|Home | About | Journals | Submit | Contact Us | Français|
The human polyomavirus BK virus (BKV) is a common virus for which 80 to 90% of the adult population is seropositive. BKV reactivation in immunosuppressed patients or renal transplant patients is the primary cause of polyomavirus-associated nephropathy (PVN). Using the Dunlop strain of BKV, we found that nuclear factor of activated T cells (NFAT) plays an important regulatory role in BKV infection. Luciferase reporter assays and chromatin immunoprecipitation assays demonstrated that NFAT4 bound to the viral promoter and regulated viral transcription and infection. The mutational analysis of the NFAT binding sites demonstrated complex functional interactions between NFAT, c-fos, c-jun, and the p65 subunit of NF-κB that together influence promoter activity and viral growth. These data indicate that NFAT is required for BKV infection and is involved in a complex regulatory network that both positively and negatively influences promoter activity and viral infection.
Polyomavirus-associated nephropathy (PVN) is prevalent in 5 to 10% of kidney transplants, resulting in a 50% or greater incidence of allograft loss or dysfunction in immunosuppressed renal transplant recipients (19, 40, 42). PVN is associated with the human polyomavirus BK virus (BKV). BKV is a member of the Polyomavirdae family, which also includes JC virus (JCV) and simian virus 40 (SV40). Polyomaviruses are characterized by small (40 to 50 nm in diameter), nonenveloped viruses containing circular double-stranded DNA (18, 23). The mode of transmission is unknown, but initial infection takes place in early childhood and is restricted to the epithelium of renal tubules, ureter, and urinary bladder (16, 50). BKV initiates an asymptomatic, lifelong persistent infection in approximately 80 to 90% of healthy human adults (6, 27, 48). The reactivation of BKV occurs during immunosuppression following kidney or bone marrow transplantation (17, 24). Transplant recipients receiving more intensive immunosuppressive regimens are at a higher risk of viral reactivation (3). This can permit high levels of viral replication, resulting in renal damage and ultimately graft failure (18, 20, 40). Currently, the treatment of PVN includes a reduction in immunosuppressive therapy in combination with the antiviral agents leflunomide and cidofovir (11, 26, 45, 54).
Like those of other members of the polyomavirus family, the BKV life cycle is highly regulated. The virus binds to cells via an N-linked glycoprotein with α(2,3)-linked sialic acid and gangliosides GT1b and GD1b (29) and is internalized via caveola-mediated endocytosis (8, 10). The genome is comprised of three functional regions. The noncoding control region (NCCR) mediates viral replication and transcription while dividing the genome into the early gene products small and large T antigens and the late gene products viral capsid proteins VP1, VP2, and VP3 (47). In the archetype strain (WW), the NCCR is a highly conserved bidirectional promoter composed of five sequence blocks labeled in alphabetical order as O, P, Q, R, and S. The O block contains a palindrome of two inverted repeat sequences, a 20-bp A/T region, and the start codon for early genes, P, Q, and R are three transcription factor binding blocks, followed by the S block, which contains the start codon for agnoprotein (38). Any deletions, duplications, or rearrangements in the NCCR are classified as rearranged forms compared to the archetype (39). The occurrence of rearrangements is not well understood, and they are not considered unique strains but rather adaptations in variable cellular environments and tissue culture systems (1). The rearranged NCCR of the BKV Dunlop strain is composed of triplicate P blocks (NCCR structure, O-P-P1-7;26-68-P1-64-S) where Q and R blocks have been deleted (7, 51).
Previous research in our laboratory identified nuclear factor of activated T-cells (NFAT) as an important transcription factor in the regulation of both JCV and SV40 infection (34, 35). The NFAT family of transcription factors is composed of five proteins, of which NFAT1 (NFATc2), NFAT 2 (NFATc1), NFAT 3 (NFATc4), and NFAT4 (NFATc3) are regulated by calcium (5, 22, 30, 31). The N terminus of the protein is highly conserved, containing the NFAT homology region and the DNA binding domain (21, 30, 44). These proteins are expressed in various immune cells and have been detected in a variety of tissues, such as kidney, thymus, and spleen (21, 37). NFAT is phosphorylated in resting cells; dephosphorylation occurs by the calcium activation of a calmodulin-dependent serine phosphatase calcineurin, which exposes the nuclear localization signal (31, 44, 46, 49). NFAT translocates to the nucleus, where it regulates the expression of various genes (25, 46). During T-cell activation, NFAT promotes gene transcription by syngerstic interactions with other transcription factors, such as activator protein 1 (AP-1).
In this study, we examined the role of NFAT in BKV infection and transcription. Luciferase reporter assays and chromatin immunoprecipitation assays (ChIP) revealed that NFAT4 binds to the Dunlop promoter and regulates promoter activity in Vero cells. Mutational analysis demonstrates that all three NFAT sites are required for transcriptional activity. However, mutations of NFAT site 1 or NFAT site 2 increased early promoter activity and viral propagation. These two sites can both activate and repress transcription, depending on whether the AP1 or NF-κB site also is bound by fos/jun or p65, respectively.
Vero cells were maintained in minimum essential medium (E-MEM: Mediatech Inc., Herdon, VA) containing 5% heat-inactivated fetal bovine serum (Mediatech Inc.) at 37°C in a 5% CO2 incubator. The BKV Dunlop strain used in these experiments was purchased from the American Type Culture Collection (ATCC; Manassas, VA). Cell-permeable NFAT inhibitor peptide (11R-VIVIT) was purchased from Calbiochem (San Diego, CA). The NFAT inhibitor peptide sequence was HRRRRRRRRRRRGGGMAGPHPVIVITGPHEEOH (41).
Cells were grown on coverslips to 70% confluence and infected with BKV at a multiplicity of infection (MOI) of 5 in the presence of 2% fetal bovine serum for 1 h at 37°C. Seventy-two hours postinfection, cells were fixed with 2% paraformaldehyde for 20 min at room temperature and permeabilized for 20 min in 0.5% Triton X-100. Cells were stained with monoclonal antibody PAB 416 (Ab-2) to detect BKV large T-antigen (T-Ag) (Oncogene Research Products, Cambridge, MA). Late viral protein VP-1 was detected with monoclonal antibody PAB 597 made against SV40, which cross-reacts with BKV and JCV. Secondary antibody Alexa Fluor 488-labeled goat anti-mouse antibody was purchased from Invitrogen (Carlsbad, CA). Cells were mounted on slides using Vectashield with 4′,6′-diamidino-2-phenylindole (DAPI) (Vector Labs, Burlingame, CA) and observed using a Nikon epifluorescence microscope (Eclipse E800; Nikon Inc.).
pluc2/DunlopE and pluc2/DunlopL were made by the PCR amplification of the promoter region of the pBR322 Dunlop construct (ATCC) with BglII and XhoI sites at either end of the primers to insert the promoter into the pGL4.10 luciferase construct (Promega, Madison, WI) in either the early or late orientation. The phMGFP construct expresses only green fluorescent protein (GFP) driven by the cytomegalovirus (CMV) promoter and was purchased from Promega. The pGFP-VIVIT construct contains the NFAT inhibitor peptide sequence (MAGPHPVIVITGPHEE) subcloned into pEGFP at the N terminus (2). The pRL-TK plasmid was purchased from Promega and expresses the Renilla luciferase from the thymidine kinase promoter of herpes simplex virus. The pluc2/Dunlop reporter constructs and BKV/puc19 plasmid were mutated using a Stratagene QuikChange mutagenesis kit. NFAT sequences in the viral promoter were mutated from GGAAA to TAGAT (28). To confirm mutations, samples were sent to Genewiz, South Plainfield, NJ, for sequencing with RVprimer3 (Promega). The primers used to mutate BKV/puc19 are the following: NFAT site 1 forward, 5′-CCATGACCTCAGGAATAGATGTGCATGACTCACAGGGGAATGCAG-3′; reverse, 5′-GCTGCATTCCCCTGTGAGTCATGCACATCTATTCCTGAGGTCATGG-3′; NFAT site 2 forward, 5′-CCATGACCTCAGGAATAGATGTGCATGACTCACAGGGAGGAGCTGC-3′; reverse, 5′-GCAGCTCCTCCCTGTGAGTCATGCACATCTATTCCTGAGGTCATGG-3′; and NFAT site 3 forward, 5′-CCAAACCATGACCTCAGGAATAGATGTGCATGACAGACATGTTTTGC-3′; reverse, 5′-GCAAAACATGTCTGTCATGCACATCTATTCCTGAGGTCATGGTTTGG-3′ (nucleotides in bold denote NFAT binding site mutations).
To assess NFAT4 binding during infection, Vero cells were grown to 70% confluence in 75-cm2 flasks and infected with BKV at an MOI of 5 in the presence of 2% fetal bovine serum for 1 h at 37°C. Cells were maintained in medium with 5% serum for 3 and 9 days postinfection. Chromatin immunoprecipitation (ChIP) was conducted by following the Santa Cruz Biotechnology protocol. Early promoter-containing NFAT mutants and wild-type (WT) early promoters were transfected in Vero cells using Fugene (Roche, Branchburg, NJ). Cells were treated with 2 μM ionomycin and 80 ng/ml PMA 48 h posttransfection. ChIP was carried out by following the protocol obtained from Santa Cruz Biotechnology. Cells were harvested by mechanical scraping and were washed twice with 1× phosphate-buffered saline (PBS). DNA was crossed-linked using 1% formaldehyde for 8 min at room temperature. To terminate cross-linking, cells were treated with 0.25 M glycine for 5 min on ice. Cells were resuspended three times with lysis buffer (Santa Cruz Biotech, Inc.), and crude nuclear extracts were resuspended in 1 ml of high-salt lysis buffer (Santa Cruz Biotechnology, Inc.). Nuclear extracts were sonicated using a Branson Sonifier 150 at a power setting of 2 six times for 1 min each on ice. A concentration of 500 μg of chromatin was precleared by protein A/G Plus-agarose (Santa Cruz Biotechnology, Inc.). Chromatin was incubated with primary antibody overnight at 4°C. Primary antibodies were purchased from Santa Cruz Biotechnology, Inc. Protein A/G Plus-agarose was added at 4°C for 2 h to harvest immune complexes. Beads were resuspended and washed with lysis buffer twice, followed by three washes with wash buffer (Santa Cruz Biotechnology, Inc.). Samples were eluted using elution buffer (Santa Cruz Biotechnology, Inc.) and incubated at 67°C overnight. DNA was extracted using a Qiagen PCR purification kit according to the manufacturer's protocol. SYBR green quantitative PCR (Applied Biosystems, Foster City, CA) was used to amplify the immunoprecipitated DNA. The primers used to detect the Dunlop promoter were (antisense) 5′-TGAGCTCCATGGATTCTTCC-3′ and (sense) 5′-ATTCCTAGGCTCGCAAAACA-3′.
Transient transfections were achieved using Fugene (Roche, Branchburg, NJ). Plasmids were cotransfected with the control luciferase plasmid Renilla (pRL-TK) from Promega. Cells were treated with 2 μM ionomycin and 80 ng/ml PMA 48 h posttransfection. Luciferase samples were measured using a Berthoid Lumat LB9501 luminometer. Luciferase results represent data from three independent transfections. The results of the firefly luciferase measurements were normalized using Renilla luciferase measurements.
Vero cells were transfected with 1 μg of linearized WT or viral promoter containing NFAT mutant DNA as previously described (9). Transfected cells were fixed and stained for V antigen (V-Ag) 4, 7, 10, 13, 16, and 19 days posttransfection. Virus-containing supernatants were collected at day 19. Vero cells were treated with supernatant as previously described (9). Cells were fixed and stained for V antigen expression 72 h postinfection.
To determine whether NFAT activity is required for BKV infection, we examined the effects of a cell-permeable NFAT inhibitor peptide, 11R-VIVIT, on BKV infection (2, 41). The NFAT inhibitor peptide binds to the calcium-dependent phosphatase, calcineurin, and inhibits its ability to dephosphorylate NFAT but does not disrupt the calcineurin phosphatase activity (2, 41). Vero cells were pretreated with the 11R-VIVIT peptide for 6 h and then infected with the BKV Dunlop strain for 1 h. Cells were maintained in medium containing the 11R-VIVIT peptide for 72 h. Infectivity was assessed by indirect immunofluorescence using antibodies specific for early viral protein T-Ag and late viral protein V-Ag (Fig. (Fig.1).1). We observed a dose-dependent decrease of infection in the presence of the 11R-VIVIT peptide. This indicates that NFAT contributes to BKV infection.
NFAT has been shown previously to regulate JCV and SV40 transcription (34, 35). To determine whether NFAT mediates BKV transcription, Vero cells were cotransfected with either the early promoter (pluc2/DunlopE) or the late promoter (pluc2/DunlopL) and a construct that expresses the NFAT inhibitor peptide sequence (pGFP-VIVIT) or a control construct expressing only GFP (phMGFP). Forty-eight hours posttransfection, cells were stimulated with ionomycin and phorbol 12-myristate 13-acetate (PMA) to activate the calcineurin-NFAT pathway (Fig. 2A and B). In the presence of pGFP-VIVIT there was a significant decrease in both early and late promoter activity. These results demonstrate that NFAT contributes to the regulation of BKV transcription.
The NFAT family is composed of five proteins, of which NFAT1 to NFAT4 are regulated by calcium signaling. As Vero cells express endogenous NFAT4 (35), we asked whether NFAT4 could bind to the viral promoter during infection using chromatin immunoprecipitation assays. Chromatin from infected cells was immunoprecipitated with an NFAT4 antibody at 3 and 9 days postinfection. NFAT4 bound to the promoter with a 3.5-fold increase at 3 days postinfection and a 9-fold increase at 9 days postinfection compared to that of DNA immunoprecipitated with an NFAT3 antibody or a nonspecific IgG control antibody (Fig. (Fig.3).3). This suggests that NFAT4 binds to the viral promoter during infection.
The Dunlop promoter contains a palindrome, two inverted repeat sequences, and a 20-bp A/T region followed by triplicate P regions encompassing three NFAT binding sites. To determine whether these sites are required for BKV transcription, we mutated the NFAT binding sites in the early promoter (pluc2/DunlopE) to be nonfunctional. The NFAT binding sequence was changed from GGAAA to TAGAT by site-directed mutagenesis (Fig. (Fig.4)4) (28). Mutations of individual NFAT binding sites or various combinations of NFAT sites as well as mutations of all three sites on the promoter were constructed. Mutational analysis demonstrated that mutations of these NFAT binding sites displayed a diverse range of promoter activity. Vero cells were transfected with either the wild-type Dunlop early promoter or mutant early promoter. We expected that mutations of these NFAT binding sites would result in a decrease in early promoter activity; however, mutations in NFAT site 1 demonstrated a 20-fold increase in early promoter activity compared to that of the WT (Fig. (Fig.5A).5A). Interestingly, mutations in NFAT site 2 demonstrated a greater increase in early promoter activity, increasing 54-fold compared to results for the WT (Fig. (Fig.5A).5A). Mutations of both NFAT site 1 and NFAT site 2 also led to enhanced promoter activity compared to that of the wild type. These findings suggest that the negative regulation of the promoter by NFAT has been disrupted by these mutations. Early promoter activity was reduced when NFAT site 3 was mutated (Fig. (Fig.5A).5A). Mutations of NFAT site 1 and NFAT site 3 showed wild-type activity. Mutations of site 2 and 3 showed significantly reduced activity compared to wild type. When all three NFAT binding sites were mutated, early promoter activity was abolished.
As luciferase data showed enhanced promoter activity with mutations in NFAT site 1 or NFAT site 2, we tested whether NFAT still could bind to the mutated viral promoters. Cells were transfected with wild-type early promoter or the relevant early promoter containing NFAT binding site mutations, and chromatin immunoprecipitation assays were performed. ChIP assays confirmed that NFAT4 bound to the wild-type promoter with a 7-fold increase compared to results for the IgG control (Fig. (Fig.5B).5B). When site 1 was mutated, NFAT binding to the promoter was increased (Fig. (Fig.5B).5B). When site 2 was mutated, NFAT binding was similar to binding to the wild-type promoter (Fig. (Fig.5B).5B). However, when site 3 was mutated, NFAT binding to the promoter again was increased (Fig. (Fig.5B).5B). These data suggest a hierarchy of NFAT binding, with functional NFAT site 1 being the dominant site and functional NFAT site 2 being a possible repressor site. Wild-type levels of NFAT binding when NFAT site 2 is mutated suggests that NFAT binds to NFAT site 1 and, possibly, site 3 under these conditions. Interestingly, NFAT4 could not bind to an NFAT site when site 1 and site 3 were mutated, suggesting that another transcription factor binds to NFAT site 2.
Luciferase assays and chromatin immunoprecipitation assays demonstrate that mutations of the NFAT binding sites in the early promoter enhance promoter activity and increase NFAT binding. To further investigate these effects, mutations of NFAT binding sites were generated in the viral genome to test the effect on viral spread and infectivity. Mutated viral genomes were transfected into cells, and cells were stained for late viral protein expression (V-Ag) (Fig. (Fig.6A).6A). Cells were fixed and stained at 3-day intervals 4 days posttransfection for a duration of 19 days (Fig. (Fig.6A).6A). At 4 days posttransfection, mutations of NFAT site 1 or NFAT site 2 demonstrated an increased level of viral spread compared to that of the WT, while mutations in NFAT site 3 showed a slight decrease. These data correlate with the luciferase data (Fig. (Fig.5A),5A), suggesting that any disruption of NFAT binding site 1 or NFAT binding site 2 results in enhanced viral promoter activity and viral spread at time points 2 to 4 days posttransfection. Mutations in NFAT site 1 or site 2 accelerated viral spread at later time points up to 19 days posttransfection (Fig. (Fig.6A).6A). Images depict the accelerated viral spread of mutations in NFAT binding site 1 (panel II) and NFAT binding site 2 (panel III) compared to that of the WT (panel I). The viral spread of mutated viral genome in NFAT binding site 3 was similar to that of the wild type at day 19 (Fig. 6A and B, panel IV). To assess the infectivity of each mutated viral genome, supernatants were collected at day 19 posttransfection and used to infect Vero cells. Mutations in NFAT site 1 or NFAT site 2 were shown to be more infectious than the wild type (Fig. (Fig.6C).6C). Taken together, viral propagation, luciferase, and ChIP data indicate that any disruption of NFAT binding site 1 or NFAT binding site 2 results in elevated levels of viral spread and enhanced promoter activity.
To elucidate the mechanisms of increased promoter activity due to NFAT binding site mutations on the viral promoter, we investigated the binding of AP-1 and subunit p65 of NF-κB to their recognized consensus sites on the early promoter. Functional synergy between NFAT and AP-1 has been shown in NFAT-dependent promoters in various immune cells (22, 32). It has been established previously that AP-1 and subunit p65 activate BKV early transcription (12, 36). In the Dunlop promoter there are AP-1 binding sites following NFAT site 1 and NFAT site 2 and a binding site for the subunit p65 of NF-κB upstream from the initiation start site (12). To test the effect of mutations of the NFAT binding sites in the viral promoter on AP-1 and NF-κB activation, cells were transfected with wild-type early promoter or early promoter containing NFAT binding site mutations, and ChIP was performed. Chromatin was immunoprecipitated for heterodimer c-Fos or c-Jun or for subunit p65, and results were compared to those for the IgG control antibody. As shown in Fig. 7A and B, the level of the binding of c-Jun and c-Fos increased when NFAT site 1 or NFAT site 2 was mutated. Interestingly, results demonstrated the increased binding of subunit p65 when NFAT site 1 or NFAT site 2 was mutated (Fig. (Fig.7C).7C). Mutations in NFAT site 3 did not result in any changes in the binding of c-Jun, c-Fos, or p65 to the promoter (Fig. 7A, B, and C). These data positively correlate with the luciferase data in Fig. Fig.5A,5A, suggesting that the transactivation of the early viral promoter is due to the enhanced activity of AP-1 and NF-κB when NFAT site 1 or site 2 is mutated.
The clinical importance of NFAT in BKV infection is highlighted by its dependence on calcineurin for activation. Calcineurin inhibitors and the immunosuppressive agents tacrolimus (FK506) and cyclosporine A are commonly used during kidney and bone marrow transplantations to reduce the risk of acute rejection (13, 33). The reactivation of BKV occurs under these conditions, resulting in the onset of the disease polyomavirus-associated nephropathy. Previous studies have demonstrated that NFAT contributes to the regulation of polyomaviruses, JC virus, and SV40 (34, 35).
In this study, we investigated the role of NFAT in the regulation of BKV infection. We found that a cell-permeable NFAT inhibitor peptide, 11R-VIVIT, dose dependently inhibits the infection of Vero cells by BKV. The NFAT inhibitor peptide specifically targets NFAT activity and does not inhibit AP-1 or NF-κB activation (2, 41), indicating that the reduction in infection is through a calcium-calcineurin-dependent pathway. Previous work has shown that Vero cells express calcium-regulated isoform NFAT4 (35). Our findings indicated that NFAT4 bound to the viral promoter during infection in Vero cells.
In the Dunlop promoter, we identified three NFAT binding sites located within each transcription factor binding triplicate P regions. A common feature of NFAT-dependent promoter activity is the presence of multiple NFAT binding sites. This is seen in the interleukin-2 (IL-2), interleukin-4 (IL-4), and CD95 promoters (4, 28, 43, 52). We analyzed the function of each NFAT binding site by mutating the consensus sequence from GGAAA to TAGAT, mutating each individual site as well as all possible combinations of sites. When all three sites were mutated, the promoter was nonfunctional. However, mutations of NFAT site 1 or NFAT site 2 resulted in a significant increase in promoter activity, suggesting that one or both of these sites negatively influenced promoter activity (Fig. 8B and C). These mutations led to the increased binding of fos/jun heterodimers and p65 to AP1 and NF-κB sites, respectively, and were correlated with the increased activity of the promoter. Mutations in NFAT site 3 slightly reduced promoter activity and led to enhanced NFAT binding to the promoter, presumably to sites 1 and 2 (Fig. (Fig.8D).8D). In the context of the mutations in NFAT site 3, AP-1 and p65 binding to the promoter was not increased.
NFAT proteins synergistically interact with the transcription factor AP-1. AP-1 heterodimers c-Fos and c-Jun are activated by the mitogen-activated protein kinase pathway to translocate to the nucleus, form NFAT-AP-1 complexes, and promote gene transcription. AP-1 and NF-κB have been shown to contribute to the transcriptional regulation of BKV (12, 36). Our findings show that enhanced promoter activity by mutations of NFAT site 1 or NFAT site 2 is due to the increased activity of NFAT, AP-1, and subunit p65 of NF-κB. The disruption of a potential repressor site may contribute to the increase in NFAT, AP-1, and NF-κB binding. In addition, promoter activity was abolished when all NFAT sites were mutated, indicating that NFAT sites are required for promoter activity and cooperation with these transcription factors.
Taken together, these data suggest that the BKV promoter is sensitive to changes in the relative levels of activators and that different combinations of these factors can both negatively and positively influence BKV transcription and infection. The regulation of the promoter likely is much more complex and may involve other factors. Previous studies have shown that BKV large T antigen binds to retinoblastoma family members regulating cellular proliferation in host cells by disrupting pRb-E2F repression (15). The phosphorylation of pRb disrupts the pRb-E2F complex and promotes the transcription of E2F family members E2F1 to E2F4 and E2F6 (14, 53). These factors bind to sequences that are similar to the NFAT binding sequences that we identified in the BKV promoter (53).
In summary, this study demonstrates that NFAT contributes to the regulation of BK virus infection and viral transcription. Multiple NFAT binding sites in the Dunlop promoter support the hypothesis that BKV early viral promoter activity is NFAT dependent. Additionally, NFAT cooperates with transcription factors AP-1 and NF-κB to regulate viral transcription. Individual NFAT sites in the viral promoter are activators and repressors of transcription to promote a low level of viral replication. Mutations of these sites disrupt the coordinated regulation of BKV transcription by NFAT, AP-1, and NF-κB and reveal important regulatory interactions between these factors. These mutations may mimic viral reactivation upon immunosuppression after organ transplantation. Future studies will determine whether other factors, such as E2F family members, bind to the viral promoter via NFAT binding sites and participate in the transcriptional regulation of BKV.
We thank all of the Atwood laboratory members for their critical discussion during the course of this work. We also thank Tammy Glass, Wendy Virgadamo, and Jamie Rees for their administrative assistance.
This project was supported by a grant from the National Cancer Institute (R01 CA71878).
Published ahead of print on 2 December 2009.