|Home | About | Journals | Submit | Contact Us | Français|
BK polyomavirus (BKV) establishes persistent, low-level, and asymptomatic infections in most humans and causes polyomavirus-associated nephropathy (PVAN) and other pathologies in some individuals. The activation of BKV replication following kidney transplantation, leading to viruria, viremia, and, ultimately, PVAN, is associated with immune suppression as well as inflammation and stress from ischemia-reperfusion injury of the allograft, but the stimuli and molecular mechanisms leading to these pathologies are not well defined. The replication of BKV DNA in cell cultures is regulated by the viral noncoding control region (NCCR) comprising the core origin and flanking sequences, to which BKV T antigen (Tag), cellular proteins, and small regulatory RNAs bind. Six nuclear factor I (NFI) binding sites occur in sequences flanking the late side of the core origin (the enhancer) of the archetype virus, and their mutation, either individually or in toto, reduces BKV DNA replication when placed in competition with templates containing intact BKV NCCRs. NFI family members interacted with the helicase domain of BKV Tag in pulldown assays, suggesting that NFI helps recruit Tag to the viral core origin and may modulate its function. However, Tag may not be the sole target of the replication-modulatory activities of NFI: the NFIC/CTF1 isotype stimulates BKV template replication in vitro at low concentrations of DNA polymerase-α primase (Pol-primase), and the p58 subunit of Pol-primase associates with NFIC/CTF1, suggesting that NFI also recruits Pol-primase to the NCCR. These results suggest that NFI proteins (and the signaling pathways that target them) activate BKV replication and contribute to the consequent pathologies caused by acute infection.
Human polyomavirus BK (BKV) persistently and asymptomatically infects approximately 80 to 90% of humans (25, 41). Kidneys are the major sites of replication, where BKV DNA is maintained at low levels (<0.01 copy/cell, on average) (20, 35) by the microRNA (miRNA)-mediated downregulation of the viral T antigen (Tag) (79) and the evasion of immune recognition (6). The activation of high levels of BKV replication in allografts occasionally occurs following kidney transplantation and can lead to viral titers exceeding 1,000 copies/cell (74), with concomitant viruria, viremia, and polyomavirus-associated nephropathy (PVAN), a major source of allograft loss. The causes of and mechanisms for the activation of viral DNA replication that occurs in the shift from persistent infection with low levels of replication to acute infection are not understood.
BKV replication in cell cultures is controlled by the viral noncoding control region (NCCR), within which the “core origin” (core-ori) serves as the initial binding site for the viral initiator-helicase protein, Tag, and small noncoding RNAs (21, 69, 84) (Fig. 1). Adjacent to the core-ori are the early flanking (EF) and the late flanking sequences (the “enhancer”), to which histones, cellular transcription factors, and perhaps also small noncoding RNAs bind and which control viral gene transcription and DNA replication (46, 52, 84, 85). The BKV archetype enhancer, comprised of four single-copy sequence blocks, termed P68, Q39, R63, and S63, rearranges by duplication, deletion, and insertion in late-stage PVAN or after passage in cell culture, providing a replication advantage and, perhaps, enhanced tropism (10, 30, 78).
Binding sites for numerous cellular transcription factors, including nuclear factor I (NFI) (14–16, 22, 47), Sp1 (14, 22, 47), NFAT (40), AP1 (15, 22, 47), Smad3 (1), ERE and GRE/PRE (53), p53 (80), NF-κB (28), and C/EBP (28), have been identified in the archetype BKV enhancer and rearranged BKV variants, with experimental evidence supporting the importance of some of these sites for viral transcription and replication. Also, putative binding sites for Ets1, PEA3, AP-2, CREB, and granulocyte-macrophage colony-stimulating factor (GM-CSF) have been predicted by sequence homology (52, 75), but their functional importance is unproven. Notably, multiple NFI binding sites occur in the BKV archetype enhancer (Fig. 1) as well as in rearranged enhancers (14, 22, 47), suggesting that these sites may be functionally important. While some of these NFI sites regulate BKV early and late promoter activities (15, 16, 31, 42), the direct involvement of NFI sites in viral DNA replication has not been demonstrated.
NFI was originally identified as a cellular factor that stimulates adenovirus (Ad) DNA replication by recruiting the viral DNA polymerase to the viral origin of replication and distorting its structure (19, 62, 64). Subsequent studies indicated that NFI is a family of four isotypes, NFIA, NFIB, NFIC, and NFIX (or NFID), with almost identical N-terminal DNA binding/dimerization domains that bind to “TGGN5~7GCCAA” sequences (32, 33). The expression pattern of NFI isotypes is cell type dependent and changes during differentiation and development (17, 43). NFI sites occur in many cellular promoters and enhancers as well as in viral genomes, including those of BKV (14–16, 22, 47), human JC virus (JCV) (55), variant murine polyomavirus (mPyV) (13, 86), human papillomavirus (HPV) (68), herpes simplex virus 1 (HSV-1) (44), and cytomegalovirus (CMV) (34). The functional importance of these NFI sites in regulating gene transcription is well established, but whether they also regulate DNA replication (other than adenoviral) has not been determined.
Here, we provide evidence for the functional importance of NFI for BKV archetype DNA replication: NFI sites placed proximal to the core-ori stimulate BKV DNA replication, and the mutation of NFI sites in the BKV enhancer diminishes replication in assays in which the mutant templates are in competition with the wild type (WT) for limiting factors. Furthermore, NFI interacts with two key replication proteins, BKV Tag and the p58 subunit of DNA polymerase-α primase (Pol-primase), as detected by in vitro pulldown assays and coimmunoprecipitation (co-IP) assays, and NFIC/CTF1 stimulates BKV DNA replication when Pol-primase is limiting in in vitro assays. We suggest that NFI and cognate signaling pathways help activate BKV replication and convert persistent infections into acute infections.
Test replication templates pUC-wt-BKV and pUC-Δen-BKV were generated by the insertion of PCR fragments of the intact BKV NCCR (positions 5031 to 282) and the NCCR without the enhancer (positions 5031 to 32) of the archetype Dik strain (GenBank accession number AB211369) into the XmaI/PstI sites of pUC18. pUC-6mtNFIs-BKV was generated by the ligation of a synthesized mutant BKV NCCR (GenScript) into the XmaI/PstI sites of pUC18. pUC-5mtNFIsW1-BKV, pUC-5mtNFIsW2-BKV, pUC-5mtNFIsW3-BKV, and pUC-5mtNFIsW6-BKV were derived from pUC-6mtNFIs-BKV by using the QuikChange site-directed mutagenesis kit (Stratagene). To construct test templates pUC-BKV-1fNFI, pUC-BKV-1rNFI, pUC-BKV-2rNFI, and pUC-BKV-4rNFI, the synthetic oligonucleotides 5′-CACATGGAATGTAGCCAAAACTGCA-3′ and 5′-GTTTTGGCTACATTCCATGTGTGCA-3′ (LNF1 consensus sites are underlined) were annealed, self-ligated, and inserted into the PstI site of pUC-Δen-BKV. Competition template pBC-wt-BKV was generated by the insertion of PCR fragments of the intact NCCR (positions 5031 to 282), enhancer (positions 33 to 282), and core origin (positions 5103 to 32) of BKV (Dik) into the XmaI/PstI, NotI/PstI, and NotI/XhoI sites of pBC-Sk(+). pBC-BKV-A89G, pBC-BKV-A143G, and pBC-BKV-A141T were mutated by using QuikChange site-directed mutagenesis.
The expression vector for BKV Tag, pCMV-BKTAg-Flag, lacking a simian virus 40 (SV40) origin, was described previously (46). NFI expression vectors pCH-NFIA, pCH-NFIB, pCH-NFIC, pCH-NFIX, and pCH-empty were kindly provided by Richard Gronostajski (18). pCH-hNFIC/CTF1 was generated by the insertion of the human NFIC/CTF1 cDNA (amplified by PCR from pCMV-CTF-1ΔUTR, kindly provided by Nicolas Mermod) (49), using the primers 5′-AGCTGGGCCCATGGATGAGTTCCAC-3′ and 5′-TTGCGCTAGCCTATCCCAGATACCAGGAC-3′, into the NheI/ApaI sites of the pCH-empty vector. The Pol-primase subunits were cloned into the expression vector pCMV with a T7 epitope tag at their N termini. The expression plasmids were kindly provided by E. Sock and M. Wegner (University of Erlangen—Nürnberg).
Bacterial expression vectors for truncated BKV Tag (pGEX3X-BKTHD, pGEX3X-BKTHDHR, and pGEX3X-BKTHR) were generated by the insertion of PCR-amplified fragments into the EcoRI/BamHI sites of pGEX3X. Primers used for PCR were BKTHD (5′-ATAGGATCCCAGGCTTAAAGGAGCATGATTTTAAC-3′ and 5′-CGGCCAATTCTTAATCAAGAATACATTTCCCCATG-3′), BKTHDHR (5′-ATAGGATCCCAGGCTTAAAGGAGCATGATTTTAAC-3′ and 5′-ACGCGAATTCTTATTTTGGGGGTGGTGTTTTAG-3′), and BKTHR (5′-ATATGGATCCCAATTACAAGAGAAGAGGATTCAG-3′ and 5′-ACGCGAATTCTTATTTTGGGGGTGGTGTTTTAG-3′).
HK-2 human proximal tubular kidney cells were cultured in RPMI 1640 medium (Invitrogen) supplemented with 10% fetal bovine serum (HyClone), 4 mM l-glutamine, 100 μg/ml penicillin, and 100 μg/ml streptomycin (Lonza). HEK293 cells and HeLa cells were cultured in Dulbecco's modified Eagle's medium (DMEM) (low glucose) (Invitrogen) supplemented with 10% fetal bovine serum (HyClone), 4 mM l-glutamine, 100 μg/ml penicillin, and 100 μg/ml streptomycin (Lonza). RPTECs (renal primary tubular cells) were cultured in renal epithelial growth medium with supplements (Lonza). All cells were grown at 37°C with 5% CO2 in a humidified incubator.
Antibodies used for Western blotting included anti-NFI (sc-870; Santa Cruz), anti-p53 (Cell Signaling), anti-c-Jun (sc-1694; Santa Cruz), anti-Sp1 (sc-59; Santa Cruz), anti-Ets1 (sc-111; Santa Cruz), anti-NF-κB (sc-372; Santa Cruz), anti-CREB (sc-58; Santa Cruz), anti-Smad3 (sc-8332; Santa Cruz), M2 anti-Flag (Sigma), anti-β-actin (sc-47778; Santa Cruz), anti-hemagglutinin (HA) (Roche), and anti-T7 (Novagen) antibodies. Polyclonal rabbit antiserum against recombinant human primase p58-p48 expressed in Escherichia coli cells was prepared and purified as previously described (87).
DNA replication assays with transfected cells were performed as previously described (46, 84), with transfection procedures optimized by luciferase/β-galactosidase reporter assays. The in vitro monopolymerase assays were performed as previously described (46, 85), with the following slight modifications: the assay mixture included 0.5 μg of BKV template DNA, 50 ng topoisomerase I, 1 μg replication protein A (RPA), and Pol-primase (as indicated in the figure legends) in a solution containing 30 mM HEPES (pH 7.8); 7 mM magnesium acetate; 0.1 mM EGTA; 0.5 mM dithiothreitol (DTT); 200 μM each UTP, GTP, and CTP; 4 mM ATP; 100 μM each dATP, dGTP, and dTTP; 10 μM dCTP; 40 mM creatine phosphate; 1 μg creatine kinase; 0.1 mg/ml heat-treated bovine serum albumin (BSA); and 5 μCi [α-32P]dCTP (3,000 Ci/mmol; Perkin-Elmer) in 40 μl. The amount of added NFI protein is indicated in the figure legends. Purified BKV Tag (0.4 μg) was added to start the reaction, and after incubation for 60 min at 37°C, the reaction products were precipitated with cold 10% (wt/vol) trichloroacetic acid (TCA) containing 2.5% (wt/vol) sodium pyrophosphate, spotted onto glass fiber filters (GF/C; Whatman), washed with 1 M HCl, and analyzed by scintillation counting.
HEK293 cells from 10 liters of a suspension culture were purchased from The National Cell Culture Center (NCCC) (Minneapolis, MN). Cell pellets were washed in 5 packed cell volumes (PCV) of a solution containing ice-cold hypotonic buffer (10 mM HEPES [pH 7.9], 1.5 mM MgCl2, 10 mM KCl, 0.2 mM phenylmethylsulfonyl fluoride [PMSF], 0.5 mM DTT) and centrifuged in a Beckman GH-3.7 rotor at 3,000 rpm for 5 min at 4°C. Collected cells were resuspended in 3 PCV of ice-cold hypotonic buffer and allowed to swell on ice for 10 min; swollen cells were transferred into a chilled glass Dounce tissue grinder, homogenized with 10 to 15 strokes using a type B pestle, and centrifuged in a Beckman GH-3.7 rotor at 3,500 rpm for 15 min at 4°C; the supernatants were removed; the pellets were resuspended in 1 packed nuclear volume (PNV) of low-salt buffer (20 mM HEPES [pH 7.9], 25% glycerol, 1.5 mM MgCl2, 0.2 mM EDTA, 0.2 mM PMSF, 0.5 mM DTT); and 0.183 PNV of 5 M NaCl was added dropwise to the resuspended nuclei while mixing gently by swirling, placed onto a rotating platform for 30 min at 4°C, and centrifuged at 16,000 rpm (37,000 × g) in a Sorvall SA-600 rotor for 1 h at 4°C. The supernatants were collected and dialyzed in dialysis buffer (20 mM HEPES [pH 7.9], 20% glycerol, 5 mM NaCl, 0.2 mM EDTA, 0.2 mM PMSF, 0.5 mM DTT). The protein concentration of the extracts was adjusted to 1.4 μg/μl, determined by a Bradford protein assay (9a). Small-scale preparations of HK-2 cell, HeLa cell, RPTEC, and HEK293 cell nuclear extracts for Western blotting were made by using nuclear extract kits (Active Motif).
Competitive electrophoretic mobility shift assays (EMSAs) were performed according to previously described procedures (11), with the following modifications: 4 pmol competitor oligonucleotides, 2 μg of HeLa cell nuclear extracts (sc-2120; Santa Cruz) or 0.25 μg of purified human NFIC (hNFIC)/CTF1 (Abcam), and 50 ng of poly(dI · dC) were incubated in 1× HEPES binding buffer (25 mM HEPES [pH 7.5], 6 mM MgCl2, 0.5 mM EDTA, 0.5 mM DTT, 50 mM NaCl, 0.5 mM PMSF, 5% glycerol) for 10 min at room temperature (RT); 20 fmol biotin-labeled probe for the NFI site was then added to the reaction mixture and incubated for 20 min at room temperature. For antibody supershift EMSAs, 1 μl of NFI antibody (sc-5567; Santa Cruz) was added after the binding reaction, and the mixture was incubated for another 10 min at room temperature. The reaction products were fractionated in 5% nondenaturing polyacrylamide gels (prerun in 0.5× Tris-borate-EDTA [TBE] at 100 V for 30 min) in 0.5× TBE at 100 V until the bromophenol blue dye reached the bottom of the gel. The products were transferred onto nylon membranes in 0.5× TBE at 65 V for 30 min, UV cross-linked for 15 min, and visualized by chemiluminescent nucleic acid detection (Pierce).
BKV Tag with an N-terminal Flag epitope was expressed by using the Bac-to-Bac baculovirus system (Invitrogen) as follows: 1.5 × 107 infected Hi-Five insect cells were harvested at 48 h postinfection (p.i.) and lysed in 1 ml of 0.5% NP-40 lysis buffer (50 mM Tris-Cl [pH 7.5], 150 mM NaCl, 5 mM KCl, 1.0 mM MgCl2, 0.5% NP-40, 10% glycerol, 1× PhosSTOP phosphatase inhibitors [Roche], 1× complete protease inhibitor cocktail [Roche]) by incubation on a rotating platform at 4°C for 30 min and homogenization with a glass Dounce tissue grinder. Lysates were cleared by centrifugation at 20,000 × g in a Sorvall SA-600 rotor for 30 min at 4°C. Supernatants were incubated with 60 μl of an anti-Flag M2 affinity gel (Sigma) by rotation at 4°C for 2 h and washed three times with 1 ml of ice-cold phosphate-buffered saline (PBS); each of the gel suspensions was then incubated with extracts of infected and uninfected Hi-Five cells and then incubated with 1 ml of HEK293 cell nuclear extracts (1.4 μg/μl) plus 1× complete protease inhibitor cocktail (Roche) by rotation at 4°C for 12 h. The gel was washed three times with 1 ml of ice-cold PBS and boiled for 5 min with 50 μl of 1× SDS sample buffer.
Glutathione S-transferase (GST)-truncated BKV Tag proteins were expressed in E. coli Rosetta 2 cells (Novagen) cultured in LB medium on a shaking platform at 225 rpm at 25°C. Expression was induced with 1 mM isopropyl-β-d-thiogalactopyranoside (IPTG) when the A600 reached 0.4 to 0.5. After induction, E. coli cells were cultured at 25°C with shaking at 225 rpm for 20 h and then sonicated (twice for 300 s each with a 20% duty cycle at maximum power) in L1 buffer (50 mM Tris-Cl [pH 8.0], 250 mM NaCl, 1 mM DTT, 10% glycerol, 1 mM PMSF, 1× complete protease inhibitor cocktail [Roche]). NP-40 (0.1%) was added to the lysates after sonication and incubated on ice for 10 min. After centrifugation at 20,000 × g in a Sorvall SA-600 rotor for 30 min at 4°C, supernatants were incubated with glutathione beads (GE Healthcare) at 4°C for 2 h, followed by washing 2 times with 20 bed volumes of L1 buffer. A small portion (1/20) of the beads for each fusion protein was boiled with SDS sample buffer, and the quantity of bound fusion proteins was determined by SDS-PAGE and Coomassie blue staining. Beads with approximately equal amounts of each fusion protein (~50 μg) were incubated with 800 μl of HEK293 cell nuclear extract (1.4 μg/μl) supplemented with 1× complete protease inhibitor cocktail (Roche) by rotation at 4°C for 12 h. The beads were washed 5 times with 1 ml of L1 buffer and then boiled for 5 min with 100 μl 1× SDS sample buffer. All samples were fractionated by SDS-PAGE and analyzed by Coomassie blue staining, followed by Western blotting.
HEK293 cells (approximately 6 × 106 cells) were transfected with expression vectors by using LipofectAMINE and Plus reagent (Invitrogen) in 100-mm-diameter plates as previously described (46). At 48 h, the cells were lysed with 750 μl of a solution containing 1% Triton lysis buffer (50 mM Tris-Cl [pH 8.0], 150 mM NaCl, 1% Triton X-100, 30 μM ethidium bromide, 1 mM PMSF, complete protease inhibitor cocktail [Roche], 1× PhosSTOP phosphatase inhibitors [Roche]) by rotation for 1 h at 4°C, and lysates were cleared by centrifugation at 20,000 × g in a Sorvall SA-600 rotor for 30 min at 4°C. A sample (10 μl) was taken for the protein input control, and the remaining supernatants were incubated with 50 μl of an anti-Flag M2 affinity gel (Sigma) or 50 μl of an anti-HA affinity matrix (Roche) by rotation at 4°C for 2 h, followed by four washes with 1 ml of 1% Triton lysis buffer. Following the last wash, pelleted beads were suspended in 40 μl of 1× SDS sample buffer and boiled for 5 min. Samples of 20 μl of the supernatant were analyzed by SDS-PAGE and Western blotting.
Six putative NFI sites occur in the BKV archetype enhancer (Dik strain) (Fig. 1), and EMSAs were employed to evaluate NFI binding to these sites. As a positive control for NFI protein-oligonucleotide complexes, an EMSA was performed with a biotin-labeled oligonucleotide with the sequence of the NFI site in the adenovirus inverted terminal repeat (ITR) (45). As expected, a shift of the migration of the oligonucleotide caused by NFI binding was observed in the absence of a competitor (Fig. 2A, lanes 1 and 6), and the addition of a 200× excess of the unlabeled oligonucleotide fully competed for NFI binding (Fig. 2A, lanes 2 and 7); however, an oligonucleotide with a point mutation in the NFI site did not alter the band shift (Fig. 2A, lane 3). The addition of an anti-NFI antibody to the binding reaction mixture caused a supershift (Fig. 2A, lane 4, dashed arrow), confirming that the band shift was caused by NFI. An excess of BKV enhancer DNA also competed for NFI binding (Fig. 2A, lane 5), confirming that the BKV archetype enhancer contains functional NFI sites. Oligonucleotides with sequences of the six NFI binding sites in the BKV archetype enhancer (Fig. 1 and and3A)3A) competed for NFI binding with different efficiencies (Fig. 2A, lanes 8 to 13), suggesting that these sites have different affinities for NFI proteins. Oligonucleotides with sequences of NFI sites closer to the BKV core-ori competed more efficiently than did those with distal sites.
Sequences corresponding to the first NFI site in the P block (Fig. 1) in different orientations and copy numbers were substituted for enhancer sequences (Fig. 2B) and assayed for Tag-dependent DNA replication in HK-2 cells (Fig. 2C, lanes 2 to 5). The addition of synthetic NFI sites to templates lacking the enhancer stimulated their replication, with the extent correlating with the number of sites (Fig. 2C, lanes 6 to 8) but not their orientations (Fig. 2C, lanes 3 and 4). The replication of templates containing multiple synthetic NFI sites was always weaker than the replication of templates with the archetype enhancer (Fig. 2C, lanes 1, 5, 7, and 8), suggesting that other elements also stimulate BKV DNA replication in these cells or that a particular spatial configuration of NFI sites is required.
Point mutations were introduced into the consensus sequence “TGGN5~7GCCAA” of each of the six NFI sites in the BKV enhancer, with the design of the mutations confirmed with MatInspector (72) to ensure that no new transcription factor binding sites were created (Fig. 3A). Oligonucleotides with mutated NFI sites were demonstrably defective for NFI binding in competitive EMSAs (Fig. 3B). A 200-fold excess of the BKV enhancer DNA with mutations in all six NFI sites (6mtNFIs BK enhancer) only slightly reduced the NFI binding to the adenovirus ITR oligonucleotide (Fig. 3C, compare lanes 1 and 3), whereas the wild-type (WT) enhancer sequence completely abolished NFI binding (Fig. 3C, lane 2). These results indicate that the mutation of the six identified NFI sites greatly reduces NFI binding to the BKV archetype enhancer.
Assays of the replication of BKV templates containing mutations of all six NFI sites (pUC-6mtNFIs-BKV) compared with the replication of the wild-type template (pUC-wt-BKV) in HK-2 cells transfected with each template separately indicated that both templates replicated with similar efficiencies (data not shown). However, the use of a competitive assay in which a competitor (3.4-kb) template, pBC-wt-BKV, containing the archetype NCCR was cotransfected into HK-2 cells together with test mutant templates (2.6 kb) (Fig. 4A) and a Tag expression vector (pCMV-BKT-Flag), revealed that the level of replication of the test templates with the enhancer containing six mutant NFI sites (pUC-6mtNFIs-BKV) was greatly reduced (Fig. 4B, compare lane 3 with lane 1), as was also observed for a test template without the enhancer (Fig. 4B, lane 2). Test templates with mutations in Ets1 binding sites in the enhancer (pUC-mtEts1s-BKV) replicated with an efficiency similar to that of the wild-type template (Fig. 4B, compare lane 5 with lane 1), indicating that the effect of the mutations on DNA replication in the competitive assay is attributable to the NFI site.
To determine whether a specific NFI site is responsible for the stimulatory effect on DNA replication, pUC-6mtNFIs-BKV templates with single NFI sites reverted to the WT were assayed for replication. Wild-type NFI sites 1 and 2 (closer to core-ori) appeared to be more stimulatory for BKV DNA replication than NFI sites 3 and 6 (Fig. 4C, lanes 3 to 7).
The requirement of a competitor template to observe the stimulatory effect of NFI sites on BKV DNA replication might be explained by the recruitment of factors required for DNA replication that are made limiting by use of the competitor template. The observation that NFI sites closer to the core-ori are more stimulatory for replication than are sites distal to the core-ori suggests that NFI may target components of the initiation complex, such as Tag, Pol-primase, replication protein A (RPA), or topoisomerase I. Evidence favoring this notion is provided by the experiments described below.
We tested whether NFI interacts with BKV Tag by antibody pulldown assays. Full-length Flag-tagged BKV Tag (BKT-Flag) was mixed with HEK293 nuclear extracts, protein complexes were collected by using Flag antibody resin, and associated candidate proteins were detected with specific antibodies to each one (Fig. 5A). Initially, a pan-NFI antibody recognizing all NFI isotypes was used in the pulldown assays; other antibodies also detected the association of BKV Tag with p53, Sp1, and c-Jun but not Ets1, CREB, NF-κB p65, or Smad3 (Fig. 5A). Domains of BKV Tag that interact with these transcription factors were determined with GST-tagged truncated BKV Tags (Fig. 5B): the BKV Tag helicase/ATPase domain (HD) pulled down NFI, p53, c-Jun strongly, and Sp1 weakly but not CREB or NF-κB (Fig. 5C, lanes 4 and 5), indicating that NFI, p53, and c-Jun interact with the helicase/ATPase domain, while Sp1 may also interact with other Tag domains in addition to the HD. The latter observations are consistent with previous reports that BKV Tag and SV40 Tag complex with p53 (80), the SV40 Tag origin binding domain (OBD) interacts with Sp1 (39), and SV40 interacts with c-Jun (7). None of the transcription factors tested appeared to interact with the BKV Tag C-terminal region (Fig. 5C, lane 3). Truncated BKV Tags did not pull down factors not observed to complex with full-length BKV Tag, including Ets1, NF-κB, CREB, and Smad3 (Fig. 5A and C and data not shown).
Isotype-specific antibodies directed against NFIA, NFIB, and NFIC were used to attempt to distinguish isotype-specific interactions with BKV Tag, but only NFIA was detected at a high level in the HEK293 cell extracts and was pulled down by Tag (data not shown). To determine other isotypes of NFI that interact with BKV Tag, HA-tagged NFI isotypes and Flag-tagged BKV Tag were overexpressed in HEK293 cells, and their interactions with Tag were assessed by co-IP assays using either anti-HA or anti-Flag antibodies (Fig. 5D). Four HA-tagged NFI isotypes were expressed at similar levels (Fig. 5D); however, NFIB strongly upregulated the expression of Tag driven by the CMV promoter (Fig. 5D, lane 3), leading to a higher level of coprecipitation of Tag with NFIB than with other NFI isotypes. In contrast, HA-NFIA only weakly coprecipitated with Tag (Fig. 5D, lane 2), which might be due to a high level of endogenous NFIA competing against the transiently expressed NFIA in an interaction with Tag (Fig. 5D, lane 2, and data not shown). In control assays, no interaction between Tag and the HA-tagged Gal4 DNA binding domain (Gal4dbd) was detected, even though Tag was expressed at an extremely high level (Fig. 5D, lane 6), and no interaction of Tag with endogenous NF-κB p65 was detected (Fig. 5D). These results are in agreement with data from in vitro pulldown assays (Fig. 5A and C), indicating a specific interaction of BKV Tag with HA-tagged NFI isotypes.
The interaction of BKV Tag with NFI provides a possible basis for the stimulatory effect upon replication by NFI sites in the enhancer. Initially, BKV Tag was expressed at a low level in the competitive DNA replication assays (Fig. 4B); to investigate if BKV Tag is the only limiting factor in these assays, increasing amounts of a pCMV-BKV Tag expression vector were introduced into HK-2 cells, and the levels of DNA replication of wild-type and mutant (NFI site) templates were compared (Fig. 6). BKV Tag expressed from 400 ng of pCMV-BKTAg saturated replication (Fig. 6, lane 7 to 9), but this amount, and even the addition of 900 ng pCMV-BKTAg, only partially rescued the replication of the template with the mutant NFI sites (pUC-6mtNFIs-BKV), compared with the robust replication of the wild-type template (pUC-wt-BKV) (Fig. 6, compare lane 9 with land 7 and compare lane 12 with lane 10). Furthermore, templates with a deleted enhancer (pUC-Δen-BKV) replicated poorly, even when BKV Tag was highly expressed (Fig. 6, lanes 8 and 11). These observations indicate that high levels of BKV Tag alone cannot correct the replication deficiency of templates with mutant NFI sites and suggest that BKV Tag is not the sole limiting factor in these replication assays.
BKV Tag interacts with different NFI isotypes in co-IP assays (Fig. 5D), but their importance for BKV replication is difficult to test in vivo due to the concurrent expression of multiple endogenous NFI isotypes. The NFIC/CTF1 isotype stimulates the initiation of adenovirus DNA replication by recruiting adenovirus DNA polymerase to the origin of replication (9, 19), and the proline-rich transactivation domain of the NFIC/CTF1 isotype stimulates SV40 DNA replication when tethered to the SV40 origin (61). These observations prompted us to assess whether NFIC/CTF1 can stimulate BKV DNA replication in the monopolymerase replication system that contains Pol-primase, RPA, and topoisomerase I (46, 85) in the absence of other cellular factors.
When Pol-primase was made limiting in the monopolymerase system, the DNA replication of the wild-type BKV template was stimulated strongly by NFIC/CTF1 in a concentration-dependent manner (Fig. 7, lanes 4 to 6). In contrast, with high levels of Pol-primase, the addition of NFIC/CTF1 had no effect on the replication of the wild-type BKV template (Fig. 7, lanes 1 to 3). Furthermore, no stimulation was observed with the NFI binding-site mutant BKV template regardless of the Pol-primase levels (Fig. 7, lanes 7 to 12). These findings indicate that NFIC/CTF1 stimulates the initiation of BKV DNA replication in vitro only when Pol-primase is limiting and suggest that Pol-primase is a limiting factor targeted by NFI to stimulate BKV DNA replication in vivo in the competitive replication assays.
To further investigate the stimulation of replication by NFI, we studied whether it and Pol-primase proteins interact. Pol-primase consists of four subunits: two smaller subunits, p48 and p58, constitute the primase, and two larger subunits, p180 and p68, constitute the DNA polymerase-α catalytic and regulatory subunit p68, respectively (50, 51, 65, 82). HA-tagged human NFIC/CTF1 was ectopically expressed with each of four T7-tagged human Pol-primase subunits in HEK293 cells (Fig. 8A), and their interaction was examined by co-IP assays. The expression of p180 and p68 subunits was much more efficient than the expression of the primase subunits (as detected with the anti-T7 antibody) (Fig. 8A, lanes 3 to 6), but no interaction between NFIC/CTF1 and either the p180 or p68 subunits was detected (data not shown). Because the p58 expression level was low (Fig. 8A, lane 4), a more sensitive antibody against p58 was used, instead of the anti-T7 antibody, for the detection of p58 (Fig. 8B). In this assay, p58 was found to coprecipitate with NFIC/CTF1 (Fig. 8B, lane 1), and p58 co-IP was not detected in three control experiments, in which only p58 (Fig. 8B, lane 2), only NFIC/CTF1 (Fig. 8B, lane 3), or neither (Fig. 8B, lane 4) was expressed. No other significant distinct band was detected in the coprecipitated fraction. Unfortunately, despite several attempts, the level of expression of p48 was extremely low. An interaction between p48 and NFIC/CTF1 was not detected (Fig. 8A, lane 3, and data not shown), but we cannot exclude that p48 might bind to NFI.
Almost 2 decades after the etiology of PVAN was linked to acute BKV replication, the cause of the activation of BKV replication in kidney allografts still remains elusive. Immune suppression is associated with the activation of BKV replication (2, 26, 89), and stress-related injury, repair, regeneration, and differentiation are also associated (27, 37), but the responsible mechanisms have not been defined.
We have determined that NFI binds to six NFI sites in the BKV archetype NCCR and stimulates DNA replication. NFI sites in P24–37 (NFI-1) and at the P68-Q13 junction (NFI-2), proximal to the core-ori, appear to have a higher affinity for NFI and also stronger stimulatory effects on BKV DNA replication than distal sites. Almost all rearranged viruses contain the P block and P-Q junction region spanning these two NFI sites (30, 54, 70, 71, 75), supporting the notion that they might be particularly important for efficient viral replication in vivo. Other NFI sites have been implicated in the early-late transcription switch (NFI-3) (42), the regulation of viral gene transcription in response to the induction of transforming growth factor β (TGF-β) in kidney allografts (NFI-4) (1), and the modulation of the hormone-mediated stimulation of BKV replication (NFI-5 and -6) (53).
We attempted to distinguish the binding of different NFI isotypes to NFI sites using isotype-specific antibodies. Unfortunately, none of the available isotype-specific antibodies work in EMSAs. As NFI isotypes have an identical DNA binding domain that determines the specificity of binding to consensus sites, we speculate that the different NFI isotypes have similar binding affinities in vitro. However, the binding activity of NFI isotypes in vivo is likely affected by their interaction and/or competition with other transcription factors (66, 73, 77) and is challenging to demonstrate with in vitro assays using purified proteins.
The topography of NFI sites in the archetype BKV enhancer resembles that of NFI sites in the archetype JCV enhancer, except that JCV lacks the NFI-4 site overlapping the Smad3 site (48). The NFI site closest to the core origin of JCV also stimulates JCV DNA replication in vivo (81). JCV also persistently infects the kidney, and the reactivation of JCV in immunocompromised individuals causes progressive multifocal leukoencephalopathy (PML) (88). Previous analyses of rearranged JCV enhancers in PML patients also revealed a trend similar to that for rearranged BKV enhancers in PVAN: sequences close to the core origin (A to C for JCV and P and the P-Q junction for BKV), which contain the first two NFI sites (NFI-1 and NFI-2), are usually preserved and duplicated (29, 30). NFI isotype-specific expression determines the tropism of JCV (55, 76), but functions for different NFI isotypes in replication have not been identified.
A characterization of the NFI isotype-specific function for BKV DNA replication has been attempted with an in vivo replication system, but the results are complicated by the endogenous expression of different NFI isotypes/splicing variants (data not shown). Using the in vitro monopolymerase assay, we have defined the stimulatory activity of the NFIC/CTF1 isotype, the prototype of the NFI family of transcription factors, on the initiation of BKV DNA replication (Fig. 7). The role of other NFI isotypes in DNA replication will be tested with similar systems as purified NFI isotype proteins and antibodies become available.
As NFI stimulates adenovirus (Ad2/5) DNA replication through the recruitment of the Ad pol-pTP (adenovirus DNA polymerase-preterminal protein) complex to the replication origin (9, 19, 60) and/or the stabilization of the preinitiation complex (59), we suggest that similar mechanisms promote BKV DNA replication via the recruitment of BKV Tag and Pol-primase to the replication origin. In support of this, we observed that NFI stimulated the initiation of BKV DNA replication in vitro only at low concentrations of Pol-primase and NFI, whereas at high concentrations of Pol-primase, no stimulation was observed (Fig. 7), which is reminiscent of the stimulation of adenovirus DNA replication in vitro by NFI (60) and which is also consistent with the results of our in vivo competitive replication assays. Furthermore, NFI forms complexes with BKV Tag and Pol-primase; swine NFI also binds to calf primase and stimulates primase activity in a concentration-dependent manner in biochemical assays (24). All four NFI isotypes were found to interact with BKV Tag in co-IP assays (Fig. 5D).
These results suggest that the formation of NFI-primase and NFI-Tag complexes is important for the initiation of DNA replication, but elucidating the functions of individual complexes is challenging due to the coexpression of different NFI isotypes/splicing variants. Further studies are required to characterize the nature and functions of these interactions.
Although the NFI sites in the BKV NCCR are not required for BKV DNA replication in the absence of a competitor template, NFI sites in the enhancer stimulate BKV DNA replication when Tag or Pol-primase is limiting. This stimulatory activity might be essential in persistent infections, where BKV replicates at very low levels in kidney tubular epithelial cells when low levels of Tag and Pol-primase are expressed (20, 35, 74), and for the reactivation of BKV replication. The tubular epithelial cells in normal kidneys are terminally differentiated quiescent cells that divide at a low rate (8, 63) and express small amounts of Pol-primase (36, 83). Ubiquitously expressed NFI may facilitate the low-level replication of persistent BKV in kidney epithelial cells by increasing the level (and activity) of Pol-primase at the core-ori. Also, signaling mediated through TGF-β (3, 4), tumor necrosis factor alpha (TNF-α) (4), and oxidative stress (5, 56–58) induced by kidney ischemia/reperfusion injury and/or inflammatory responses (12, 23, 38) during kidney transplantation or by the administration of immunosuppressive drugs such as tacrolimus and cyclosporine (67) might alter NFI isotype expression or activity and thereby promote the NFI-mediated recruitment of Tag and/or Pol-primase to the viral core-ori. These notions can be tested experimentally.
We thank members of the Folk laboratory, especially Sarah Scanlon and Olga Kenzior for their advice and assistance and David Pintel, Mark Hannink, David Setzer, and Michael Imperiale for their constructive criticisms.
This work was supported by funds from the University of Missouri—Columbia (W.R.F.), grants from the Science Foundation Ireland (H.P.N.), an NUIG student fellowship, and a Thomas Crawfort Hayes fellowship (both I.T.).
Published ahead of print 28 December 2011