Zika Virus Infection of the Human Glomerular Cells: Implications for Viral Reservoirs and Renal Pathogenesis

Abstract

Background

Zika virus (ZIKV) infection in the human renal compartment has not been reported. Several clinical reports have describe high-level persistent viral shedding in the urine of infected patients, but the associated mechanisms have not been explored until now. The current study examined cellular components of the glomerulus of the human kidney for ZIKV infectivity.

Methods

I infected primary human podocytes, renal glomerular endothelial cells (GECs), and mesangial cells with ZIKV. Viral infectivity was analyzed by means of microscopy, immunofluorescence, real-time reverse-transcription polymerase chain reaction (RT-PCR), and quantitative RT-PCR (qRT-PCR), and the proinflammatory cytokines interleukin 1β, interferon β, and RANTES (regulated on activation of normal T cells expressed and secreted) were assessed using qRT-PCR.

Results

I show that glomerular podocytes, renal GECs, and mesangial cells are permissive for ZIKV infection. ZIKV infectivity was confirmed in all 3 cell types by means of immunofluorescence staining, RT-PCR, and qRT-PCR, and qRT-PCR analysis revealed increased transcriptional induction of interleukin 1β, interferon β, and RANTES in ZIKV-infected podocytes at 72 hours, compared with renal GECs and mesangial cells.

Conclusions

The findings of this study support the notion that the glomerulus may serve as an amplification reservoir for ZIKV in the renal compartment. The impact of ZIKV infection in the human renal compartment is unknown and will require further study.

Zika virus (ZIKV) is a single-stranded RNA virus of the Flaviviridae family, genus Flavivirus, which also includes dengue, West Nile, Japanese encephalitis, and yellow fever viruses [1, 2]. ZIKV virus disease is transmitted to humans primarily through the bites of infected Aedes species mosquitoes and is emerging on a global scale [3, 4]. ZIKV outbreaks in Brazil during 2015 have quickly spread to >70 countries in the Americas and the Caribbean, infecting >2 million persons [5–7]. The geographic distribution of ZIKV continues to expand because of the widespread distribution of the mosquito vector and travel-associated infections [8]. ZIKV infection has been associated with a sporadic increase in the incidence of Guillain-Barré syndrome and microcephaly in infants [9–13]. There are data strongly supporting the notion that ZIKV can be sexually transmitted [14–16]. The significance of sexual transmission of the virus and the asymptomatic carrier state is unclear. Patient clinical presentations include fever, headache, arthralgia, rash, retro-orbital pain, myalgia, malaise, lymphadenopathy, and diarrhea, but in most cases symptoms were found to be mild and self-limiting [17].

Currently, there is no treatment or vaccine for ZIKV infection. A number of reports show high levels of infectious ZIKV excreted in urine of infected patients [18–20]. Viremia, as demonstrated by quantitative real-time reverse-transcription polymerase chain reaction (qRT-PCR), has been observed 2 days after disease onset; however, there was persistent shedding of high levels of ZIKV RNA in urine for up to 15 days after symptom onset [21, 22]. Study findings suggest that urine samples may be preferred to serum samples for detecting ZIKV RNA, because of the long duration of excretion, higher RNA levels, and less invasive specimen collection [21, 22]. Pathological effects of ZIKV infection in the urinary tract have not been reported, and the cellular targets for ZIKV replication in the renal compartment are unknown. The long-term consequences of urinary tract shedding of ZIKV require further investigation.

The current study examined ZIKV replication in cells of the glomerulus of the human kidney. I have identified in vitro primary human glomerular podocytes, renal glomerular endothelial cells (GECs), and mesangial cells as being permissive for ZIKV lytic replication. Here I examined ZIKV replication in cellular components of the glomerulus and cytokines known to be dysregulated in the blood of patients infected with ZIKV in the primary glomerular cells of the kidney. I found that podocytes, GECs, and mesangial cells are all likely to serve as reservoirs for ZIKV dissemination in the renal compartment, which probably contributes to the high-level persistent viruria observed in ZIKV-infected patients.

MATERIALS AND METHODS

Cells

Primary human renal mesangial cells and human renal GECs were obtained from ScienCell and cultivated in mesangial cell and endothelial cell media from ScienCell respectively. Mesangial cells and human renal GECs were maintained at passage level 3. Human glomerular podocytes were obtained from Moin A. Saleem [23] and were cultured as described elsewhere [24]. All cells were trypsinized and plated on uncoated 4.2 cm²-well glass chamber slides at a density of 2.5 × 10⁵ cells per well.

Virus and Virus Cultivation

The ZIKV strain PRVABC59 used in this study was originally isolated from a human serum specimen from Puerto Rico in December 2015 (nucleotide [GenBank] KU501215; ZIKV strain PRVABC59, complete genome) [25–27]. The virus was cultivated in Vero cells (Cercopithecus aethiops, kidney cell line), the infectious supernatant was filtered using a 0.22-µm filter, and the serum content was adjusted to 15%. Stock viral titers were produced by fluorescent focus assays on Vero cells using the 4G2 antibody and were adjusted to approximately 1 × 10⁵ particles per 5 µL of infectious culture supernatant. Heat-killed ZIKV was prepared by heating the viral inoculum to 65°C for 30 minutes in a water bath [28]. The mild heat inactivation used is unlikely to cause a global effect on thermolabile viral proteins. ZIKV has also been shown to be partially resistant to UV inactivation [29]. All experiments were carried out under biosafety level 2 containment, as recommended by the Meharry Medical College Institutional ReviewBoard and the Institutional Biosafety Committee. The use of ZIKV was approved by the Meharry Medical College Institutional Review Board and the Institutional Biosafety Committee.

Immunofluorescence

Immunofluorescence staining was performed as described elsewhere [30]. Briefly, chamber slide cultures containing ZIKV-infected human glomerular podocytes, renal GECs, and mesangial cells were washed twice with phosphate-buffered saline (pH 7.4), air dried, and fixed in absolute methanol for 20 minutes at −20 C. Cells were air dried for 15 minutes, hydrated in Tris-buffered saline (pH 7.6) for 5 minutes, and incubated separately for 1 hour with monoclonal antibodies from Millipore, including 4G2, von Willebrand factor, α–smooth muscle actin, and synaptopodin. All antibodies were diluted 1:50 in phosphate-buffered saline (pH 7.4). ZIKV-exposed podocytes, GECs, and mesangial cells, along with mock-infected or heat-killed ZIKV controls, were incubated for 1 hour with monoclonal antibodies to the 4G2 group antigen. Immunofluorescence staining was performed as described elsewhere [31]. Negative controls were produced using mock-infected cells stained with the 4G2 antibody and isotype-matched irrelevant primary antibodies for cell-specific biomarkers.

RT-PCR

Total RNA was extracted from ZIKV-infected podocytes, GECs, and mesangial cells along with the respective mock-infected and heat-killed ZIKV control cells using a Qiagen RNeasy Mini Kit (Qiagen). RNA was treated with DNase before elution on the column, according to the manufacturer’s recommendations. Messenger RNA (mRNA) in 0.5 μg of each sample was primed using random hexamers and reverse-transcribed with a high-capacity complementary DNA (cDNA) reverse-transcription kit (Applied Biosystems). Gene-specific primer pairs included ZIKV forward primer 5’TTYGAAGCCCTTGGATTCTT3’ and ZIKV reverse primer 5’CYCGGCCAATCAGTTCATC3’ [32] and 50 ng of cDNA for RT-PCR amplification, performed with PuReTaq Ready-To-Go PCR beads (GE Healthcare). PCR was carried out in a MJ Mini thermal cycler (Bio-Rad Laboratories) with a final volume of 25 μL.

The cycling protocol used was 95°C for 5 minutes, 55°C for 30 seconds, and 72°C for 1 minute for 36 cycles, with a final extension at 72°C for 10 minutes. PCR products were electrophoresed in 1.5% agarose, and DNA bands were visualized using ethidium bromide. Primers for glyceralaldehyde 3-phosphate dehydrogenase (GAPDH)—forward primer 5′-TGATGACATCAAGAAGGTGGTGAA-3′ and reverse primer 5′-TCCTTGGAGGCCATGTGGGC CAT-3′ (256 base pairs [bp])—were used to amplify mRNA in mock-infected and infected cells as a loading and quality control. Using ZIKV-infected cell total RNA, I amplified a 364-bp fragment with the above primers, respectively to positions 1538–1558 and 1902–1883 of the ZIKV genome sequence AY632535 [33].

qRT-PCR

Total RNA was extracted separately from ZIKV-infected podocytes, GECs, and mesangial cells, along with the respective mock-infected and heat-killed ZIKV control cells, using a Qiagen RNeasy Mini Kit (Qiagen), as described above. mRNA in 0.5 μg of each sample was primed using oligo-dT and reverse-transcribed with a high capacity cDNA reverse-transcription kit (Applied Biosystems). Real-time quantitative PCR was performed as described elsewhere, using gene-specific primers for ZIKV: forward 5′ AGGATCATAGGTGATGAAGAAAAGT 3′ and reverse 5′ CCTGACAACACTAAGATTGGTGC 3′ [33]. GAPDH primers were used for qRT-PCR, as described elsewhere [31].

Statistical Analysis

Experiments presented in this study were performed in triplicate; mock-infected, ZIKV-infected, and heat-killed ZIKV-exposed glomerular podocytes, renal GECs, and renal mesangial cells were used for RT-PCR and qRT-PCR amplification of ZIKV cDNA. Unpaired t tests were used to compare the mean values between 2 groups. Differences were considered statistically significant at P < .05. Data are presented as means with standard deviations. qRT-PCR experiments were replicated 3 times and normalized to GAPDH.

RESULTS

Glomerular Podocytes, Renal GECs, Mesangial Cells and ZIKV Infectivity

High-level persistent viruria has been observed in ZIKV-infected patients, and the mechanisms involved have not been explored [34]. I first examined podocyte cultures and found podocytes to be morphologically typical of podocytes cell lines cultured in vitro (Figure 1A, top left). These podocytes stained positive for the podocyte biomarker synaptopodin (Figure 1A, top right). Podocytes were then exposed to ZIKV at a multiplicity of infection of 0.1. At 72 hours after exposure, I observed ZIKV cytopathic effects that included rounding and sloughing of cells as well as cytolysis (Figure 1A, middle left). I also confirm podocyte infectivity with ZIKV by using immunofluorescence staining with the 4G2 antibody at 72 hours after exposure (Figure 1A, middle right). Virus-infected podocytes showed perinuclear staining with the Flavivirus 4G2 antibody (Figure 1A, middle right). Controls for immunofluorescence staining included mock-infected podocytes stained with the 4G2 antibody (Figure 1A, bottom left) and an isotype control for the synaptopodin antibody (Figure 1A, bottom right).

Figure 1.

Podocytes and Zika virus (ZIKV) infectivity. A, Phase-contrast images of mock-infected podocytes (top left), immunofluorescence staining of mock-infected glomerular podocytes stained with the synaptopodin antibody (top right), ZIKV-infected podocytes 72 hours after infection (middle left), immunofluorescence staining of ZIKV-infected podocytes with the 4G2 antibody 72 hours after infection (middle right), and mock-infected podocytes stained with the 4G2 antibody (bottom left) or the synaptopodin antibody (bottom right). All images were obtained using a Nikon TE2000S microscope mounted with a charge-coupled device (CCD) camera at ×200 magnification. For fluorescent images, 4’,6-diamidino-2-phenylindole was used to stain the nuclei blue. IFA, immunofluorescence assay. B, Reverse-transcription real-time polymerase chain reaction (RT-PCR) analysis of mock-infected podocytes, podocytes exposed to heat-killed (HK) ZIKV, and podocytes infected with wild-type (WT) ZIKV for 72 hours; bp, base pairs. C, Quantitative RT-PCR (qRT-PCR) analysis of mock-infected podocytes, podocytes exposed to HK ZIKV, and podocytes infected with WT ZIKV for 72 hours. Glyceralaldehyde 3-phosphate dehydrogenase (GAPDH) was used a loading control for RT-PCR analysis, and qRT-PCR results were normalized to GAPDH. Phase images were obtained using a Nikon TE2000S microscope mounted with a CCD camera at ×200 magnification.

ZIKV infection of glomerular podocytes was confirmed by RT-PCR using ZIKV gene-specific primers (Figure 1B). I show semiquantitative RT-PCR amplification of a 364-bp DNA fragment using ZIKV-specific primers and no amplification using cDNA from total RNA obtained from mock-infected podocytes or podocytes exposed to heat-killed ZIKV (Figure 1B). GAPDH was amplified and used as a loading control, represented as a 256-bp DNA fragment (Figure 1B). I also confirmed ZIKV infection of podocytes with qRT-PCR, which revealed a >650000-fold increase in ZIKV transcription compared with mock-infected and heat-killed control cells (Figure 1C).

I then examined human renal GECs and human renal mesangial cells and found that both were permissive for ZIKV infection. Normal GECs exhibited a characteristic cobblestonelike morphological appearance (Figure 2A, top left) and stained positive for von Willebrand factor (Figure 2A, top right). At 72 hours after ZIKV exposure, these cells showed a characteristic ZIKV cytopathological appearance by phase microscopy (Figure 2A, middle left) and showed perinuclear staining with the 4G2 antibody (Figure 2A, middle right). Controls for immunofluorescence staining included mock-infected GECs stained with the 4G2 antibody (Figure 2A, bottom left) and an isotype control for the von Willebrand factor antibody (Figure 2A, bottom right). I demonstrate semiquantitative RT-PCR amplification of a 364-bp DNA fragment using ZIKV-specific primers and no amplification using cDNA from total RNA obtained from mock-infected GECs or GECs exposed to heat-killed ZIKV (Figure 2B). GAPDH was amplified and used as a loading control, represented as a 256-bp fragment (Figure 2B). I also confirmed ZIKV infection of GECs with qRT-PCR, which revealed a >692000-fold increase in ZIKV transcription compared with controls (Figure 2C).

Figure 2.

Zika virus (ZIKV) infection and transcriptional expression in glomerular endothelial cells (GECs). A, Phase-contrast images of mock-infected confluent monolayer of normal GECs (top left), immunofluorescence staining of normal GECs with an antibody to von Willebrand factor (VWF; top right), GECs infected with ZIKV for 72 hours (middle left), immunofluorescence staining of ZIKV-infected GECs with the 4G2 antibody (middle right), mock-infected GECs stained with the 4G2 antibody (bottom left), and mock-infected GECs stained with the VWF antibody (bottom right). IFA, immunofluorescence assay. B, Real-time reverse-transcription polymerase chain reaction (RT-PCR) analysis of mock-infected GECs, GECs exposed to heat-killed (HK) ZIKV, and GECs infected with wild-type (WT) ZIKV for 72 hours; bp, base pairs. C, Quantitative RT-PCR (qRT-PCR) analysis of mock-infected GECs, GECs exposed to HK ZIKV, and GECs infected with WT ZIKV for 72 hours. Glyceralaldehyde 3-phosphate dehydrogenase (GAPDH) was used a loading control for RT-PCR analysis, and qRT-PCR results were normalized to GAPDH. Phase images were obtained using a Nikon TE2000S microscope mounted with a charge-coupled device camera at ×200 magnification.

Finally, I examined renal mesangial cells for ZIKV infectivity. Normal human mesangial cells appeared typical as described by the manufacturer (ScienCell Laboratories) (Figure 3A, top left) and stain positively for a mesangial cell biomarker, α–smooth muscle actin (Figure 3A, top right). Mesangial cells exposed to ZIKV for 72 hours showed no clear evidence of ZIKV-associated cytopathological changes (Figure 3A, middle left), but I observed high levels of virus replication, as demonstrated by perinuclear staining of ZIKV-infected mesangial cells with the 4G2 antibody (Figure 3A, middle right). No significant background staining was observed in mock infected mesangial cells stained with the 4G2 antibody (Figure 3A, bottom left), or mock infected mesangial cells stained with an α-SMA IgG isotype antibody (Figure 3A, bottom right). In addition, RT-PCR amplified the ZIKV-specific 364-bp DNA fragment in mesangial cells (Figure 3B). I also demonstrated ZIKV infectivity in mesangial cells by qRT-PCR, which revealed an increase of >328000-fold in ZIKV transcription compared with mock-infected controls and >240000-fold compared with heat-killed controls (Figure 3C).

Figure 3.

Zika virus (ZIKV) infection and transcriptional expression in mesangial cells. A, Phase-contrast images of mock-infected confluent monolayer of mesangial cells (top left), immunofluorescence staining of normal mesangial cells with an antibody to α–smooth muscle actin (α-SMA; top right), mesangial cells infected with ZIKV for 72 hours (middle left), immunofluorescence staining of ZIKV in mesangial cells with the 4G2 antibody (middle left). IFA, immunofluorescence assay. Mock infected mesangial cells stain with the 4G2 antibody (antibody control) (bottom right). Mock infected mesangial cells stained with an α-SMA isotype antibody (isotype control) (bottom left). B, Real-time reverse-transcription polymerase chain reaction (RT-PCR) analysis of mock-infected mesangial cells, mesangial cells exposed to heat-killed (HK) ZIKV, and mesangial cells infected with wild-type (WT) ZIKV for 72 hours. C, Quantitative RT-PCR (qRT-PCR) analysis of mock-infected mesangial cells, mesangial cells exposed to HK ZIKV, and mesangial cells infected with WT ZIKV for 72 hours. Glyceralaldehyde 3-phosphate dehydrogenase (GAPDH) was used a loading control for RT-PCR analysis, and qRT-PCR results were normalized to GAPDH. Phase images were obtained using a Nikon TE2000S microscope mounted with a charge-coupled device camera at ×200 magnification.

Glomerular Cells and ZIKV Infection Kinetics

To examine the level of virus produced in the 3 renal cell types I performed a 12-, 24-, and 72-hour growth yield analysis using a fluorescent focus assay for ZIKV-infected podocytes, GECs, and mesangial cells (Figure 4). Independent infections were performed in triplicate along with mock-infected controls for each time point (Figure 4). Cells positive for the 4G2 antibody were scored with fluorescent microscopy for each time point, and the results are shown in Figure 4. I observed no ZIKV-infected cells among all 3 cell types at either 12 or 24 hours after infection (Figure 4). At 72 hours after infection, I observed the highest number of ZIKV-infected cells in infected podocytes cultures, compared with ZIKV-infected GECs and mesangial cell cultures (Figure 4).

Figure 4.

Zika virus (ZIKV) replication kinetics in glomerular cells by fluorescent focus assay. The graph shows the number of ZIKV-infected 4G2-positive × 10³ (ZIKV fluorescent focus-forming units [FFU] per milliliter) for podocytes (open bars), glomerular endothelial cells (GECs; gray bars), and mesangial cells (black bars) over a time course of 12, 24, and 72 hours after infection. Fluorescein isothiocyanate–stained cells were scored with a Nikon TE2000S microscope. Error bars represent standard errors of the mean for triplicate experiments.

I observed approximately 44-fold and 14-fold increases in ZIKV (measured in focus-forming units per milliliter) in ZIKV-infected podocytes, compared with ZIKV-infected GECs and mesangial cells, respectively (Figure 5). A point longer duration (eg, ≥96 hours) with podocytes was not possible owing to the extensive cytopathological effects observed in podocytes at 96 hours after infection (data not shown). I then validated these results by performing qRT-PCR on ZIKV-infected podocytes, GECs, and mesangial cells, along with mock-infected and heat-killed control cells (Figure 5). I observed 14-fold and 7-fold increases in ZIKV mRNA expression in podocytes, compared with GECs and mesangial cells, respectively (Figure 5A, 5B, and 5C). No significant ZIKV mRNA expression was detected in mock-infected or heat-killed-infected control cells (Figure 5A, 5B, and 5C).

Figure 5.

Zika virus (ZIKV) transcriptional expression in glomerular cells. A, Expression of ZIKV messenger RNA as shown by quantitative real-time reverse-transcription polymerase chain reaction (qRT-PCR) in human podocytes. B, C, Human glomerular endothelial cells (B) and human mesangial cells (C) exposed to wild-type ZIKV along with heat-killed (HK) ZIKV and mock-infected controls. All cells were exposed to ZIKV for 72 hours. Total RNA was extracted from infected cells, followed by complementary DNA amplification and qRT-PCR. Fold expression was normalized to glyceraldehyde 3-phosphate dehydrogenase. Error bars represent standard errors of the mean for triplicate experiments. The units are comparative mRNA transcriptional units.

ZIKV Induction of Interleukin 1β, Interferon β, and RANTES in Infected Glomerular Cells

ZIKV-infected patients show high levels of proinflammatory cytokines in their serum samples during both acute and recovery phases of disease [35, 36]. No reported studies have examined cytokine profiles in the renal compartment of ZIKV-infected patients. Podocytes are most essential to renal function, and their dysregulation, injury, or functional failure initiates progressive renal disease as well as progression to end-stage renal failure [37–39]. Therefore, I initially focused on ZIKV-induced dysregulation of podocyte cytokine expression profiles in this study. I examined the levels of interleukin 1β (IL-1β), interferon (IFN) β, and RANTES (regulated on activation of normal T cells expressed and secreted), which are known to be dysregulated in patients infected with ZIKV [35, 36]. I performed qRT-PCR on podocytes infected with ZIKV along with mock-infected and heat-killed ZIKV control cells.

After 24 hours, there was no significant induction of IL-1β, IFN-β, or RANTES compared with controls (Figure 6A). Because there is no detectable virus by immunostaining at 12 or 24 hours (Figure 4) after infection in all 3 renal cell types, I then examined the 3 cell types with qRT-PCR 72 hours after infection (Figure 6B). Higher levels of induction of IL-1β, IFN-β, and RANTES were observed in ZIKV-infected podocytes than in mock-infected and heat-killed ZIKV control cells (Figure 6B). I also observed significantly higher levels of IFN-β and RANTES mRNA expression in mesangial cells infected with ZIKV than in controls, and marginal induction of RANTES, which was not significant, was observed in GECs infected with ZIKV compared with controls (Figure 6B)

Figure 6.

Zika virus (ZIKV) induction proinflammatory cytokines interleukin 1β (IL-1β), interferon (IFN) β, and RANTES (RANTES (regulated on activation of normal T cells expressed and secreted) in human glomerular cells. A, Transcriptional expression OF IL-1β, IFN-β, and RANTES in ZIKV-infected podocytes as assessed by quantitative real-time reverse-transcription polymerase chain reaction (qRT-PCR) analysis at 24 hours after infection, compared with heat-killed (HK) ZIKV and mock-infected controls. B, Transcriptional expression of IL-1β, IFN-β, and RANTES in ZIKV-infected podocytes, human glomerular endothelial cells (GECs), and human mesangial cells 72 hours after infection. Results represent cells exposed to medium only (mock; blue bars), cells exposed to HK ZIKV (black bars), and cells exposed to wild-type ZIKV (red bars). Fold expression was normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH), and error bars represent standard errors of the mean for triplicate experiments.

DISCUSSION

I have developed a hypothetical model that illustrates how ZIKV disseminates in the glomerulus of the human kidney (Figure 7). In this mechanistic model, I propose how ZIKV interacts with glomerular cells of the renal compartment shown in Figure 7, wherein ZIKV is depicted as black dots. In the model, I propose that virus enters the bloodstream via a blood meal from an infected mosquito. In the viremic phase of infection, ZIKV-containing blood enters the glomerulus via the afferent arteriole and glomerular capillaries, leading to infection of the renal corpuscle and subsequently the GECs of the kidney (Figure 7).

Figure 7.

Model of Zika virus (ZIKV) entry and existence in the glomerulus (modified with permission from Pearson Education Inc. 2013 [unpublished data]). ZIKV, depicted by black dots, enters the bloodstream via a blood meal from an infected mosquito. In the viremic phase of infection, blood enters the glomerulus via the afferent arteriole and glomerular capillaries, leading to infection of the renal corpuscle and subsequently the glomerular endothelial cells of the kidney. The virus spreads from infected glomerular endothelial cells to glomerular parenchyma. Mesangial cells and podocytes are highly exposed to infectious ZIKV. Podocytes and mesangial are highly permissive for ZIKV infection and together probably serve as amplification reservoirs in the glomerulus, resulting in high-level persistent viruria, as measured by means of real-time polymerase chain reaction for ZIKV RNA.

The virus spreads from infected GECs to the glomerular parenchyma (Figure 7). Mesangial cells and podocytes are highly exposed to infectious ZIKV. Podocytes and mesangial cells are also highly permissive for ZIKV infection. Prolonged viral shedding ensues via the voided urine of infected patients (Figure 7). These findings suggests that cellular components of the glomerulus together probably serve as amplification reservoirs in the kidney, resulting in high-level persistent viruria, as measured by RT-PCR for ZIKV RNA. Findings of the current study suggest that podocytes have the greatest ZIKV infection burden in the glomerulus, compared with mesangial cells and GECs (Figures 4, 5, and 6). The current findings also show that cytokines (IL-1β, IFN-β, and RANTES) are induced at higher levels in podocytes, which correlates with the increased viral replication observed in podocytes (Figures 4 and 5).

In 2016, several reports were published on mouse models showing ZIKV infection in the kidney. In a 129/Sv mouse model, deficient in both IFN-α/β and IFN-γ receptors (AG129), Zmurko et al [40] showed high levels of ZIKV RNA in spleen, liver, and kidneys, in comparison with control animals. In a study by Aliota et al [41], ZIKV infection of AG129 mice resulted in high tissue viral loads in multiple organs, including the kidneys, liver and spleen. In a mouse model for ZIKV infection of dexamethasone-immunosuppressed animals, Chan et al [42] demonstrated ZIKV infection of the kidney, with acute tubulitis and inflammatory exudation in tubular lumens and a moderate degree of interstitial inflammation. Additional studies in immunosuppressed animals, by Lazear et al [43] and Rossi et al [44], showed higher ZIKV viral loads in tissues of the spleen, liver, kidney, serum, testes, brain, and spinal cord, in comparison with control animals.

Among other flaviruses associated with renal disease in humans, dengue virus infection (DVI) is known to cause renal disease in infected patients. DVI can result in acute kidney injury, glomerulonephritis, and nephrotic-range proteinuria. Increased morbidity and mortality rates associated with DVI-induced renal disease are more likely to occur in patients who are older and obese and have other preexisting infections [45]. Possible mechanisms for dengue virus induction of renal disease include direct infection of glomerular parenchymal cells and the production of virus-induced antigen antibody complexes in the glomerulus [45]. Severe DVI can result in acute renal failure in 2%–5% of cases, with high mortality rates [46]. Acute proteinuria can be observed in as many as 74% of individuals with DVI. Acute kidney injury due to DVI is underreported and is often documented only on clinical presentation of microscopic hematuria, proteinuria, or electrolyte imbalance [46]. The association of West Nile virus infection with renal disease is controversial. The presence of persistent West Nile virus shedding in the urine of infected individuals has been observed in small cohorts, and the impact on renal disease is unknown [47].

The mechanisms associated with high-level persistent viruria observed in ZIKV-infected patients have not been explored until now. The current report provides information that identifies human glomerular podocytes, renal GECs, and mesangial cells as important target cells in the renal compartment for ZIKV infection and replication. The findings of this study suggest that there is a high viral burden in the glomerulus of the kidney that directly correlates with long-term persistent viral shedding in urine. The ZIKV-associated cytopathological changes in glomerular cells are likely to be controlled by immune surveillance, because patients recover and viruria subsides over time. There are probably host immune factors that limit viral infection and dissemination in the glomerulus, resulting in self-limiting disease that remains subclinical throughout the disease course. However, the long-term effects of ZIKV replication in the glomerulus of adults and infants, as well as in immunocompromised patients, are unknown.

The next steps for this study will be to directly examine renal biopsy specimens or archived renal tissue from patients with both acute and reconvalescent ZIKV disease to determine viral dissemination patterns and the possible long-term persistence of viral genomes within the glomerulus. In addition, this study is insufficient to draw correlations regarding ZIKV detection in urine and the duration of detection. An animal model system for ZIKV infection in the renal compartment is a logical next step for these investigations.

The risk of ZIKV disease in solid organ transplant (SOT) recipients has not been fully examined. Disease severity associated with ZIKV infection in the context of immunosuppression for allograft maintenance is unknown and the impact of ZIKV disease on allograft function has not been reported. Nogueira et al [48] reported the first small case series of ZIKV infection in SOT recipients positive for ZIKV by RT-PCR and negative for other arboviruses, including dengue virus and chikungunya virus. ZIKV infection in both liver and renal transplant recipients resulted in clinical complications, most notably bacterial infections [48]. None of the ZIKV-infected patients presented with a rash, conjunctivitis, or neurological symptoms but 3 of 4 were anemic, and all 4 had thrombocytopenia [48] Patients presented at admission with fever, myalgia, and adynamia along with signs of acute liver or renal damage [48]. This was a small sample of transplant recipients infected with ZIKV, and the findings are not sufficient to clarify the impact of ZIKV infection in SOT recipients, which will require larger case control studies and further investigation.

I am also aware that acute renal disease has not been reported in ZIKV-infected patients and that although I observe full lytic replication of ZIKV in primary human glomerular cells in vitro, there have, to our knowledge, been no reports of renal disease directly associated with ZIKV infection in nonimmunosuppressed patients. There is probably an underlying host immune defense mechanism in vivo that limits ZIKV-associated disease in the glomerulus. These proposed host factors that would limit ZIKV pathogenesis in the absence of immunosuppression could be important for the development of antiviral strategies to prevent or limit ZIKV disease. The incidence of such disease in SOT recipients will probably increase over time and may require widespread recommendations for mandatory screening of allograft donors and recipients from ZIKV-endemic regions.

Notes

Acknowledgments. Thanks to Agnes Fogo, MD for reading this manuscript and to Waldemar Popik, PhD and Atanu Khatua, PhD for technical assistance with qRT-PCR.

Disclaimer. The funders did not participate in the design, preparation, data analysis, or decision to publish the manuscript.

Financial support. This work was supported by a Zika virus research startup award from James E. K. Hildreth, PhD, MD, president of Meharry Medical College (D. J. A.) and the National Center For Advancing Translational Sciences of the National Institutes of Health (grant 3UH2TR000491-02S1 to D. J. A.).

Potential conflicts of interest. Author certifies no potential conflicts of interest. The author has submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.