Glucocorticoids Do Not Enhance BK Polyomavirus Replication in Human Renal Proximal Tubular Epithelial Cells In Vitro

ABSTRACT

BK polyomavirus (BKPyV) infection may lead to BKPyV‐associated nephropathy (BKPyVAN) and subsequent graft loss in kidney transplant recipients. Experimental studies have suggested that glucocorticoids (GCs) may enhance BKPyV replication through a steroid‐responsive element within the viral genome. However, clinical observations have not consistently supported this association, prompting re‐investigation of the effect of GCs on BKPyV replication in primary human renal epithelial cells (PTECs). PTECs were infected with BKPyV and subsequently exposed to increasing concentrations of methylprednisolone (MPS) or dexamethasone (Dexa). Viral replication was quantified by quantitative polymerase chain reaction (qPCR). Relative differences in viral load were analyzed using univariate analysis of variance. Activation of the GC pathway was assessed by measuring FKBP5 and PER1 gene expression. Inflammatory responses were evaluated by enzyme‐linked immunosorbent assays (ELISA) for IL‐6, IL‐8, and TNF‐α. Treatment of BKPyV‐infected PTECs with increasing concentrations of MPS or Dexa did not affect viral replication compared to untreated controls. While treatment‐related changes in gene expression of FKBP5 and PER1 were observed, confirming GC‐signaling pathway activation in BKPyV‐infected cells. Cytokine production (IL‐6 and IL‐8) was unaffected. In primary human PTECs infected with BKPyV, treatment with MPS or Dexa did not enhance BKPyV replication. These findings contrast with earlier in vitro studies and suggest that glucocorticoids do not directly promote BKPyV replication under the conditions tested. Further studies are needed to determine whether these findings extend to other cell models and in vivo.

1. Introduction

The BK polyomavirus (BKPyV) is of significant concern in kidney transplant recipients. It can cause BK polyomavirus associated nephropathy (BKPyVAN), which can lead to graft damage and even graft loss. Due to its clinical impact, routine screening for BKPyV DNAemia, a precursor for BKPyVAN, is standard care after kidney transplantation [1]. Upon detection, treatment mainly consists of reduction of immunosuppression, as there are currently no approved treatments [1, 2]. As a consequence, the risk of developing de novo donor specific antibodies (dnDSA) and acute rejection increases [3].

Interestingly, several reports have described patients with concurrent BKPyVAN and acute rejection [4, 5]. In these cases, treatment with anti‐rejection therapy—most notably high‐dose methylprednisolone (MPS)—was associated with stabilization or even a decline in BKPyV DNAemia levels. These findings are in apparent contrast with an in vitro study by Moens et al., which raised concerns about the use of glucocorticoids (GCs) in BKPyV infection due to the presence of GC‐responsive elements in the BKPyV genome, potentially enhancing viral replication [6]. However, this study is limited by the use of nonrenal and nonhuman cell lines, rather than the primary target cells of BKPyV infection, namely human proximal tubular epithelial cells (PTECs) of the kidney [7, 8].

To address this discrepancy, we designed an in vitro model using PTECs infected with BKPyV, and assessed the effects of MPS and dexamethasone (Dexa) on viral replication. This study aimed to clarify whether exposure to GCs directly enhances, inhibits, or has no effect on BKPyV replication in its primary renal host cell type.

2. Methods

2.1. Primary Human Tubular Epithelial Cell Culture

PTECs derived from a single donor (donor 43; Biopredic) were cultured in renal epithelial basal medium (REBM; Lonza, CC‐3191) supplemented with REGM SingleQuots (Lonza, CC‐4127) and 10% fetal bovine serum (FBS; Serana, S0FBS‐CO‐015), hereafter referred to as complete REGM. All mediums were prepared by charcoal treatment to deplete endogenous steroid components. Tissue culture flasks and 12‐well plates were precoated according to standard procedures. After thawing (passage 3 or 4), cells were seeded into coated flasks and cultured at 37°C in 5% CO₂. Cells were passaged once more before experimental use, and only cells up to passage 6 were included in experiments. Culture medium was refreshed every 2–4 days. The infection experiments were performed three times independently.

2.2. Passaging and Seeding

Cells were passaged using TrypLE Express (Thermo Fisher Scientific). Detached cells were neutralized with PBS containing 10% FBS, pelleted by centrifugation, and resuspended in complete REGM. Cell number was determined using an automated cell counter. For experiments, cells were seeded in 12‐well plates at a density (200.000 cells per well) sufficient to reach optimal confluence within 24 h. Plates were incubated overnight prior to experimental procedures.

2.3. BK Polyomavirus Infection

PTECs (passage 6) were exposed to BK polyomavirus (BKPyV; ATCC VR‐837, American Type Culture Collection) diluted 1:3000 in REGM for 2 h at 37°C. Virus‐containing medium was then removed, cells were washed twice with washing medium (complete REGM), and fresh complete REGM was added.

2.4. Drug Treatment

Three days postinfection, cells were treated for 96 h with either brincidofovir (1 μM) as a positive control for inhibition of viral replication [9], methylprednisolone (0.1–100 nM), or dexamethasone (0.1–100 nM). Concentrations were selected based on clinically relevant plasma levels and previous in vitro studies. Vehicle‐treated BKPyV‐infected cells served as controls. Each condition was tested in duplicate.

2.5. Harvesting and DNA Extraction

At the end of the treatment period, both supernatant and cell‐associated fractions were collected and combined. Samples were subjected to three freeze–thaw cycles and stored in liquid nitrogen. DNA was extracted from lysates for quantitative PCR (qPCR) analysis.

2.6. qPCR Detection of BKPyV

DNA from supernatant and cell lysates was analyzed using qPCR. The BKPyV assay targeted a 90 bp fragment within the VP1 gene using primers 440BKVs (5′‐GAAAAGGAGAGTGTCCAGGG‐3′) and 441BKVas (5′‐GAACTTCTACTCCTCCTTTTATTAGT‐3′) with probe 576BKV‐TQ‐FAM (5′‐CCAAAAAGCCAAAGGAACCC‐3′‐BHQ1), as previously described [10]. Prior to each qPCR reaction, samples were diluted 1:1000 to ensure accurate quantification within the linear range of the assay. BKPyV qPCR included a negative and positive control to validate assay performance.

2.7. Gene Expression Analysis

To assess activation of the GC signaling pathway, the expression of the GC‐responsive genes FKBP5 and PER1 was quantified by qPCR. Total RNA was isolated from treated and control conditions using TriPure Isolation Reagent, according to the manufacturer's instructions. qPCR was performed using gene‐specific primers for FKBP5 (forward: 5′‐TGGAAAGAAGTTTGATTCCAGTCAT‐3′; reverse: 5′‐CATGGTAGCCACCCCAATGT‐3″), PER1 (forward: 5′‐CACTCCTGCGACCAGGTA‐3′; reverse: 5′‐TAGGGGGCCACTCATGTCT‐3′), and the housekeeping gene GAPDH (forward: 5′‐CCACATCGCTCAGACACCAT‐3′; reverse: 5′‐GCAACAATATCCACTTTACCAGAGTTAA‐3′). All samples were analyzed in duplicate.

Relative gene expression levels were calculated using the comparative ΔΔCq method. Briefly, Cq values of FKBP5 and PER1 were normalized to GAPDH (ΔCq), and relative expression levels were calculated compared with the untreated control condition (ΔΔCq). Fold changes in gene expression were expressed as 2^−ΔΔCq. Fold changes were used to evaluate GC pathway activation across treatment conditions.

2.8. Cytokine Measurement by ELISA

To evaluate the effects of GC treatment on inflammatory signaling in BKPyV‐infected PTECs, the concentrations of tumor necrosis factor–α (TNF‐α), interleukin‐6 (IL‐6), and interleukin‐8 (IL‐8) were quantified in cell culture supernatants and cell lysates using commercially available enzyme‐linked immunosorbent assay (ELISA) kits (BD OptEIA, BD Biosciences, San Diego, CA, USA), according to the manufacturers' instructions. Prior to analysis, samples were serially diluted in assay buffer to ensure that optical density (OD) values fell within the dynamic range of the standard curves. OD values were converted to cytokine concentrations (ng/mL) by interpolation from the corresponding standard curves.

2.9. Statistical Analysis

qPCR results were log₁₀‐transformed to approximate a normal distribution, as assessed by visual inspection of histograms and Q–Q plots. For each experimental condition, mean and standard deviations in log₁₀ viral loads (copies/mL) were calculated. To evaluate differences in viral replication across treatment conditions, a univariate analysis of variance (ANOVA) was performed using treatment condition as the fixed factor and experiments as a random factor to account for between‐experiment variability. Each condition was compared against the untreated control, which was set as the reference category. p‐Values below 0.05 were considered statistically significant. Statistical analysis was performed using SPSS (version 25).

Gene expression analyses were performed to confirm GC pathway activation and are presented descriptively. Cytokine values were log₁₀‐transformed to approximate a normal distribution, and differences between experiments were assessed using one‐way ANOVA.

3. Results

3.1. BKPyV Replication in PTECs and Effect of Glucocorticoid Exposure

To investigate the effect of GCs on BKPyV replication in its natural host cell, we established an in vitro infection model using primary human PTECs. In the untreated conditions, this model consistently yielded high levels of BKPyV replication, with mean log₁₀ copies/ml values of 5.9, 4.7, and 7.0 in three consecutive experiments respectively (Table 1 and Figure 1). These findings indicate a robust and reproducible infection model.

TABLE 1.

Effect of methylprednisolone and dexamethasone on BKPyV replication.

Conditions	Experiment 1	Experiment 2	Experiment 3	Univariate ANOVA analysis log10 copies/mL	p
	Mean qCPR (log10 copies/mL ±SD)	Mean qCPR (log10 copies/mL ±SD)	Mean qCPR (log10 copies/mL ±SD)
Control	5.9 (0.3)	4.7 (0.3)	7.0 (0.2)	Reference
MPS 0.1 nM	6.5 (0.2)	4.5 (0.1)	6.6 (0.0)	−0.02 (−0.4 to 0.4)	0.928
MPS 1 nM	5.6 (0.3)	4.7 (0.2)	6.6 (0.1)	−0.2 (−0.6 to 0.1)	0.212
MPS 10 nM	5.8 (0.0)	4.6 (0.0)	6.6 (0.2)	−0.2 (−0.6 to 0.2)	0.326
MPS 100 nM	5.3 (0.2)	4.5 (0.1)	6.8 (0.4)	−0.3 (−0.7 to 0.1)	0.093
Dexa 0.1 nM	5.8 (0.0)	4.3 (0.1)	6.4 (0.1)	−0.4 (−0.8 to −0.01)	0.043
Dexa 1 nM	5.7 (0.0)	4.4 (0.0)	6.5 (0.6)	−0.3 (−0.7 to 0.04)	0.077
Dexa 10 nM	5.6 (0.1)	4.3 (0.0)	7.0 (0.9)	−0.3 (−0.6 to 0.1)	0.182
Dexa 100 nM	5.8 (0.9)	4.7 (0.4)	6.6 (0.3)	−0.2 (−0.6 to 0.2)	0.326
Brincidofovir 1 μM	4.7 (0.1)	3.6 (0.0)	5.8 (0.2)	−1.2 (−1.6 to −0.8)	< 0.001

Note: Each condition was tested in duplicate across three independent experiments. For each condition, the mean log₁₀ BKPyV qPCR copies/mL ± standard deviation (SD) is shown per experiment. Statistical comparison was performed using univariate ANOVA with condition as fixed factor and experiment as random factor. The untreated control group (no medication) was used as the reference condition. Statistically significant effects (p < 0.05) are indicated in bold.

Abbreviations: Dexa, dexamethasone; MPS, methylprednisolone.

FIGURE 1.

BKPyV replication in infected PTECs upon exposure to GCs and brincidofovir. Mean log10 BKPyV qPCR copies/ml across three independent experiments for each condition. Error bars represent the 95% confidence interval. (*) indicate statistically significant differences compared with the control condition. Abbreviations: MPS. Methylprednisolone, Dexa. Dexamethasone, BKPyV. BK polyomavirus.

To validate the responsiveness of the model to antiviral intervention, cells were treated with brincidofovir (1 μM), which resulted in a strong and statistically significant reduction in BKPyV replication (−1.2 log₁₀ copies/mL; p < 0.001), confirming that viral replication could be effectively inhibited under these experimental conditions.

Subsequently, the effect of GC exposure on BKPyV replication was assessed. Across all conditions, treatment with MPS or Dexa resulted in BKPyV qPCR values that were comparable to or lower than those observed in untreated controls (Table 1 and Figure 1).

In the univariate ANOVA analysis across three independent experiments, a modest but statistically significant reduction in BKPyV replication was observed for Dexa 0.1 nM (mean difference: −0.4 log₁₀ copies/mL; 95% CI: −0.8 to −0.01; p = 0.043), while no significant effects were observed for other GC concentrations. Accordingly, no dose–response relationship could be established for either MPS or Dexa. Overall, GC exposure did not result in increased BKPyV replication in this model (Table 1 and Figure 1).

3.2. GC Receptor Activation

To confirm that GCs exerted their expected biological effects in this model, we assessed the expression of GC‐responsive genes. As shown in Figure 2, exposure to both MPS and Dexa resulted in increased expression of FKBP5 and PER1, genes that are induced upon GC receptor activation, when compared with untreated controls. This effect was dose‐dependent, with higher GC concentrations associated with greater fold induction.

FIGURE 2.

Relative expression of FKBP5 and PER1 in BKPyV infected PTECs following treatment with dexamethasone and methylprednisolone at different doses (A) Shows FKBP5 mRNA expression and (B) shows PER1 mRNA expression, quantified by qPCR and normalized to GAPDH expression. Relative expression levels were calculated using the 2^−ΔΔCt method and expressed as fold change compared with untreated controls. Data represent mean ± SD. Dexa, Dexamethasone; MPS, Methylprednisolone.

3.3. Cytokine Production

To further assess the functional effects of GCs on the inflammatory response in PTECs, we measured the production of proinflammatory cytokines. TNF‐α concentrations were below the lower limit of detection in the majority of samples and were therefore not included in quantitative analyses.

For IL‐6 and IL‐8, no statistically significant differences were observed across the different experimental conditions, with p = 0.400 and p = 0.624, respectively (Tables 2a and 2b; Figure 3b). These findings indicate that, in contrast to the clear transcriptional activation of GC‐responsive genes, GC exposure did not result in a measurable suppression of IL‐6 or IL‐8 production in this model.

TABLE 2a.

IL‐6 concentrations in BKPyV‐infected PTECs under different experimental conditions.

Experimental conditions	Average ± SD IL‐6 (ng/mL)
Controle	20.3 (10.4)
Brincidofovir	24.3 (9.7)
MPS 0.1 nM	14.0 (4.5)
MPS 1 nM	17.9 (14.8)
MPS 10 nM	11.7 (7.8)
MPS 100 nM	9.8 (3.6)
Dexa 0.1 nM	23.3 (7.6)
Dexa 1 nM	11.9 (3.4)
Dexa 10 nM	16.0 (7.5)
Dexa 100 nM	9.4 (4.0)

Note: Mean IL‐6 concentration across experimental conditions. Data represents the mean ± SD of the three experiments preformed in duplicate.

Abbreviations: Dexa, dexamethasone; MPS, methylprednisolone; SD, standard deviation.

TABLE 2b.

IL‐8 concentrations in BKPyV‐infected PTECs under different experimental conditions.

Experimental conditions	Average ± SD IL‐8 (ng/mL)
Controle	9.6 (5.9)
Brincidofovir	17.9 (8.1)
MPS 0.1 nM	6.5 (2.9)
MPS 1 nM	7.2 (4.8)
MPS 10 nM	5.3 (4.6)
MPS 100 nM	6.6 (4.2)
Dexa 0.1 nM	6.7 (6.0)
Dexa 1 nM	4.9 (3.2)
Dexa 10 nM	6.6 (5.0)
Dexa 100 nM	7.0 (4.4)

Note: Mean IL‐8 concentration across experimental conditions. Data represents the mean ± SD of the three experiments preformed in duplicate.

Abbreviations: Dexa, dexamethasone; MPS, methylprednisolone; SD, standard deviation.

FIGURE 3.

(a) IL‐6 concentration in BKPyV‐infected PTECs under the experimental conditions. (b) IL‐8 concentration in BKPyV‐infected PTECs under the experimental conditions.

4. Discussion

In the present study, we observed that exposure of PTECs to MPS and Dexa after BKPyV infection did not lead to increased viral replication. These findings suggest that, contrary to prior studies, GC exposure did not enhance BKPyV replication in primary human PTECs under the experimental conditions tested.

In our experiments, 100 nM Dexa led to a nonsignificant decline in BKPyV replication, and with 0.1 nM Dexa even significantly reduced viral replication. This is in contrast to the 11‐fold increase in viral yield reported in a previous study with ~125 nM Dexa in Vero cells [6]. Moens et al. used multiple treatment strategies (preinfection, postinfection, pre‐ and postinfection, and continuous exposure), with the highest virus yield observed in the postinfection arm, most comparable to our approach. However, there was a difference in timing; Moens et al. initiated treatment 12 h postinfection and harvested the cells 4 days later, whereas in our experiments, treatment began at 72 h postinfection, whereupon the cells were harvested 3 days later. Early exposure could potentially affect early transcriptional events or host responses that shape subsequent viral yield. However, whether this difference in experimental setup accounts for the observed discrepancies remains uncertain. It is, for example, conceivable that the choice of cell type contributed substantially to the observed differences. Primary human PTECs likely preserve antiviral responses and intracellular signaling pathways that more closely reflect the human in vivo environment during BKPyV infection, in contrast to Vero cells used in the experiments of Moens et al., which are derived from African green monkey kidney epithelium and may exhibit altered steroid responsiveness or viral permissiveness. Future experiments directly comparing Vero cells and primary human PTECs under otherwise identical experimental conditions may help clarify whether the discrepant findings are primarily related to differences in cellular context. Such studies could also further assess whether the effects of glucocorticoids depend on the timing of viral infection and glucocorticoid exposure.

Importantly, despite confirmed activation of the GC pathway, indicated by upregulation of GC‐responsive genes (PER 1 and FKBP5), BKPyV replication remained unchanged. A possible explanation is that PTECs may have intrinsic resistance to steroid‐mediated viral activation. Previous work by de Haij et al. demonstrated that cytokine production by primary human PTECs is largely insensitive to Dexa [11]. In line with these findings, we did not observe significant changes in IL‐6 and IL‐8 concentrations following GC exposure. This suggests that while glucocorticoid receptor activation is present, the subsequent suppressive effect on the NF‐κB pathway does not seem to follow. This decoupling could explain why BKPyV replication did not increase in our experiments. In addition, PTECs are reported to mount relatively limited innate antiviral responses to BKPyV, including Type I interferon responses. Thus, the limited engagement of these antiviral pathways in PTECs may further reduce the potential for glucocorticoids to modulate BKPyV replication through suppression of innate antiviral signaling [12]. Furthermore, primary PTECs are highly permissive to BKPyV infection, which may have limited our ability to detect an additional stimulatory effect of glucocorticoids on viral replication [13]. However, viral loads in untreated conditions varied between 4.7 and 7.0 log₁₀ copies/mL across experiments, indicating that the assay was not saturated and that a further increase in viral replication could in principle have been detected.

Another possible explanation for the observed stable or slightly reduced BKPyV replication may lie in the suppression of synthesis of TNF‐α by corticosteroids. Yi‐Lung et al. demonstrated that TNF‐α stimulates BKPyV replication [14]. As GCs reduce TNF‐α synthesis in many cell types, this anti‐inflammatory effect could paradoxically reduce viral replication in certain settings. However, TNF‐α remained largely undetectable in our experiment. This is in line with prior observations that nonactivated primary human PTECs produce relatively low levels of this cytokine [15]. Thus, our findings do not allow us to determine whether TNF‐α contributes to the effects of GCs on BKPyV replication. Future experiments modulating TNF‐α signaling pathways could provide further insight into the effect of GCs on BKPyV replication.

Our findings prompt a reevaluation of the presumed stimulatory effect of GCs on BKPyV. Prior work from our group has shown that patients with rejection following BKPyV DNAemia often receive less rejection treatment than protocoled, presumably due to fear of viral reactivation. Yet, in this group, rejection remains the main cause of graft loss [16]. Because GCs do not seem to exacerbate BKPyV replication, this clinical hesitancy may be unfounded and warrants reconsideration.

Several limitations should be acknowledged. First, inter‐experiment variability in BKPyV yield was observed across the three replicate studies. Second, as an in vitro model, our system does not capture the complex interplay between GCs, host immunity, and viral dynamics seen in vivo. In particular, the absence of immune cells precludes assessment of indirect effects of GCs on BKPyV replication mediated through altered cytokine production or immune‐cell signaling. Therefore, our findings demonstrate that GCs do not directly enhance BKPyV replication in primary human PTECs under the experimental conditions tested, but cannot exclude indirect effects that may occur in vivo. However, this controlled experimental model also allowed us to isolate the direct effect of GCs on BKPyV replication in primary human PTECs, without confounding effects from other cell types or immune‐mediated signaling. Third, all experiments were performed with PTECs from a single donor. Although the findings were consistent across three independent infection experiments, interindividual variation in glucocorticoid receptor expression or responsiveness cannot be excluded. Lastly, the translational applicability of our findings to clinical decision‐making must be approached with caution and requires further investigation.

In this in vitro model, treatment with MPS or Dexa after BKPyV infection did not enhance viral replication. These findings contrast with earlier studies and suggest that GCs do not directly promote BKPyV replication in primary human PTECs. Further investigations are needed to determine whether these findings extend to other cell models and in vivo settings.

Author Contributions

W.T.M., R.A.L., J.P., A.P.J.V., E.W., C.K., M.C.W.F., and J.I.R. wrote the manuscript; W.T.M., J.P., A.P.J.V., C.K., M.C.W.F., and J.I.R. designed the research; W.T.M., R.A.L., and E.W. performed the research; W.T.M., R.A.L., J.P., C.K., M.C.W.F., and J.I.R. analyzed the data. All authors reviewed and approved the final manuscript.

Funding

Wouter T. Moest, Mariet C.W. Feltkamp, and Joris I. Rotmans are supported by The TTVguideIT project, which is funded by the European Union's Horizon 2020 research and innovation program under grant agreement 896932.

Disclosure

No AI tools were used for data analysis, figure generation, or interpretation of results.

Conflicts of Interest

The authors declare no conflicts of interest.