key: cord-0714740-gnudzunk authors: Menon, R.; Otto, E. A.; Sealfon, R.; Nair, V.; Wong, A. K.; Theesfeld, C. L.; Chen, X.; Wang, Y.; Boppanna, A.; Kasson, P. M.; Schaub, J. A.; Berthier, C. C.; Eddy, S.; Lienczewski, C. C.; Godfrey, B.; Dagenais, S. L.; Sohaney, R.; Hartman, J.; Fermin, D.; Subramanian, L.; Looker, H. C.; Mariani, L. H.; Naik, A. S.; Nelson, R. G.; Troyanskaya, O. G.; Kretzler, M. title: SARS-CoV-2 receptor networks in diabetic kidney disease, BK-Virus nephropathy and COVID-19 associated acute kidney injury date: 2020-05-13 journal: medRxiv : the preprint server for health sciences DOI: 10.1101/2020.05.09.20096511 sha: 7571eb2ef3f2162807ab471011ac0dc56dbeb42e doc_id: 714740 cord_uid: gnudzunk COVID-19 morbidity and mortality is significantly increased in patients with diabetes and kidney disease via unknown mechanisms. SARS-CoV-2 uses angiotensin-converting enzyme 2 (ACE2) for entry into human host cells, and ACE2 levels in target cells may influence SARS-CoV-2 susceptibility. We investigated how pre-existing conditions and drug treatments alter receptor expression in kidney tissue. Using single cell RNA profiling (scRNAseq) to assess ACE2 and associated SARS-CoV-2 proteases in healthy living donors (LD) kidneys, diabetic kidney disease (DKD), and in kidney injury during viral infection, ACE2 expression was primarily associated with proximal tubular epithelial cells (PTEC). ACE2 mRNA expression levels were significantly upregulated in DKD versus LD, however, ACE2 levels were not altered by exposures to renin angiotensin aldosterone system (RAAS) inhibitors. ACE2+ expression signatures were defined by differential expression analysis and characterized by Bayesian integrative analysis of a large compendium of public -omics datasets, resulting in the identification of network modules induced in ACE2 positive PTEC in DKD and BK virus nephropathy. These ACE2 upregulated cell programs were linked to viral entry, immune activation, endomembrane reorganization, and RNA processing and overlapped significantly with the cellular responses induced by SARS-CoV-2 infection. Similar cellular programs were activated in ACE2-positive PTEC isolated in a urine sample from a COVID19 patient with acute kidney injury, suggesting a consistent ACE2-coregulated expression program that may interact with SARS-Cov-2 infection processes. The SARS-CoV-2 receptor associated gene signatures could seed further research into therapeutic strategies for COVID-19. Functional networks of gene expression signatures are available for further exploration to researchers at HumanBase (hb.flatironinstitute.org/covid-kidney). COVID-19 disproportionally affects individuals with diabetes, hypertension, and kidney disease [1] [2] [3] . Yet the underlying molecular and physiological causes of this association are unknown, and could be as varied as drugs used to treat these conditions 4 , disease biology [5] [6] [7] , direct infection of relevant organs by the virus 8, 9 and consequent tissue destruction, and cytokine storm that occurs secondary to infection 1 . While upper and lower airway tissues are the primary sites of infection, additional sites are likely to include kidney tissue, and initial data suggest that patients with kidney disease suffer significantly higher mortality than age-matched individuals without these conditions [10] [11] [12] . Understanding the disease-specific molecular processes associated with COVID-19 in patients with kidney disease and diabetes can have a significant impact on public health. COVID-19 develops from infection with SARS-CoV-2, a betacoronavirus with a single stranded RNA genome. It gains entry into specific cell types through interaction of surface spike protein with a cell surface receptor 13 . Studies of severe acute respiratory syndrome (SARS) in the early 2000s identified angiotensin-converting enzyme 2 (ACE2) as the primary cell-entry receptor for the SARS coronavirus (SARS-CoV) in humans 14, 15 . Recent studies from the COVID-19 pandemic demonstrate that ACE2 is also the primary cell-entry receptor for SARS-CoV-2 13, 16, 17 and, similar to SARS-CoV, higher ACE2 expression may lead to higher risk of SARS-CoV-2 infection 8, 18 . ACE2 is a membrane bound zinc metallopeptidase and a master regulator of the reninangiotensin aldosterone system (RAAS). Normally, ACE2 is expressed as a cell-surface protein to which both SARS-CoV and SARS-CoV-2 can bind 4, 19 . Cleavage of spike protein is required for fusion of the virus with host membranes and occurs in a manner similar to SARS-CoV: TMPRSS2 appears to be the primary means of this activation, at least in lung epithelial cells; however cleavage can also occur in endosomes by acid-activated proteases such as cathepsin L (CTSL) 17, 20, 21 . Direct access to SARS-CoV-2 target tissues combined with the granularity of single cell RNA sequencing (scRNAseq) technology can locate specific cell types with enhanced ACE2 expression to determine whether these cells also possess additional molecular machinery that facilitate initial viral entry and subsequent cellular cytotoxicity. Characterization of these molecular processes in existing tissue from relevant cohorts could greatly accelerate the identification and development of therapeutic options for SARS-CoV-2 infection. In this study, we explored the expression and associated biological processes of ACE2 and other viral co-receptors in kidney cells from healthy living donors (LD), and from patients with diabetic kidney disease (DKD), BK virus nephropathy (BKVN) and COVID-19 associated AKI (COV-AKI). Using scRNAseq data, we defined the cellular expression of ACE2 in proximal tubular epithelial cells (PTEC), characterized the cellular programs associated with ACE2 and its regulation in DKD, BKVN and COV-AKI, mapped the ACE2 associated changes to emerging data on SARS-CoV-2 induced cellular responses and tested the associations with RAAS medication exposure (Figure 1 ). Our study focused on DKD since COVID-19 disproportionally affects individuals with diabetes and kidney disease 1, 3, 7, 11, 22, 23 , and research kidney biopsies were available in a comprehensive DKD cohort for scRNAseq profiling and analysis. The DKD analyses could identify underlying mechanisms of excessive risk associated with COVID-19 in patients with diabetes or chronic kidney disease, with the studies in BKVN and COV-AKI establishing the link between DKD and virus-associated kidney disease. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. An early DKD cohort was selected as the discovery cohort given the impact of COVID-19 on individuals with diabetes and kidney disease 1, 3, 7, 11, 22, 23 . Between 2013 and 2017, 141 American Indians (39 men, 102 women) from southern Arizona with type 2 diabetes were enrolled in an observational study of DKD that included annual research examinations. Glomerular filtration rate (GFR) was measured at each of these examinations by the urinary clearance of iothalamate 24 . At baseline, each participant also underwent a research kidney biopsy used for quantitative morphometry and for compartment-specific tissue expression studies. For kidney biopsies conducted in the last 44 participants enrolled between 2016 and 2017, processing of the kidney specimens was modified to enable scRNAseq studies using kidney biopsy tissues. These 44 participants (DKD) were included in the present study. At each annual research examination, medicines taken by the participants were recorded. Blood pressure (BP) was measured with the participant resting in a seated position, and height and weight were obtained with the participant dressed in light clothing without shoes. Body mass index was defined as weight in kilograms divided by the square of height in meters. Urine and serum creatinine were measured by an enzymatic method. Urinary albumin excretion was assessed by nephelometric immunoassay and expressed as the albumin/creatinine ratio (ACR) in mg/g creatinine. GFR and HbA1c were measured by high performance liquid chromatography. Because of the level of obesity in the participants, absolute GFR (ml/min) was reported, uncorrected for body surface area 25 . All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Transcriptomic Atlas study, which procures an additional biopsy core for research and tissues from these biopsies were used in the present study. Reference healthy tissue from 7 LD biopsies were obtained prior to perfusion and before placement in the recipient. To evaluate the kidney epithelial response to viral infection 3 biopsies from 2 patients with active BK viremia at the time of their surveillance biopsies were also included in the study. BKVN biopsies were obtained before, at the time of, or after peak BK viremia by which time immunosuppression was reduced. LD biopsies were selected as healthy reference tissue. Participant age was 49.3 ± 12, range , five of the 7 donors were female (71.4%), mean spot urine protein to creatinine ratio at the time of evaluation was 0.07 ± 0.03 g/g with range (0.04-0.12), mean iothalamate GFR was 87 ± 5 ml/min/1.73 m 2 with range (81-96). All donors were white, nondiabetic and nonhypertensive. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 13, 2020. . https://doi.org/10.1101/2020.05.09.20096511 doi: medRxiv preprint BK Virus Nephropathy (BKVN) biopsies from the transplant atlas cohort study were used, with the first BKVN patient being the recipient of a deceased donor kidney transplant, with a peak viral load >10,000 copies/ml and immunohistochemistry staining positive for SV40 (peak biopsy). After net reduction in immunosuppression the BK viral load dropped to <5000 copies/ml and a subsequent biopsy became negative for SV40 staining (post peak biopsy). The second patient was the recipient of a living donor kidney transplant and had a viral load of >10,000 copies/ml that had still not reached its peak at time of biopsy, but with positive SV40 biopsy staining (pre peak biopsy). Both patients were in the 3 rd and 4 th decade of life respectively. Clinical characteristics of the transplant atlas cohort participants are provided in Supplementary Table 1 . In response to the COVID-19 pandemic at the University of Michigan Medical Center a protocol to capture urine samples from SARS-CoV-2 patients was established to allow cell and molecular biology studies of the kidney manifestations in COVID-19. Patients are offered participation in the study upon admission to the hospital and throughout their hospital course. 50ml of a spot random urine sample is procured specifically for the study, immediately transported to the research laboratory on ice and processed for cell and molecular studies and participants are prospectively followed thereafter. COVID-19 associated AKI (COV-AKI) was studied in urine derived PTEC from a male patient in the 5 th decade of his life who had received a combined pancreas and kidney transplant two decades ago. The patient was diagnosed with COVID-19 eleven days before sample collection. Serum creatinine at time of collection was 2.3 mg/dl compared to baseline of 1.6 mg/dl. Urinalysis was significant for proteinuria, but with no RBCs on microscopy. No urinary sediment analysis was available at time of sample procurement. At the time of COVID-19 diagnosis the All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 13, 2020. . https://doi.org/10.1101/2020.05.09.20096511 doi: medRxiv preprint patient's immunosuppression was reduced to dual immunosuppression with tacrolimus and prednisone with mycophenolate mofetil withhold. His kidney function returned to baseline 17 days after admission, 6 days after sample collection (Supplementary Figure 1 ). Single cell transcriptomes were generated with 2-3 mg of the biopsy core samples from 44 CryoStor® (Stemcell Technologies) preserved DKD, 7 LD and 3 BKVN samples. Tissue processing and single cell isolation were performed following our published protocol 26 . Briefly, the tissue samples were thawed at room temperature and dissociated to single cells using Liberase TM TL (research grade, Roche) at 37 0 C. Single cell samples were immediately transferred to the University of Michigan Advanced Genomics Core facility for further processing. Urine single cell preparation followed the protocol by Arazi et al. with various modifications 27 . Briefly, COV-AKI urine was filtered through a 30 µm strainer into 50 ml tubes and centrifuged at 200 x g at 4°C for 10 minutes. After removing and storing the supernatant the cell pellet was washed once with 1 ml cold X-VIVO™10 medium (Cat#: 04-743Q, Lonza) and centrifuged in 1. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Individual cell barcoding, reverse RNA transcription, library generation and single cell sequencing using Illumina were all performed using the 10X Genomics protocol 26 . The output from the sequencer was first processed by CellRanger, the proprietary 10X Chromium single cell gene expression analysis software (https://support.10xgenomics.com/single-cell-geneexpression/software/pipelines/latest/what-is-cell-ranger). Data analyses were performed on the CellRanger output data files using the Seurat 3 R package (https://cran.rproject.org/web/packages/Seurat/index.html). As a quality control step, cells with less than 500 or greater than 5000 genes per cell were filtered out, with the lower cut off as a threshold for cell viability and the higher cut off to remove cell duplets. This study used a cutoff of < 50% mitochondrial reads per cell as an additional threshold for cell viability. A combined analysis of the single cell datasets generated from the different sample sources (LD and DKD) was Further functional analyses were focused on PTEC with ACE2 (ACE2+) and without ACE2 (ACE2-) mRNA expression in the different samples studied. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 13, 2020. . https://doi.org/10.1101/2020.05.09.20096511 doi: medRxiv preprint Differentially expressed genes were identified between ACE2+ and ACE2-PTEC in DKD, BKVN, and COV-AKI samples using the FindAllMarkers Seurat function. For DKD and BKVN signatures, all genes with adjusted p-value < 0.05 and found in at least 10% of cells were selected. Due to the comparatively small number of cells in COV-AKI sample, ACE2+ coregulated genes that passed a nominal uncorrected p-value threshold of 0.05 and were found in at least 10% of cells were selected (Supplementary Tables 2-5). SARS-CoV-2 relevant gene sets were compiled from multiple published sources: 1) Bojkova_proteome is a list of differentially expressed proteins in Caco-2 cell line following SARS-Cov-2 infection (p < 0.05 at any time point) 30 ; 2) Bojkova_translatome is a list of proteins with differential translatome kinetics following SARS-CoV-2 infection (p < 0.05 at any time point) 30 ; 3) Gordon_interactome is a set of host proteins identified as physically interacting with SARS-CoV-2 viral proteins in HEK-293T cells 31 ; 4) Zhou_interactome is a literature-curated list of genes related to diverse coronaviruses 17 ; 5) Blanco-Melo_A459 is a list of differentially expressed genes in response to SARS-CoV-2 infection in human alveolar adenocarcinoma cells 32 ; 6) Blanco-Melo_NHBE is a list of differentially regulated genes in response to SARS-CoV-2 infection in normal human bronchial epithelial cells 32 . ACE2 was removed from ACE2+ co-expressed gene sets before computing overlaps. P-values were computed using the hypergeometric test with a count of 20,000 genes used as background. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 13, 2020. . To determine the biological processes and pathways in the ACE2+ differentially expressed gene sets, functional network clustering of the PTEC gene functional network from GIANT 2.0 33, 34 . This network was generated through regularized Bayesian integration of 61,400 publicly available expression, physical interaction and other omics experiments to generate a fully connected weighted graph representing functional relationships in biological pathways in PTEC. Community clustering in the network was performed to identify tightly connected sets of genes using HumanBase.io module detection function 35 Table 5 ). For DKD and BKVN signatures, all genes with adjusted p-value < 0.05 and found in at least 10% of cells were selected. Due to the comparatively small number of cells in COV-AKI sample, ACE2+ co-regulated genes that passed a nominal uncorrected p-value threshold of 0.05 and were found in at least 10% of cells were selected. The Gene Ontology enrichment outputs are Supplementary Tables 7-10 and can be interactively explored at hb.flatironinstitute.org/covid-kidney. Spearman correlation were used to evaluate the association of the steady state average PTEC specific gene expression levels of ACE2 and age at time of biopsy for the 44 DKD samples. The Mann-Whitney non parametric test was applied to evaluate the gender and treatment difference of ACE2 expression in the DKD cohort. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 13, 2020. A dynamic user-friendly interface at HumanBase (hb.flatironinstitute.org/covid-kidney) is available for researchers to explore the functional networks of gene expression signatures. We studied ACE2 expression and viral infection associated processes through scRNASeq and functional network analysis (Figure 1 ). Kidney cell expression profiles were obtained from early DKD (n=44), LD (n=7), and BKVN (n=3) kidney biopsies and a urine sample from a COV-AKI patient. Baseline clinical characteristics for All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 13, 2020. . https://doi.org/10.1101/2020.05.09.20096511 doi: medRxiv preprint the DKD cohort are provided in Table 1, for BKVN in Supplementary Table 1, All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. To define the functional context of ACE2 expression signatures in kidney disease, we first identified gene expression programs specifically co-activated with the SARS-CoV-2 receptor expression. To this end the expression profiles of ACE2+ PTEC were compared to the ACE2-PTEC cell population in both DKD and BKVN. The resulting gene expression signatures (with pvalue adjusted Bonferroni cutoff of 0.05 and expressed in at least 10% of cells) identified sets of genes whose expression is specifically induced in ACE2+ cells in LD (Supplementary Table 2 ), Table 3 ) and BKVN (Supplementary Table 4 ). Critically, we then focused on identifying shared gene expression programs in ACE2+ PTEC in both DKD and BKVN samples (Supplementary Table 6 ). We also analyzed the overlap of the ACE2+ signatures with SARS-CoV-2-implicated genes and ACE2 signature genes from multiple publicly available, independent studies and found significant shared components across SARS-CoV-2 datasets (Table 2) . Specifically, genes associated with ACE2+ expression in DKD were also significantly enriched among host proteins identified to interact with SARS-CoV-2 proteins 33 , host genes upregulated in SARS-CoV-2infected cells 30, 32 , and host proteins previously reported to interact with multiple coronaviruses 17 (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. To functionally characterize these ACE2+ expression signatures, we projected these genes onto networks that represent pathways active in normal PTEC. Intuitively, the HumanBase functional network is constructed by probabilistically integrating a large compendium of thousands of public omics datasets (including expression, protein-protein interaction and motifs) to represent how likely it is that two genes act together in biological processes that function in PTEC 33, 34 . We clustered genes differentially expressed in ACE2+ vs ACE2-PTEC of the DKD kidneys in this functional network ( Figure 3A, Supplementary Table 7) . The resulting modules contain key processes of tubular function and failure, including renal tubule development and ion transport, general cellular processes such as cell cycle control and translational regulation, as well as disease signals including inflammatory pathway activation (IL-2 production, JAK/STAT signaling, and Wnt signaling). Importantly, we also identify key Table 8 ). The congruence between the DKD and BKVN signals at both gene and functional process levels supports the robustness of this ACE2+ molecular signature and suggests that at least some of the programs related to establishment of productive viral infection and to innate immune responses seen in BKVN are also present in the DKD ACE2+ PTEC even without an active viral infection ( Figure 3B, Supplementary Table 9 ). We hypothesize that some of these All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 13, 2020. . expression programs could be permissive to viral infection, along with the upregulation of ACE2 receptor. We then assessed differential ACE2 expression in PTEC in DKD compared to LD and found a significant overexpression of ACE2 steady state mRNA levels in DKD (adjusted p-value < 11.79 e-23; in cells with normalized read count >= 1.25) ( Figure 3A ). We focused on cells with ACE2 expression above 1.25 to exclude the ambient ACE2 mRNA expression (~ 1 normalized read count) observed in most cell types as seen in Figure 2 . Importantly, proteases thought to be permissive for coronavirus infection including ANPEP, BSG, CTSL, DPP4, ENPEP, FURIN and TMPRSS2 appear to co-express with ACE2 in PTEC ( Figure 4A, B) . The expression levels of ACE2 transcripts in PTEC did not show a significant association with baseline DKD participant characteristics (age/gender, Figure 4C and D) or treatment exposures to renin angiotensin aldosterone system (RAAS) inhibitors ( Figure 4E-G) . Finally, the PTEC cellular response in COVID-19 was assessed using urinary scRNAseq analysis constrained to PTEC gene expression clusters. Single cell data analysis of the cells isolated from the urine of a SARS-CoV-2 infected patient yielded 273 cells that passed our quality control threshold. Unsupervised clustering of these cells produced three kidney cell clusters, including PTEC, and two immune cell clusters ( Figure 5A and B) . 42% of the cells in the PTEC cluster showed ACE2 expression. In addition to ACE2, proteases including ANPEP, BSG, CTSL, DPP4, ENPEP, FURIN and TMPRSS2 were expressed in PTEC from the COV-AKI sample ( Figure 5C, Supplementary Table 5 ). BSG and ANPEP were found to be expressed in all proximal cells identified ( Figure 5C ). Supplementary Table 11 provides the PTEC markers identified in COV-AKI. Notably, genes nominally upregulated in ACE2+ vs ACE2-COV-AKI All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Figure 5) . With the pandemic spread of COVID-19 and the increased COVID-19 morbidity and mortality in patients with diabetes and kidney disease it is imperative to define the underlying mechanism of the excessive risk in patients with diabetes and rapidly develop protective strategies for risk reduction. Autopsy studies of patients with COVID-19 are consistent with direct viral infection of the kidney and that even within the kidney 12, 36 , expression of molecules that mediate viral entry may be cell type-specific 12 . Susceptibility to viral entry may be enhanced by DKD associated changes in SARS-CoV-2 receptor expression in kidney tissue. Our study set out to establish the cell population within the kidney expressing ACE2 and associated SARS-CoV-2 receptors mRNA in healthy individuals and in patients with DKD and viral associated kidney diseases. ACE2 intra-renal gene expression was localized narrowly to PTEC in surveillance kidney biopsies obtained from LD and in research kidney biopsies from patients with DKD. PTEC are the prime localization of SARS-CoV-2 infection in the kidney 36 , supporting the link between receptor expression and cellular toxicity in kidneys, as well. In contrast to findings in other cell types, little co-expression of ACE2 with TMPRSS2 co-receptor used for SARS-CoV-2 viral entry was observed, suggesting other cell surface proteases may be associated with viral infection and uptake in the kidney 12, 36 . Despite the fact that none of the diabetic patients in this study were infected with SARS-CoV-2 at the time of biopsy, our pathway analysis showed that processes including infection, protein All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 13, 2020. . https://doi.org/10.1101/2020.05.09.20096511 doi: medRxiv preprint processing and antigen presentation were enriched in high versus low ACE2 expressing PTEC in DKD. The results from this analysis suggest that ACE2 is co-expressed with a set of genes also having functions in viral entry, establishment of viral replication, and innate immunity. A key finding of our study is that ACE2+ proximal tubule cells of patients with DKD had significantly higher ACE2 mRNA levels than those of LDs. These higher ACE2 levels may make these cells more permissive to initial viral entry and subsequent increased cellular cytotoxicity in DKD, especially as viral replication and processing machinery also appear to be upregulated in the ACE2+ PTEC. This includes genes involved in endomembrane transport and lysosomal compartments, which we hypothesize could facilitate viral entry and exit, and proteins that interact with viral transcription and translation. It is intriguing that a number of the ACE2+ signature genes interact with predicted or demonstrated SARS-CoV-2 anti-host defense response proteins 31 , including M (suppresses type1 interferon response and refine cytokine responses) 37, 38 , N (inhibit autophagy and apoptosis of infected cells) 39 , Orf3a (modulates IL-1B) 40 , and Nsp6 (inhibits phagosome expansion and targeting of viral products to lysosome) 41 . These interactions may simply reflect an element of the functional host response program to cellular stress in DKD. However, if these interactions also involve negative functional regulation of host-defense proteins, ACE2+ proximal tubule cells could be impaired in their ability to mount an appropriate host response, providing one explanation of their observed vulnerability to infection to SARS-CoV-2. The inflammatory signals observed in the DKD and BKVN PTEC are expected in the context of chronic kidney disease as reported previously by studies in this and other DKD patient populations 42, 43 . However, these data also could support the hypothesis that this inflammatory state could be maladaptive for a response to viral infection. Recent work on lung epithelial cells infected with SARS-CoV-2 has shown a blunted interferon response despite robust cytokine All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 13, 2020. . https://doi.org/10.1101/2020.05.09.20096511 doi: medRxiv preprint production still being present 32 ; a potential hallmark of SARS-CoV-2 pathology. If a comparably asymmetric inflammatory response is already present in DKD kidney tissue, viral infection could amplify the pre-existing inappropriate stress response, increasing cytopathology and enabling further viral propagation. If ACE2 upregulation is indeed correlated with a cellular inflammatory state that permits enhanced viral pathogenesis and replication in multiple organ systems, this linkage could provide an evolutionary driving factor to select ACE2 for viral entry. The PTEC specific ACE2+ networks observed in DKD, BKVN and COV-AKI allow for a wide range of exploratory studies to generate starting points for functional evaluation. Given the urgent need for novel therapeutic options, intersecting the ACE2 regulatory networks with the complex landscape of interactions between drugs and the gene signature defined here can help to identify therapeutic targets for further study. We identified ACE2+ co-regulated genes annotated in DrugBank 44 as targets of medications regularly taken by patients with DKD. For example, insulin (INSR, in M5 in Figure 3B ) and the IGF-1 receptor (IGF1R, in M3 in Figure 3B) are overexpressed in ACE2+ PTEC as is DPP4 (in M5 in Figure 3B ), a protease associated with entry of other coronaviruses and a key factor in glucose metabolism, targeted by DPP4 inhibitors already in clinical use. These interactions suggest that the ACE2+ co-expression signature may be influenced by a number of commonly used drugs and can form the starting points for further mechanistic and epidemiological studies. As data emerged on the role of ACE2 as a receptor for SARS-CoV-2, significant concerns were raised about risks of RAAS inhibitors frequently prescribed in patients with diabetes and chronic kidney diseases. Based on animal models RAAS inhibitors were predicted to increase ACE2 levels raising the potential concern for higher COVID-19 morbidity and mortality 22, 45, 46 . However, our findings in the DKD cohort do not support the animal model data and do not provide any evidence of transcriptional upregulation of ACE2 mRNAs in PTEC by RAAS All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 13, 2020. . inhibition. In parallel, a series of case control, database and Electronic Health Record studies did not find an association of RAAS inhibitors with poor outcomes in three independent cohorts of patients with COVID19 and RAAS exposures [47] [48] [49] , consistent with our scRNAseq studies. ACE2 gene expression studies of kidney tissue to date have been inconclusive concerning regulation in kidney disease [50] [51] [52] [53] An important element of this study is the confirmation of ACE2 and associated genes and modules in PTEC derived from a COV-AKI sample. The association of ACE2 expression with proteins involved in innate immune response and viral entry and replication provides an intriguing hypothesis for how SARS-CoV-2 can cause increased cytopathology in ACE2+ cells. However, these findings have to be considered exploratory, as they are based on a small population of PTEC which have lost their structural kidney tissue context in a single patient with COV-AKI. Confirmation of these findings in a larger cohort of COVID-19 patients and ideally from kidney tissue is needed. Our study is also limited by focusing on mRNA levels, which only capture one of several levels of regulation of ACE2 function. For this reason, our analyses focused on the definition of co-All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 13, 2020. In summary, our work identifies the regulation and associated cellular machinery of ACE2 and associated SARS-CoV-2 co-receptors in PTECs in kidney health, metabolic and viral disease. The SARS-CoV-2 receptor associated networks are now available to seed further research into urgently needed therapeutic strategies for COVID-19. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. To identify SARS-CoV-2 receptor networks in diabetic kidney disease (DKD), BK-Virus nephropathy (BKVN) and COVID-19 associated acute kidney injury (COV-AKI), we used scRNAseq profiling and integrative network analyses. (A) Biopsy samples from DKD, BKVN, and living donors (LD) were processed for scRNAseq profiling and cell-type clusters were identified from each sample. (B) For each dataset, we identified genes that are differentially expressed in ACE2+ compared to ACE2-proximal cells (PTEC). We compared DKD and BKVN ACE2+ co-expression sets to identify a disease-specific signature, and we also compared each signature to published SARS-CoV-2-relevant gene sets. (C) In parallel, we evaluated the association of ACE2 expression levels with clinical characteristics, including exposure to RAAS blockers and ACE inhibitors in DKD (D) For each ACE2+ co-expression signature, we identified modules in a proximal tubule-specific functional network and found modules enriched with biological processes relevant to viral infection. (E) We compared expression of ACE2 and key proteases between LD and DKD PTEC. (F) Finally, we explored this ACE2+ signature in PTEC cells in BKVN kidney tissue and a COV-AKI urine sample. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 13, 2020. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Supplementary Table 1 : Clinical characteristics of BKVN cohort. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 13, 2020. . https://doi.org/10.1101/2020.05.09.20096511 doi: medRxiv preprint COVID-19 Pandemic, Corona Viruses, and Diabetes Mellitus COVID-19 And Older Adults: What We Know Prevalence of Underlying Diseases in Hospitalized Patients with COVID-19: a Systematic Review and Meta-Analysis COVID-19, ACE2, and the cardiovascular consequences COVID-19, diabetes mellitus and ACE2: The conundrum Glomerular Localization and Expression of Angiotensin-Converting Enzyme 2 and Angiotensin-Converting Enzyme: Implications for Albuminuria in Diabetes Coronavirus infections and type 2 diabetes-shared pathways with therapeutic implications Integrated analyses of single-cell atlases reveal age, gender, and smoking status associations with cell type-specific expression of mediators of SARS-CoV-2 viral entry and highlights inflammatory programs in putative target cells Organ distribution of severe acute respiratory syndrome (SARS) associated coronavirus (SARS-CoV) in SARS patients: implications for pathogenesis and virus transmission pathways Chronic kidney disease is associated with severe coronavirus disease 2019 (COVID-19) infection Kidney impairment is associated with in-hospital death of COVID-19 patients Ultrastructural Evidence for Direct Renal Infection with SARS-CoV-2 Angiotensin Converting Enzyme 2: SARS-CoV-2 Receptor and Regulator of the Renin-Angiotensin System ACE2 receptor expression and severe acute respiratory syndrome coronavirus infection depend on differentiation of human airway epithelia Trilogy of ACE2: A peptidase in the reninangiotensin system, a SARS receptor, and a partner for amino acid transporters Coronavirus Disease 2019 (COVID‐19) and Cardiovascular Disease: A Viewpoint on the Potential Influence of Angiotensin‐Converting Enzyme Inhibitors/Angiotensin Receptor Blockers on Onset and Severity of Severe Acute Respiratory Syndrome Coronavirus 2 Infection Systemic analysis of tissue cells potentially vulnerable to SARS-CoV-2 infection by the protein-proofed single-cell RNA profiling of ACE2, TMPRSS2 and Furin proteases Relevance of enriched expression of SARS-CoV-2 binding receptor ACE2 in gastrointestinal tissue with pathogenesis of digestive symptoms, diabetes-associated mortality, and disease recurrence in COVID-19 patients Human Kidney is a Target for Novel Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) Infection SARS-CoV-2 Cell Entry Depends on ACE2 and TMPRSS2 and Is Blocked by a Clinically Proven Protease Inhibitor Evidence that TMPRSS2 Activates the Severe Acute Respiratory Syndrome Coronavirus Spike Protein for Membrane Fusion and Reduces Viral Control by the Humoral Immune Response Should COVID-19 Concern Nephrologists? Why and to What Extent? The Emerging Impasse of Angiotensin Blockade Regulation of angiotensin converting enzyme 2 (ACE2) in obesity: implications for COVID-19 Changes in Albuminuria But Not GFR are Associated with Early Changes in Kidney Structure in Type 2 Diabetes Indexing glomerular filtration rate for body surface area in obese patients is misleading: concept and example Single cell transcriptomics identifies focal segmental glomerulosclerosis remission endothelial biomarker The immune cell landscape in kidneys of patients with lupus nephritis Fast, sensitive and accurate integration of single-cell data with Harmony Comprehensive Integration of Single-Cell Data Sofosbuvir Activates EGFR-Dependent Pathways in Hepatoma Cells with Implications for Liver-Related Pathological Processes A SARS-CoV-2-Human Protein-Protein Interaction Map Reveals Drug Targets and Potential Drug-Repurposing SARS-CoV-2 launches a unique transcriptional signature from in vitro, ex vivo, and in vivo systems GIANT 2.0: genome-scale integrated analysis of gene networks in tissues Understanding multicellular function and disease with human tissue-specific networks Genome-wide prediction and functional characterization of the genetic basis of autism spectrum disorder Renal histopathological analysis of 26 postmortem findings of patients with COVID-19 in China APOL1-associated glomerular disease among African-American children: a collaboration of the Chronic Kidney Disease in Children (CKiD) and Nephrotic Syndrome Study Network (NEPTUNE) cohorts Severe acute respiratory syndrome coronavirus M protein inhibits type I interferon production by impeding the formation of TRAF3.TANK.TBK1/IKKepsilon complex Adapting the Stress Response: Viral Subversion of the mTOR Signaling Pathway Severe acute respiratory syndrome coronavirus ORF3a protein activates the NLRP3 inflammasome by promoting TRAF3-dependent ubiquitination of ASC Coronavirus NSP6 restricts autophagosome expansion A molecular morphometric approach to diabetic kidney disease can link structure to function and outcome Modular activation of nuclear factor-kappaB transcriptional programs in human diabetic nephropathy DrugBank: a comprehensive resource for in silico drug discovery and exploration The effect of angiotensin converting enzyme inhibitors and angiotensin receptor blockers on death and severity of disease in patients with coronavirus disease 2019 (COVID-19): A meta-analysis Speculation is not evidence: antihypertensive therapy and COVID-19 Cardiovascular Disease, Drug Therapy, and Mortality in Covid-19 Renin-Angiotensin-Aldosterone System Blockers and the Risk of Covid-19 Renin-Angiotensin-Aldosterone System Inhibitors and Risk of Covid-19 ACE and ACE2 in kidney disease Exploring diseases/traits and blood proteins causally related to expression of ACE2, the putative receptor of 2019-nCov: A Mendelian Randomization analysis Decreased glomerular and tubular expression of ACE2 in patients with type 2 diabetes and kidney disease Increased ACE 2 and Decreased ACE Protein in Renal Tubules From Diabetic Mice Tissue-based map of the human proteome Organoid single cell profiling identifies a transcriptional signature of glomerular disease No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity We thank Ms. Lois Jones, RN (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint this version posted May 13, 2020. All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint this version posted May 13, 2020. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Table 11 : PTEC-specific markers in the COV-AKI sample.All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint this version posted May 13, 2020. All rights reserved. No reuse allowed without permission.(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.The copyright holder for this preprint this version posted May 13, 2020.