JAK Inhibitor Blocks COVID-19-cytokine-induced JAK-STAT-APOL1 Signaling in Glomerular Cells And Podocytopathy in Human Kidney Organoids

Abstract 

COVID-19 infection causes collapse of glomerular capillaries and loss of podocytes, terminating in a severe kidney disease called COVID-19-associated nephropathy (COVAN). The underlying mechanism of COVID-19 is unknown. We hypothesized that cytokines induced by COVID-19 trigger the expression of pathogenic APOL1 via JAK-STAT signaling, resulting in podocyte loss and COVAN phenotype. Here, based on nine biopsy-proven COVAN cases, we demonstrated for the first time that APOL1 protein is abundantly expressed in podocytes and glomerular endothelial cells (GECs) of COVAN kidneys but not in controls. Moreover, a majority (77.8%) of COVAN patients carried two APOL1 risk alleles. We showed that recombinant cytokines induced by SARS-CoV-2 act synergistically to drive APOL1 expression through the JAK-STAT pathway in primary human podocytes, GECs, and kidney micro-organoids derived from a carrier of two APOL1 risk alleles but were blocked by JAK1/2-inhibitor, baricitinib. We demonstrated for the first time that cytokine-induced JAK-STAT-APOL1 signaling reduced the viability of kidney organoid podocytes but was rescued by baricitinib. Together, our results support the conclusion that COVID-19-induced cytokines are sufficient to drive COVAN-associated podocytopathy via JAK-STAT-APOL1 signaling and that JAK-inhibitor could block this pathogenic process. These findings suggest that JAK-inhibitors may have therapeutic benefits for managing cytokine-induced APOL1-mediated podocytopathy.

Supportive Service Of Wecistanche-The largest cistanche exporter in the China:

Email:wallence.suen@wecistanche.com 

Whatsapp/Tel:+86 15292862950

Shop For More Specifications Details:

https://www.xjcistanche.com/cistanche-shop

GET NATURAL ORGANIC CISTANCHE EXTRACT WITH 25% ECHINACOSIDE AND 9% ACTEOSIDE FOR KIDNEY INFECTION

Introduction

Kidney failure is a devastating complication of COVID-19 infection. Up to 50% of inpatient and 70% of intensive care unit COVID-19 admissions are complicated by acute kidney injury (AKI), which in turn increases mortality by 30-50% (1, 2). A kidney biopsy case series revealed that collapsing glomerulopathy is the most common histopathologic diagnosis in COVID-19-associated AKI (3). A distinctive feature of COVID-19-associated collapsing glomerulopathy (abbreviated COVAN) is its near-exclusive predilection for African Americans or Blacks who carry two risk alleles of Apolipoprotein L1 (APOL1) (3, 4). The two risk alleles (named G1 and G2) emerged as coding variants in the APOL1 gene and confer protection against African trypanosomiasis. However, carriage of G1G1, G2G2, or G1G2 (termed high-risk genotypes) increases the risk of a spectrum of kidney disease and explains much of the excess risk of non-diabetic kidney disease among African Americans (5-8). An estimated 13% of African Americans carry high-risk APOL1 genotypes (7). During the COVID-19 pandemic, studies found that a remarkable 92% of biopsy-proven COVAN cases were in carriers of high-risk APOL1 genotypes, 61% of whom required dialysis at presentation (3, 9). These findings establish APOL1 variants as a major contributor to the racial disparity in COVID-19 health outcomes. Despite this impressive association, the cellular mechanism that connects high-risk APOL1 genotypes to SARSCoV-2 infection and the pathogenesis of collapsing glomerulopathy of COVAN remains unknown. 

The strong epidemiologic association between the high-risk APOL1 genotype and COVAN has led to the hypothesis that COVID-19-induced expression of APOL1 G1 or G2 in podocytes and glomerular endothelial cells-the kidney cells impacted in collapsing glomerulopathy-drives pathogenesis of COVAN. This hypothesis was supported by recent reports that transgenic overexpression of APOL1 risk alleles in mouse podocytes or glomerular endothelial cells caused podocytopathy, endotheliopathy, glomerulopathy, and clinical manifestations of kidney failure (10-14). These murine disease models suggest that the mechanism that underlies COVID-19-induced APOL1 expression would be a potential therapeutic target for COVID-19. However, there are two important unknowns. One, it is unknown whether APOL1 protein expression is upregulated in the glomeruli of patients with COVAN. Two, it is unknown whether SARS-CoV-2 induces APOL1 expression directly by viral infection of kidney cells or indirectly via the effects of SARS-CoV-2-induced cytokine storm.  

The failure to detect SARS-CoV-2 in kidney biopsies of COVAN patients provides indirect support for the hypothesis that COVAN likely results from the effects of cytokine storm rather than from direct viral infection of kidney parenchyma. The occasional detection of SARS-CoV-2 viral particles has been in autopsy kidney specimens in which the confounding effect of tissue autolysis could not be excluded (3, 4, 15, 16). Several inflammatory cytokines and chemokines have been noted to be upregulated in the sera of patients with COVID-19 and/or COVAN (4, 17, 18). This list includes cytokines such as interferons alpha, beta, gamma, and TNF which are previously known to upregulate APOL1 expression. However, the list also includes several inflammatory cytokines that are robustly upregulated by COVID-19 infection. It is unknown whether these cytokines have additive, synergistic, or antagonistic effects on APOL1 expression and associated podocytopathy. 

In the current study, we addressed these knowledge gaps by leveraging kidney biopsies obtained from nine patients with COVAN and two human control kidneys to investigate whether and where APOL1 protein is expressed in COVAN and control kidneys. We profiled eighteen COVID-19-induced cytokines to identify eight cytokines that were sufficient, in the absence of SARS-CoV-2, to synergistically induce APOL1 expression in primary human glomerular cells and cause podocytopathy in human kidney micro-organoids. This study not only offers the first proof in a human-derived experimental model that COVID-19 cytokine storm induces APOL1 expression and podocytopathy, it also identifies the common signaling pathway that mediates the pathogenic effects. This study has implications that could impact strategies for screening and treating COVID-19 in Black and Hispanic patients. It raises questions about safety of supplemental interferons as COVID-19 therapy in Black and Hispanic individuals who carry high-risk APOL1 genotypes (19, 20). 

Results 

APOL1 expression is upregulated in podocytes and glomerular endothelial cells of COVAN patients. To investigate whether patients with biopsy-proven diagnosis of COVAN have elevated expression of APOL1 protein in their podocytes and glomerular endothelial cells (GECs), we performed immunohistochemical co-staining of APOL1, synaptopodin (an actin-associated protein of differentiated podocytes), and CD31 (an endothelial cell marker) on kidney biopsies of two patients with COVAN diagnosis (Figure 1). APOL1 expression was abundant in glomeruli of both case 1, which was biopsied ten months after COVID-19 diagnosis (Figure 1A-F), and case 6, who was biopsied nine days after COVID-19 diagnosis (Figure 1G-L). In both patients, there was strong APOL1 staining in synaptopodin-positive podocytes (arrow; Figure 1B, E, H, and K) and along CD31-positive glomerular endothelium (arrowhead; Figure 1C, F, I, and L). The presence of APOL1 protein in podocytes and GECs at times 9 days and 10 months after diagnosis of COVID-19 infection suggests that APOL1 expression is induced early and may persist in the glomeruli for several months, long after the triggering COVID-19 infection has resolved. 

APOL1 expression is upregulated in biopsy tissue of COVAN cases but not in controls. To further evaluate the generalizability of these immunohistochemical findings, we identified a total of nine COVAN cases with available biopsy tissue for genotyping and IHC (Figure 2) as well as two control patients, including one autopsy control of a patient who had COVID-19 infection but did not develop AKI (Figure 2B and 2C). The classic histopathologic features of COVAN included glomerular capillary tuft collapse with adjacent podocyte hypertrophy and proliferation, often with the podocyte protein reabsorption droplets associated with glomerular proteinuria (Supple Fig1 and Supple Fig2). APOL1 IHC staining was absent in all glomeruli of controls (Figure A-C) but present in the glomeruli of all nine COVAN patients (Figure 2D-X). APOL1 staining was abundant in the cytoplasm of podocytes, GECs, and in some parietal epithelial cells (Figure 2N, hashed circle) (Figure1, Figure 2 and Suppl Fig3). APOL1 protein could be seen in glomeruli with open capillaries as well as in areas of glomerular collapse. The apparent presence of APOL1 protein within some capillary lumen likely represented circulating APOL1, which is produced primarily by the liver (21). Moreover, APOL1 was also seen in peritubular capillaries and in injured tubular epithelial cells (asterisks). The specificity and significance of this latter finding is unclear. Notably, seven of the nine COVAN cases carried a high-risk APOL1 genotype (Figure 2D-V). The other two cases, case 9 and case 4, carried low-risk G0G0 genotypes (Figure 2W and 2X). Despite being of low-risk genotype, case 9 APOL1 expression was comparable to those of the seven high-risk cases (Figure 2W). There were only two glomeruli in case 4 kidney biopsy slide, and APOL1 expression was lower in these glomeruli (Figure 2X). Together, these findings demonstrate that the basal kidney APOL1 expression is low in glomeruli of individuals without glomerular injury, even when the individual has COVID-19 infection; whereas APOL1 expression becomes upregulated in podocytes and GECs in the setting of COVAN in 89% of our cases. 

As shown in Table 1, seven of the nine patients (77.8%) with biopsy-proven COVAN self-identified as African American. Six of these seven patients (85.7%) carried high-risk APOL1 genotypes (four G1G1, one G1G2, one G2G2). By comparison, 13% of African Americans carry high-risk APOL1 genotype. The remaining two patients self-identified as White Hispanic but notably, one of them also carried a high-risk APOL1 genotype. In total, seven of the nine COVAN patients (77.8%) carried high-risk APOL1 genotypes. The median age of the cases was 51 years (range 37-60). All nine cases developed acute kidney injury and had varying degrees of proteinuria, ranging from subnephrotic to nephrotic range (1.4- 14 g/24hrs). Most biopsies were performed at least 1 month after COVID-19 infection, with the exception of case 6 who was biopsied 9 days after a positive PCR test. One patient's biopsy was not pursued until 10 months after infection secondary to incomplete recovery. All nine patient biopsies exhibited collapsing glomerulopathy, tubular injury, and interstitial inflammation. Endothelial tubular reticular inclusions were not observed in any of the cases. Notably, in the two COVAN biopsies that were tested for direct SARS-CoV-2 viral infection by IHC and in situ hybridization, no virus was detected (data not shown). 

Recombinant COVID-19-Induced cytokines synergistically upregulate APOL1 expression in primary human glomerular endothelial cells and podocytes. To investigate whether COVID-19-induced cytokine storm is sufficient to trigger APOL1 expression in human glomerular cells, we cultured primary human podocytes isolated from deceased donor kidney and primary human glomerular endothelial cells (GECs) in 1 of 18 cytokines and chemokines previously reported to be elevated in the serum of patients with SARS-CoV-2 (Figure 3A) (4, 17). Podocyte identity was confirmed with multiple podocyte markers including Wilms tumor1, synaptopodin, nephrin, and podocalyxin (Figure 3B and Suppl Fig 4A-B). GEC identity was confirmed by expression of PECAM1 relative to human embryonic kidney 293 cells (HEK) (Figure 3C). Induced APOL1 expression was quantitated by qPCR and immunoblot after 48hr treatment in GECs (Figure 3D and 3F) and podocytes (Figure 3E and 3G). Consistent with prior report (22), we found that interferons (gamma > beta > alpha) robustly induced expression of APOL1 in both GECs and in podocytes. Similarly, we found that TNF also induced a modest APOL1 expression in GECs and podocytes. Unexpectedly, we found that three cytokines-IL-6, IL-1β, and IL-18, which were previously unrecognized as inducers of APOL1 expression, also individually induced modest APOL1 expression in GECs or podocytes. Notably, the combination of all 18 recombinant cytokines produced a synergistic upregulation of APOL1 that was an order of magnitude higher than that produced by any of the interferons alone (Figure 3D). These effects were not only observed with cytokine concentration of 50ng/mL (Figures 3D and 3E) but also at 20ng/mL and 10ng/mL (Supplemental Figure 5). Cytokine conditions inducing >1.5 fold APOL1 transcript compared to media-treated control were further analyzed for significance. Significance was assessed using an unpaired t-test with Holm-Sidak correction for multiple comparisons. The P-values reported are the adjusted p-values. These results expand the list of physiologic cytokines that are capable of inducing APOL1 expression beyond the well-recognized interferons and TNF. Importantly, the findings also suggest that the synergy of COVID-19-induced cytokines may be more relevant for APOL1 regulation than the impact of an isolated cytokine alone. 

JAK-STAT signaling mediates COVID-19-cytokine-induced APOL1 expression. We next investigated whether COVID-19 induced cytokines upregulate APOL1 expression via a common intracellular signaling pathway that could be exploited as a therapeutic target. It was previously reported that interferon induction of APOL1 is mediated by JAK-STAT1/2 (22, 23). Signaling through the IL-6 receptor has been shown to be mediated by STAT3, and both IL-1β and TNF are reported to indirectly activate STAT3 (24, 25). Of interest, Meliambro et al recently reported upregulation of phospho-STAT3 in the biopsy tissue of a case of COVAN and HIV-associated nephropathy (HIVAN) compared to control (26). Based on this background information, we hypothesized that the JAK1/2-STAT1/2/3 pathways are the primary mediators of the effects of COVID-19-induced cytokines in driving APOL1 expression. To test this hypothesis, we determined the state of these signaling pathways by measuring the phosphorylated STAT1, 2, and 3 in lysates of GECs after culturing them in individual or combined cytokines for 48 hours (Figure 3F). Type I interferons (IFNα and IFNβ) increased phosphorylation of STAT1-3 while IFNγ upregulated phosphorylation of STAT 1 and 3. IL-1β, TNF, and IL-6 increased phosphorylation only of STAT3. Combined cytokines increased phosphorylation of STAT1-3. Knowing that JAK1 and JAK2 are the primary upstream protein kinases that phosphorylate STAT1-3, we hypothesized that inhibition of JAK1/2 would block APOL1 expression induced by "all cytokines". Consistent with this prediction, we found that a JAK1/2-specific inhibitor, baricitinib, significantly reduced APOL1 mRNA and APOL1 expression by all-cytokine-treated GECs and primary podocytes (Figure 3DG). Together, these results demonstrate that JAK-STAT signaling is the primary pathway that mediates COVID-19-cytokine-induced APOL1 expression. 

COVID-19-induced cytokines are sufficient to drive APOL1 expression in human iPSC-derived kidney micro-organoids via the JAK-STAT pathway. Human kidney micro-organoid is a proven platform for modeling human kidney disease and facilitating clinical translation. We asked whether the results we obtained from primary human podocytes and GECs were generalizable and validated by a human-derived kidney micro-organoid model. Therefore, we generated kidney micro-organoids from induced pluripotent stem cells (iPSCs) of an African American carrier of the G1G2 APOL1 genotype to investigate APOL1 regulation, expression, and outcomes in this model (Figure 4A). We cultured kidney microorganoids in IFNγ 10ng/mL or a combination of eight cytokines (IFNγ, IFNα, IFNβ, IL-18, IL-8, IL-6, TNF, IL-1β) each at 10ng/mL either in the absence or presence of baricitinib,10µM for 24 hours. These eight cytokines were chosen due to their observed regulation of APOL1 expression in the preceding experiments. Podocytes and tubular epithelial cells in the kidney micro-organoids were marker-confirmed (Figure 4B). Consistent with the literature (27), endothelial cells were underrepresented in the kidney micro-organoids (data not shown). We discovered that basal APOL1 protein expression was low in micro-organoids. IFNγ treatment induced substantial APOL1 expression, with the highest intensity co-localized to areas of podocyte marker, podocalyxin. The cocktail of cytokines induced an outsized and robust APOL1 expression throughout the micro-organoid structure when compared to other treatments. Groups treated with IFNγ plus baricitinib and all cytokines plus baricitinib showed no APOL1 expression, consistent with complete inhibition of cytokine effect. The APOL1 expression in kidney micro-organoid podocytes was reminiscent of that seen in podocytes of COVAN patients. However, unlike COVAN kidneys in which no significant APOL1 expression was seen in healthy tubular epithelial cells, kidney micro-organoid E-Cadherin-positive tubular epithelial cells expressed APOL1. This difference could be due to differences in membrane cytokine receptors or epigenetic factors that impact protein expression in the immature tubules of the kidney micro-organoids. In summary, human iPSC-derived kidney micro-organoids cultured with COVID-19-induced cytokines show robust upregulation of pathogenic G1G2 APOL1 protein and the expression was blocked by inhibition of the JAK-STAT signaling. 

Cytokine-induced JAK-STAT-APOL1 signaling reduced the viability of kidney micro-organoid podocytes which was rescued by JAK-inhibitor. Finally, we asked if the G1G2 APOL1 expressed in kidney microorganisms impairs podocyte viability. We hypothesized that cytokine-induced variant APOL1 protein would cause podocyte loss-a hallmark phenotype of COVAN. To test this hypothesis, we isolated podocytes from kidney micro-organoids generated from iPSCs of a carrier of G1G2. The podocytes were cultured in IFNγ (10ng/mL), or a combination of eight cytokines (10ng/mL each), both in the presence and absence of baricitinib (10µM) for 96 hours (Figure 5A and B). Cytokine treatment robustly induced APOL1 expression and this expression was blocked by baricitinib, consistent with our earlier experiments (Figure 5C). Concordantly, cytokine treatment caused significant podocyte loss as indicated by viability assay and total cellular ATP (Figure 5D and 5E). Remarkably, baricitinib completely rescued the cytokine-induced podocyte loss. Together, these results support the conclusion that COVID-19-induced cytokines trigger JAK-STAT-APOL1 signaling which in turn causes podocyte injury and loss. The protective effect of JAK-inhibition on podocyte viability strongly supports this hypothesis. 

Discussion 

The major conclusions of the current study are that several COVID-19-induced cytokines beyond interferons act synergistically via JAK-STAT signaling to drive pathogenic APOL1 expression, resulting in podocyte injury and loss which is blocked by JAK inhibition. Based on a case series, we demonstrate for the first time that APOL1 protein is abundantly expressed in podocytes and GECs of patients diagnosed with COVID-19 but not in the glomeruli of healthy controls nor of COVID-19-positive but COVAN-negative control. In three experimental models, we demonstrate that recombinant cytokines upregulated in COVID-19 infection are sufficient to drive robust APOL1 expression, and unexpectedly, that the strong synergism produced by a combination of cytokines was mediated predominantly through a common intracellular signaling pathway. Collectively, our experimental evidence strongly supports a causal relationship between cytokine-induced JAK-STAT-APOL1 signaling and in vivo COVAN glomerular phenotype and supports further investigation into this therapeutic target.

The increased frequency of high-risk APOL1 genotype (77.8%) among patients with COVAN that we report here correlates with a recent international multi-center pathology review that reported high-risk APOL1 genotype in 91.7% of COVAN patients (3). Given that the frequency of high-risk APOL1 genotype in the general African American population is 13% (28), discovering a frequency of 77-90% in COVAN is profound and comparable to the 60-70% frequency reported in HIV-associated nephropathy (HIVAN) (28-31). The existence of the remaining 20-30% of COVAN (and HIVAN) cases who do not carry high risk APOL1 genotypes suggests the possibility of an APOL1-independent pathomechanism or the possibility that in some cases COVID-19-induced supraphysiologic expression of G0 APOL1 may also cause podocytopathy. Parsing these possibilities will require further studies. Nevertheless, we previously demonstrated in human embryonic kidney (HEK) cells with tetracycline-inducible APOL1 expression system that cytotoxicity of APOL1 is both variant- and dose-dependent (32, 33). Dose-dependent APOL1 cytotoxicity was also reported by other investigators in similar cell-based systems (34). Moreover, APOL1 transgenic mouse models have not only validated the causal link between APOL1 risk alleles and podocyte injury, but have demonstrated that the degree of podocytopathy correlated with APOL1 expression levels (10, 12-14, 35). Our discovery that eight out of nine COVAN cases demonstrated robust glomerular APOL1 expression relative to controls and the evidence that expression of endogenous APOL1 risk alleles causes podocytopathy in human kidney micro-organoids supports the causal link between APOL1 and podocytopathy. Conversely, the lack of APOL1 in the glomeruli of COVID-19 positive but AKI-negative G0G0 autopsy control suggests that COVID-19 infection without APOL1 induction is not a sufficient driver of COVAN disease.