0: 0.39, RV/RV: 1.25; (p=0.03). There were 5 deaths during the study period: 1 death in the G0/G0 group (ESRD), 1 in the RV/G0 group (post-partum sepsis plus renal), and 3 in the RV/RV group (myocarditis/heart failure; ESRD; unknown). RV carrier status associated with mortality with 25% of the RV/RV group succumbing to disease vs 2% in the G0/G0 and RV/G0 groups (p=0.003; F=6.3).
Conclusion(s): Taken together, APOL1 RV associates with renal progression, organ damage accrual, and mortality in this Ghanaian SLE cohort. Despite having poorer outcomes, RV homozygotes exhibited lower dsDNA titers and similar SLEDAI scores compared to G0/G0 or RV/G0 patients suggesting a genetic effect independent of SLE activity. APOL1 genotyping could have important prognostic implications in ancestrally African SLE patients

Background A transmissible agent has long been suspected inthe pathogenesis of SLE, yet the potential contribution of thehuman intestinal microbiome has been little examined. Wetherefore characterized the gut microbiota of patients withSLE, with special interest in those with lupus nephritis (LN).Methods Blood and fecal samples from SLE patients wereobtained, with strict inclusion/exclusion of criteria. Fecal 16SrDNA sequencing, as well as cytokine and autoantibody assayswere performed. In addition, sera from two independent lupuscohorts were studied for validation. Biomarkers of gut leakiness were assessed.Results Compared to controls, the intestinal microbiome fromSLE patients (n=61) showed decreased species richness diversity with reductions in taxonomic complexity mostpronounced in those with high disease activity. Notably, SLEpatients had an overall 5-fold greater representation of a species in the Lachnospiracaea family of obligate anaerobic Grampositive cocci, with reciprocal contractions of two other commensal species with putative protective properties. Abundanceof the Lachnospiracaea species correlated with serum IgG to acell wall component, postulated to represent a lipoglycan,from a strain of this same species (p=0.002, n=61, Spearman)but not with 7 other strains. There was also a significantdirect correlation between SLEDAI scores and levels of thesecirculating anti-strain IgG antibodies (p=0.02, n=48). Levelsof antibodies to strain-specific bacterial antigen, treated withRNAse/DNAse/proteinase K, were significantly higher in thosewith active nephritis at time of sampling compared to SLEwithout renal activity (Cohort 1 p=0.01 n=48; Cohort 2p=0.006, n=28, Mann-Whitney). Levels of serum IgG antistrain antibodies also significantly correlated with high-titerserum IgG to native DNA (p<0.0001, n=27), and inverselycorrelated with C3 and C4 levels. High titers of these antibacterial antibodies were associated with active Class III, IVand V (overlap) LN (Cohort 3).Conclusions These findings suggest a novel paradigm for thepathogenesis of LN in which a common intestinal commensalbacteria may contribute to the immune-complex mediated disease process, with features akin to poststreptococcal GN butwithout outward signs and symptoms of clinical infection