Alkoxylalkyl Esters of Nucleotide Analogs Inhibit Polyomavirus DNA Replication and Large T Antigen Activities

Polyomavirus-related infections are ubiqutious in immunocompromised individuals, and in some cases are intractable and fatal. Due to a lack of approved drugs to treat polyomavirus infections, cidofovir, a phosphonate nucleotide analog approved to treat cytomegalovirus infections, has been repurposed as an antipolyomavirus agent.

ABSTRACT

Polyomavirus infections occur commonly in humans and are normally nonfatal. However, in immunocompromised individuals, they are intractable and frequently fatal. Due to a lack of approved drugs to treat polyomavirus infections, cidofovir, a phosphonate nucleotide analog approved to treat cytomegalovirus infections, has been repurposed as an antipolyomavirus agent. Cidofovir has been modified in various ways to improve its efficacies as a broad-spectrum antiviral agent. However, the actual mechanisms and targets of cidofovir and its modified derivatives as antipolyomavirus agents are still under research. Here, polyomavirus large tumor antigen (Tag) activities were identified as the viral target of cidofovir derivatives. The alkoxyalkyl ester derivatives of cidofovir efficiently inhibit polyomavirus DNA replication in cell-free human extracts and a viral in vitro replication system utilizing only purified proteins. We present evidence that DNA helicase and DNA binding activities of polyomavirus Tags are diminished in the presence of low concentrations of alkoxyalkyl ester derivatives of cidofovir, suggesting that the inhibition of viral DNA replication is at least in part mediated by inhibiting single-stranded DNA (ssDNA) and double-stranded DNA (dsDNA) binding activities of Tags. These findings show that the alkoxyalkyl ester derivatives of cidofovir are effective in vitro without undergoing further conversions, and we conclude that the inhibitory mechanisms of nucleotide analog-based drugs are more complex than previously believed.

INTRODUCTION

Polyomaviruses belong to the polyomaviridae family of viruses characterized by their nonenveloped small double-stranded circular genomic DNA of about 5 kb, packaged into icosahedral-shaped virions, and their narrow host range (1–4). The viruses have high prevalence in humans and maintain lifelong latency but immunocompetent hosts remain asymptomatic. However, in some hosts, such as patients suffering from HIV/AIDS or lymphoproliferative disorders and those on immune suppressants following kidney or hematopoietic stem cell transplants, as well as autoimmune disease patients treated with biologics and humanized monoclonal antibodies, the reactivation or primary infection of polyomaviruses has been implicated in the pathogenesis of various severe polyomavirus-associated diseases, such as polyomavirus-associated nephropathy (PVAN), hemorrhagic cystitis (HC), trichodysplasia spinulosa (TS), and progressive multifocal leukoencephalopathy (PML), among others (1–8).

The treatment of polyomavirus-associated diseases is mainly based on management and withdrawal of the predisposing conditions, because no drugs have been approved to treat the infections. However, an FDA-approved drug for the treatment of cytomegalovirus (CMV) infection in HIV/AIDS patients, 1-[(S)-3-hydroxy-2-(phosphony-lmethoxy)propyl] cytosine (HPMPC) (also called cidofovir [CDV] or Vistide) has been frequently used to manage polyomavirus-associated diseases without special approval (9–14). Cidofovir, an acyclic nucleoside phosphonate (ANP), is a dCMP analog and a prodrug which circumvents the requirement for a virally encoded thymidine kinase for activation, whereas the latter is necessary for the first critical phosphorylation step to activate nonphosphonate nucleoside analogs such as acyclovir (12, 13). Thus, CDV and other ANPs have broad-spectrum antiviral activities against DNA virus infections whether they encode a thymidine kinase or not (12, 13). To metabolize CDV to its active antiviral drug form, the prodrug is transported across the cellular plasma membrane and is subsequently intracellularly converted to CDV-diphosphate (CDV-PP) (13, 15, 16), which is the active form of the drug and competes with dCTP as an alternative substrate for viral DNA polymerases as seen for CMV and poxviruses (12, 13, 15, 16). The preliminary application of CDV to treat polyomavirus-associated diseases was performed following its successful use in the treatment of CMV and herpes simplex virus (HSV) infections, but it is worth noting that both viruses differ from polyomaviruses in their protein requirements for the replication of their genome (12, 13, 17, 18). CMV and HSV encode their own DNA polymerases to replicate their genomes, whereas polyomaviruses do not and depend on host DNA polymerase α-primase (Pol-prim) and DNA polymerase δ for the replication of their genomes (12, 13, 17–21). In contrast to the viral DNA polymerases, these cellular DNA polymerases show very low affinity for CDV-PP as an alternative substrate to dCTP, whereas CMV and HSV DNA polymerases exhibit comparatively high affinities for CDV-PP (12, 13, 15–18).

In recent years, researchers have intensified efforts to overcome the limitations of CDV, such as its low cellular uptake, lack of oral administration to patients, and accumulation in kidney tubules with resultant nephrotoxicity (15, 16). Consequently, CDV has been extensively modified, including 5-Aza substitution to form 5-Aza-CDV (aCDV), cyclization of aCDV to form 5-Aza-cyclic-CDV (acCDV), and esterification of CDV and acCDV with alkoxyalkanols to form hexadecyloxypropyl CDV (HDP-CDV or Brincidofovir) and hexadecyloxyethyl acCDV (HDE-acCDV) (22–28). Previous evaluations of the CDV derivatives report that the alkoxyalkyl ester derivative HDP-CDV shows good bioavailability after oral administration, has low risks of nephrotoxicity, and exhibits an increase in efficacy that is attributed to increased cellular uptake (22–32). The metabolism of the alkoxyalkyl ester of CDV, such as HDP-CDV, requires cellular phospholipase C enzymatic activity to release CDV in cells. The latter is subsequently converted to CDV-PP by cellular kinases (12, 33–37). Although CDV has been used to treat polyomavirus-associated symptoms, the knowledge of the targets and actual mechanism of inhibition of polyomavirus propagation is largely unknown. Therefore, it is important to gain more knowledge of the mechanistic action of these nucleotide analogs toward the propagation of polyomavirus to specifically delineate whether drug administration or the withdrawal of immunosuppressants is the main contributing factor to a reduction in polyomavirus load in patients. The withdrawal of immunosuppressants has been shown to enhance the recovery from polyomavirus-associated symptoms, and nucleotide analog treatment of patients has also always been accompanied by a reduction or total withdrawal of immunosuppressants (36, 38, 39). Various studies of the therapeutic effects of administering CDV against polyomavirus-associated symptoms in patients and cell culture gave conflicting outcomes (10, 14, 36, 38–46). This study evaluated the effects of CDV and its derivatives on the in vitro replication of polyomavirus DNA with a focus on the ATPase, helicase, and DNA binding activities of polyomavirus large tumor antigens (Tags).

RESULTS

In vitro replication of polyomavirus DNA in crude extract is inhibited by alkoxyalkyl esters of cidofovir.

The efficiency of nucleotide analogs, including cidofovir (CDV) and its derivatives (CDV-P, CDV-PP, HDP-CDV, aCDV, acCDV, and HDE-acCDV) to inhibit the replication of polyomaviruses was investigated using replication of polyomavirus DNA in human S100 extract. The data revealed that CDV and its monophosphorylated product CDV-P did not inhibit the in vitro replication of simian virus 40 (SV40) DNA, as they exhibit inhibition similar to the solvent-treated control and a high 50% inhibitory concentration (IC₅₀) of >500 µM (Fig. 1A). The diphosphorylated CDV derivative, CDV-PP, which is the active metabolite that inhibits CMV and HSV DNA polymerases (12, 33–37) and, to a lesser extent the DNA polymerase α, only slightly inhibited the SV40 DNA replication in vitro with an estimated IC₅₀ of 280 µM. In contrast, the CDV alkoxyalkyl ester HDP-CDV (hexadecyloxypropyl cidofovir, brincidofovir) showed a concentration-dependent inhibition of SV40 DNA replication in vitro with an IC₅₀ of 13.3 ± 0.07 µM (Fig. 1A; summarized in Table 1). It is important to note that the alkoxyalkanol moiety (HDP-OH) that was linked to CDV to form HDP-CDV did not inhibit SV40 DNA replication in vitro (Fig. 1A). Also, a 5-Aza substituted CDV derivative (aCDV) and its cyclic derivative acCDV did not inhibit cell-free SV40 DNA replication (Fig. 1B). However, the alkoxyalkyl ester of acCDV, HDE-acCDV (hexadecyloxyethyl acCDV) efficiently inhibited the cell-free SV40 DNA replication (Fig. 1B; summarized in Table 1) with an IC₅₀ of 12.2 ± 0.02 µM, whereas the alkoxyalkanol moiety (HDE-OH) did not inhibit SV40 DNA replication in vitro (Fig. 1B).

FIG 1.

Inhibition of in vitro SV40 DNA replication by nucleotide analogs. SV40 DNA replication was performed using SV40 Tag, HeLa cell replication extract, and SV40 origin-containing plasmid (pUC-HS) in the presence of the analogs in the concentrations indicated and set up as previously described (72, 73). The relative incorporation of dNMP was determined by calculating the percentage ratio of counts of analog-treated samples to the solvent-treated sample. The data were then fit on GraphPad using log[inhibitor] versus normalized response in order to obtain the IC₅₀. (A) The analogs CDV (cidofovir), CDV-P (CDV-monophosphate), CDV-PP (CDV-diphosphate), HDP-CDV (hexadecyloxypropyl-CDV), and HDP-OH (hexadecyloxypropanol). (B) Comparison of the inhibitory activity of CDV, aCDV (5-Aza-CDV), acCDV (5-Aza-cyclic-CDV), HDE-acCDV (hexadecyloxyethyl 5-Aza-cyclic-CDV), and HDE-OH (hexadecyloxyethanol). The untreated/solvent-treated control contained phosphate-buffered saline (PBS), in which the analogs were solubilized, and was arbitrarily set to 100. The data presented are derived from 4 independent experiments.

TABLE 1.

Effects of nucleotide analogs on the polyomavirus DNA replication and T antigen activities summary of the effects of the nucleotide analogs in various biochemical assay^a

Inhibitors	Polyomavirus/protein	In vitro replication (crude extract)	Monopolymerase (purified Proteins)	Helicase	ATPase	DNA binding
						ssDNA	dsDNA
CDV	BKV, JCV and SV40	–	–	–	–	ND	ND
CDV-PP	BKV, JCV and SV40	±	–	–	–	ND	ND
HDP-CDV	SV40	++++	++++	++++	++	++++	++++
	JCV	++++	++++	ND	+	ND	ND
	BKV	++++	++++	++++	+	ND	ND
	RPA	ND	ND	ND	ND	–	ND
HDE-acCDV	SV40	++++	++++	ND	+	ND	ND
	JCV	++++	+++	ND	+	ND	ND
	BKV	++++	++++	ND	+	ND	ND
Mal2-11B	SV40	ND	ND	ND	+	ND	ND

^aAll assays contain the polyomavirus large T antigen as the only viral protein. ++++, IC₅₀ ≤ 25 µM; +++, IC₅₀ ≤ 50 µM; ++, IC₅₀ ≤ 100 µM; +; IC₅₀ ≤ 280 µM; ±, IC₅₀ ≥ 280 µM; –, inhibition similar to solvent added; ND, not determined.

To investigate if these effects observed in SV40 DNA replication in vitro are virus specific, these analogs were further studied in the in vitro replication of human polyomaviruses JC and BK virus (JCV and BKV, respectively) DNA. The results showed that the in vitro DNA replication of JCV and BKV was similarly affected as that of SV40 (Fig. 2A and B, respectively; summarized in Table 1), and thus, the effects are not virus specific. The nucleotide analogs CDV and acCDV inhibited neither JCV (Fig. 2A) nor BKV DNA (data not shown) replication in vitro. Likewise, CDV-P, CDV-PP, aCDV, HDP-OH, and HDE-OH inhibited the in vitro replication of neither JCV nor BKV DNA (data not shown). In contrast, HDP-CDV and HDE-acCDV (which are alkoxyalkyl esters of CDV and acCDV, respectively) efficiently inhibited the in vitro replication of JCV DNA with an IC₅₀ of 12.7 ± 0.08 µM and 13.8 ± 0.1 µM, respectively (Fig. 2A; summarized in Table 1). Similarly, HDP-CDV and HDE-acCDV inhibited the BKV DNA replication in vitro with IC₅₀ values of 19.8 ± 0.07 µM and 19.5 ± 0.08 µM, respectively (Fig. 2B; summarized in Table 1).

FIG 2.

Inhibition of in vitro JCV and BKV DNA replication by nucleotide analogs. DNA replication was performed with JCV and BKV Tag (panels A and B, respectively) using S100 extracts and the respective origin-containing plasmid (pJC433 and pUC-BKVOri, respectively) in the presence of the indicated analogs as previously described (72, 73). (A) The relative incorporation of dNMPs in the presence of the analogs cidofovir (CDV), acCDV, HDP-CDV, and HDE-acCDV with JCV. (B) For BKV, only HDP-CDV and HDE-acCDV were analyzed. The untreated control containing PBS, which was used to solubilize the analogs, was arbitrarily set to 100. The data are the mean and standard deviation of dNMP incorporation derived from 3 independent experiments.

Previous studies have shown that the alkoxyalkyl ester modifications of CDV improved several of the drawbacks of the CDV prodrug by eliminating nephrotoxicity, enhancing its oral bioavailability and cellular uptake, which resulted in an increase in the drugs’ efficacies as broad-spectrum antiviral agents (15, 23–25). In vivo, the alkoxyalkyl ester conjugation enhances the transport of the prodrug across the plasma membrane barrier, and subsequently, the catalytic actions of cellular phospholipase C and kinases catalyze the release of CDV from the alkoxyalkyl ester-CDV and the conversion of the released CDV to CDV-PP in that order (23, 24, 30–32, 47, 48). In contrast to the in vivo conditions, the in vitro experimental conditions under which the nucleotide analogs were investigated in this study could have eliminated the plasma membrane barrier, implying that enhanced cellular uptake does not sufficiently explain the marked increase in inhibition due to alkoxyalkyl-ester conjugation. Additionally, the cellular enzymes required to metabolize these alkoxyalkyl esters to CDV-PP could have been eliminated or are not optimally active in the S100 replication extracts, implying that the alkoxyalkyl ester-conjugated CDV showing highly efficient inhibition of the in vitro replication of polyomavirus DNA is probably independent of these analogs undergoing further conversions. Hence, it is imperative to further investigate these analogs using purified proteins.

In vitro replication of polyomavirus DNA using purified proteins is inhibited by alkoxyalkyl esters of cidofovir.

To investigate these inhibitory activities in more detail, monopolymerase polyomavirus replication assays were performed which utilize a defined number of purified host proteins (Pol-prim, Topo I, RPA) and recombinant viral Tag to replicate polyomaviral DNA. The obtained data revealed that CDV, CDV-P, and CDV-PP did not inhibit the monopolymerase replication assays using SV40 DNA (Fig. 3A; summarized in Table 1). Similar results were found for these nucleotide analogs using the purified JCV and BKV systems (data not shown). Additionally, aCDV, acCDV, HDP-OH, and HDE-OH did not inhibit the monopolymerase replication of SV40, JCV, and BKV DNA (data not shown). In contrast, the alkoxyalkyl ester-modified CDV and acCDV, i.e., HDP-CDV and HDE-acCDV, both efficiently inhibited the monopolymerase replication of polyomavirus DNA in a concentration-dependent manner and independent of the polyomavirus species (Fig. 3A to C; summarized in Table 1). These lipid analogs, HDP-CDV and HDE-acCDV, inhibited monopolymerase replication of SV40 DNA with IC₅₀ of 21 ± 0.05 µM and 15.7 ± 0.09 µM, JCV DNA with IC₅₀ of 22.4 ± 1.08 µM and 28.0 ± 1.16 µM, and BKV DNA with IC₅₀ of 20.1 ± 1.06 µM and 24.9 ± 1.2 µM. These data strongly suggest that the alkoxyalkyl esters of CDV, HDP-CDV, and HDE-acCDV do not need to be further modified, e.g., deesterification and phosphorylation, to be highly inhibitory for polyomavirus DNA replication in vitro.

FIG 3.

Inhibition of monopolymerase replication of SV40, JCV, and BKV DNA by nucleotide analogs. SV40/JCV/BKV monopolymerase DNA replication was performed according to references 71–73 using polyomaviral Tag with the accompanying viral origin-containing plasmids and purified host proteins in the presence of the analogs in the concentrations indicated. (A) The analogs evaluated in the SV40 system were cidofovir (CDV), CDV-P, CDV-PP, HDP-CDV, and HDE-acCDV. (B and C) For the JCV (B) and BKV (C) systems, HDP-CDV and HDE-acCDV were analyzed as indicated. The means and standard deviations of dNMP incorporation are presented. SV40 data were averaged from 3 independent experiments, whereas JCV and BKV data are the averages from 2 independent experiments. The data were fitted using the GraphPad program and the model “log[inhibitor] versus response − variable slope” plus the equation $Y=\frac{100}{1+{10}^{X-\text{LogIC}50}}$. In panel B, the model “log[inhibitor] versus normalized response” and the equations $Y=-3.58+\frac{93.19}{1+{10}^{\left(X-\mathrm{LogIC}50\right)*2.49}}$ and $Y=-7.58+\frac{95.69}{1+{10}^{\left(X-\mathrm{LogIC}50\right)*1.81}}$ for HDV-CDV and HDE-acCDV, respectively, were used. In panel C, the IC₅₀ values were determined with the model “log[inhibitor] versus normalized response” plus the equations $Y=-7.79+\frac{100.91}{1+{10}^{\left(X-\mathrm{LogIC}50\right)*1.24}}$ and $Y=-1.254+\frac{108.25}{1+{10}^{\left(X-\mathrm{LogIC}50\right)*1.58}}$ for HDV-CDV and HDE-acCDV, respectively.

Polyomavirus Tag’s ATPase and DNA helicase activities are inhibited by alkoxyalkyl esters-modified derivatives of cidofovir.

To determine the target of these nucleotide analogs, their effects on the essential enzymatic activities (ATPase and helicase) of polyomavirus Tags were investigated. Here, we determined that similar to the results obtained for the cell-free viral replication assays, CDV and its derivatives, with the exception of alkoxyalkyl ester derivatives, did not inhibit the ATPase activities of SV40, JCV, and BKV Tags (Fig. 4A to C). Similarly, the ATPase activities of these Tags were not sensitive to the alkoxyalkanol moieties alone at any concentration tested (Fig. 4A to C). In contrast, the alkoxyalkyl-modified analogs, HDP-CDV and HDE-acCDV, inhibited the ATPase activities of the three Tags investigated in a concentration-dependent manner (Fig. 4A to C; summarized in Table 1). For comparison, Mal2-11B, a compound known to inhibit the ATPase activity of SV40 Tag, was investigated in parallel and in our experiments inhibited SV40 Tag to a similar extent as previously reported (Fig. 4A; 49, 50). The IC₅₀ for the inhibition of ATPase of SV40 Tag by HDP-CDV, HDE-acCDV, and Mal211B is 64.2 ± 2.1 µM, 263 ± 3 µM, and 175.6 ± 1 µM, respectively.

FIG 4.

Inhibition of the polyomavirus Tag ATPase activity by nucleotide analogs. The ATPase assays were performed as described in Materials and Methods using 240 nM polyomavirus Tag. (A) The inhibition of SV40 Tag ATPase activity by the analogs CDV, CDV-P, CDV-PP, HDP-CDV, HDE-acCDV, and HDP-OH and a known ATPase inhibitor Mal2-11B are shown (50). (B) The JCV Tag ATPase inhibition by CDV, CDV-P, CDV-PP, HDP-CDV, HDE-acCDV, and HDP-OH is presented. Panel C displays the concentration-dependent inhibition of BKV ATPase activity by HDP-OH, CDV-PP, HDP-CDV, and HDE-acCDV. The hydrolysis of ³²P-γ-ATP in the presence of PBS was arbitrarily set to 100%. The presented data are the mean and standard deviation of 3 experiments.

In addition, the helicase activity of polyomavirus Tags was investigated to determine the sensitivity of this enzymatic function to inhibition by nucleotide analogs. The data showed that the helicase activity of the Tags was not sensitive to CDV, CDV-P, CDV-PP, aCDV, acCDV, and HDP-OH (data not shown). However, HDP-CDV inhibited the Tags’ helicase activities in a concentration-dependent manner (Fig. 5A and B; summarized in Table 1). Figure 5A shows a representative gel image of inhibition of SV40 Tag helicase by HDP-CDV. Figure 5B presents inhibition data for SV40 and BKV Tag helicase by HDP-CDV fitted with the model log[inhibitor] versus normalized response and the equation

$$Y=\frac{100}{1+{10}^{X-\text{LogIC}50}}$$

as described above, which showed that the helicase activities of SV40 and BKV Tag were inhibited by HDP-CDV with IC₅₀ values of 13.6 ± 1.03 µM and 18.4 ± 1.1 µM, respectively.

FIG 5.

Effects of nucleotide analogs on the helicase activity of polyomavirus Tags. The helicase activities of the polyomavirus Tags were determined as previously described (83) and measured in the absence or presence of increasing amounts HDP-CDV. The reaction products were separated on nondenaturing polyacrylamide gels, which were subsequently analyzed using a Fuji FLA-5100 phosphor imager. (A) A representative result. Lanes 1, heat denatured (HD) substrate, and 3, no Tag, served as controls for complete denaturing of the substrate and negative control. Lane 2 presents the unwinding activity of 0.6 µg of SV40 Tag without inhibitor, whereas lanes 4 to 10 show the Tag unwinding activity in the presence of increasing amounts HDP-CDV as indicated. The bands were quantified with ImageQuant analysis software (Fujifilm Europe, Düsseldorf, Germany), and the relative DNA unwinding was calculated by determining the percentage ratio of unwound DNA in analog-treated samples to the solvent-treated sample. (B) The presented data are the mean and standard deviation of 3 experiments of the inhibition of SV40 and BKV Tag helicase activities by HDP-CDV. Both curves were fitted using the program GraphPad and the model “log[inhibitor] versus normalized response” plus the equation $Y=\frac{100}{1+{10}^{\left(X-\mathrm{LogIC}50\right)}}$.

Effects of HDP-CDV on Tag and RPA DNA binding activities.

The significance of access of DNA-interacting proteins to DNA in the processes of DNA metabolism cannot be overemphasized. The DNA-binding activity of Tag is essential for its interaction with double-stranded DNA (dsDNA) containing the viral replication origin, for duplex DNA unwinding, initiation of the viral DNA replication, and synthesis of Okazaki fragments during lagging strand DNA synthesis (51, 52). Our findings are in agreement with previous studies and showed that Tag binds with high affinity to single-stranded DNA (ssDNA) and dsDNA derived from SV40 origin sites I and II (51, 52). Here, we first determined the dissociation constants K_D for SV40 Tag binding to ssDNA and dsDNA derived from sites I and II of the SV40 replication origin. The K_D values for Tag binding to ssDNA and dsDNA were 131.9 ± 3.1 nM (Fig. 6A and B) and 120.8 ± 4.7 nM (Fig. 6C and D), respectively.

FIG 6.

ssDNA and dsDNA binding activities for SV40 Tag. The binding of SV40 Tag to ssDNA and dsDNA was analyzed as previously described (51) to optimize Tag concentrations required to investigate the inhibition of Tag-DNA interactions by nucleotide analogs. (A and C) The binding products using (A) ssDNA and (B) SV40 Tag site 2-containing dsDNA were separated on nondenaturing polyacrylamide gels, which were subsequently analyzed using a phosphor imager. The bands were quantified with ImageQuant analysis software, and the bound DNA was calculated by determining the ratio of the bound to total DNA. The data were fitted using GraphPad and the model “one site-specific binding with Hill slope” to obtain the binding constants (panels B and D). The K_D values for Tag binding to ssDNA and dsDNA are 131.9 ± 3.1 nM and 120.8 ± 4.7 nM, respectively.

Subsequently, the effects of HDP-CDV on the binding of Tag to ssDNA and dsDNA were investigated (Fig. 7A to D). The data show that the fatty ester-modified analog strongly inhibited the binding of Tag to ssDNA and dsDNA of sites I and II to similar extents in a concentration-dependent manner, but only figures for site II are shown (Fig. 7A to D; summarized in Table 1). Fitting the data with “log[inhibitor] versus response − variable slope” using the equations

$$Y=-1.09+\frac{92.8}{1+{10}^{\left(X-\mathrm{LogIC}50\right)*3.406}}$$

and

$$Y=-7.83+\frac{96.5}{1+{10}^{\left(X-\mathrm{LogIC}50\right)*2.129}}$$

FIG 7.

Effects of HDP-CDV on the DNA binding activity of SV40 Tag. Following the determination of the binding constants for SV40 Tag binding to ssDNA and dsDNA, the effects of HDP-CDV on ssDNA and dsDNA binding function of Tag were determined via electrophoretic mobility shift assay (EMSA) at the indicated analog concentrations in the presence of 300 nM Tag. The reaction Tag-ssDNA and Tag-dsDNA products were separated on nondenaturing polyacrylamide gels, which were subsequently analyzed using a phosphor imager. (A and C) Representative results. The bands were quantified with ImageQuant software, and the relative DNA binding was calculated by determining the percentage ratio of bound DNA of the HDP-CDV treated samples to the solvent-treated sample. (B and D) The mean and standard deviation from 3 independent experiments are presented. The data were fitted with the model “log[inhibitor] versus response − variable slope” on GraphPad to obtain the binding curves and inhibition constants using the equations $Y=-1.09+\frac{92.8}{1+{10}^{\left(X-\mathrm{LogIC}50\right)*3.406}}$ and $Y=-7.83+\frac{96.5}{1+{10}^{\left(X-\mathrm{LogIC}50\right)*2.129}}$ (panels B and D, respectively). The IC₅₀ values for the inhibition of binding of Tag to ssDNA and dsDNA by HDP-CDV were determined as 17.4 ± 1.02 µM and 22.1 ± 1.04 µM, respectively.

in Fig. 7B and D, respectively, yielded IC₅₀ values of 17.4 ± 1.02 μM and 22.1 ± 1.04 μM for the inhibition of Tag binding to site II-derived ssDNA and dsDNA, respectively, by HDP-CDV. Investigations of the inhibition of the ssDNA binding function of RPA were performed in parallel to determine whether the analog also has broad-spectrum inhibitory effects on cellular DNA-binding proteins (Fig. 8, summarized in Table 1). Figure 8A and B show that HDP-CDV did not inhibit the binding of RPA to ssDNA, unlike Tag binding to ssDNA and dsDNA, which is highly sensitive to the analog.

FIG 8.

HDP-CDV does not influence the DNA binding activity of RPA. (A and B) The effects of HDP-CDV on the binding of RPA to ssDNA were determined via EMSA at the indicated analog concentrations in the presence of 2 nM RPA. The free ssDNA and the RPA-ssDNA complexes were separated on nondenaturing polyacrylamide gels, which were subsequently analyzed using a phosphor imager (panel A). The bands were quantified with ImageQuant software, and the relative DNA binding was calculated by determining the percentage ratio of bound DNA of the HDP-CDV-treated samples to the solvent-treated sample. Then the results from 3 independent experiments were fitted with various nonlinear regression models using GraphPad, but none yielded reliable results. Therefore, a linear regression analysis was performed yielding the trendline with the equation $\mathit{y\; =\hspace{0.17em}}-3.5114\times 112.6{;\hspace{0.17em}r}^2\text{\hspace{0.17em}=\hspace{0.17em}0}.9776$ (panel B).

DISCUSSION

Nucleoside analogs such as ganciclovir and acyclovir that target CMV and HSV, respectively, have been successfully used to inhibit viral replication for over 3 decades but with inherent limitations, e.g., acyclovir metabolism requires virally encoded thymidine kinase to initiate the critical phosphorylation step in its activation (53–56). A new generation of nucleotide analogs, e.g., cidofovir (CDV), have recently been developed to overcome the shortcomings of acyclovir and to be independent of viral thymidine kinases (53–55, 57). CDV has been one of the most successful antiviral drugs to treat HSV. To overcome its inherent limitations of low bioavailability and nephrotoxicity, various forms of modified products of the nucleotide analog have been produced. These modified CDV products include HDP-CDV (Brincidofovir), which can be orally administered, cCDV, acCDV, and HDE-acCDV (14–16, 26, 27, 53–55, 58, 59). Brincidofovir is already in clinical trials as an antiviral drug against CMV and adenovirus (24, 31, 47, 60). To combat diseases associated with human polyomaviruses such as Merkel cell carcinoma, TS, HC, PVAN, and PML, CDV has been used, but the outcomes have been inconclusive due to concurrent withdrawal of immune-compromising conditions, which itself has been a palliative for management of polyomavirus-associated diseases (8, 61–65).

Using HeLa cell replication extracts and monopolymerase DNA replication assays with purified proteins, neither CDV, aCDV, acCDV, nor CDV-P inhibited the cell-free replication of BKV, JCV, and SV40 DNA. These results are expected, as the analogs are prodrugs or the intermediate products that are supposedly not the active metabolites. Additionally, the enzymes required for the activation of the prodrug are lacking in the experimental setup. The evaluation of the effects of the nucleotide analogs on the replication of the primate polyomavirus DNA as performed in this study revealed surprising results. On one hand, the supposedly active metabolite, CDV-PP, only marginally inhibited the in vitro DNA replication of the polyomaviruses investigated. On the other hand, alkoxyalkyl esters of CDV themselves strongly inhibited the in vitro replication of BKV, JCV, and SV40 DNA without exhibiting species specificity. These inhibitions are obtained without the drugs undergoing further chemical modifications, which was validated by using monopolymerase DNA replication that utilizes only purified replication proteins lacking any nucleotide-modifying enzymes. This view was additionally supported by the findings from the studies of CMX029, an R-enantiomer of HDP-CDV, which cannot be phosphorylated in cells, but nonetheless also efficiently inhibited in vitro polyomavirus DNA replication (data not shown). Moreover, our results revealed that in the in vitro experiments, the alkoxyalkyl esters’ inhibitory effects are independent of increased cellular uptake, and they are also not a result of the alkoxyalkanol moieties, because neither HDP-OH nor HDE-OH inhibited the replication of polyomavirus DNA at all concentrations investigated.

The investigation of the analogs with respect to ATPase and helicase activities of Tag reveals that the alkoxyalkyl esters of nucleotide analogs inhibit these activities of the polyomavirus Tags in a concentration-dependent manner. These functions of the Tag are critical for orchestrating the multiple steps involved in viral DNA synthesis (52, 66). The viral Tags possess ATPase activity that catalyzes ATP hydrolysis and dsDNA unwinding function performed by the oligomeric Tag (52, 66). This helicase function is essential for the initiation of DNA replication at the replication origin and for the continuous dsDNA unwinding activity through the elongation phase (52, 66). The inhibitory activity of the fatty ester analogs in various biochemical assays—monopolymerase, in vitro DNA replication, ATPase, helicase, and DNA binding—shows that the analog HDP-CDV most efficiently inhibits viral DNA replication, SV40 Tag’s DNA helicase, and DNA binding functions. In contrast, the amounts of the compound required to reach 50% inhibition of Tag’s ATPase activities is 3 to 5 times higher than those required to achieve 50% inhibition of polyomavirus DNA replication in vitro, whereas the IC₅₀ values describing the decrease of Tag’s ssDNA and dsDNA binding to 50% are similar to those found for the inhibition of viral DNA replication in vitro (SV40 Tag, HDP-CDV: IC_50ssDNA = 17.4 µM and IC_50dsDNA = 22.1 µM versus SV40 DNA replication IC₅₀ = 13.3 µM), suggesting that HDP-CDV inhibits polyomavirus DNA replication in vitro by blocking the binding of Tag to ssDNA and dsDNA. The inhibition of SV40 Tag binding to DNA also effects its helicase activity, and interestingly, the SV40 Tag helicase activity is as sensitive to HDP-CDV (IC₅₀ = 13.6 µM) as the cell-free SV40 DNA replication and DNA binding inhibition (see Table 1). It is important to note that the IC₅₀ values of the SV40 Tag ATPase activity with HDP-CDV in comparison with the IC₅₀ of the SV40 DNA replication, SV40 Tag’s DNA binding, and helicase activity by HDP-CDV are statistically significant, with P values of at least <0.01. Hence, the alkoxyalkyl ester analogs may inhibit polyomavirus DNA replication by targeting viral Tags at the initiation step and during the unwinding of dsDNA. The latter would be inhibited by preventing Tag from binding to ssDNA and thus diminishing its helicase activity. The hypothesis that alkoxyalkyl ester analogs may play a direct role as antiviral agents is supported by the findings that HDP-CDV is relatively stable in S9 fractions from human liver. Incubating 1 µM HDP-CDV with the extract showed that 83.3% was remaining after 90 min, whereas ∼90% of 10 µM HDP-CDV was recovered in these extracts at the same time (67). These studies suggest that further investigations into the inhibition mechanisms of nucleotide analog-based compounds are necessary.

The inhibition of SV40 Tag ATPase has also been reported for other compounds, such as Mal2-11B, which inhibits the enzyme activity with an IC₅₀ of 100 or 200 µM depending on whether the enzyme is preincubated with Mal2-11B (49, 50). A similar inhibition by Mal2-11B was also seen for the SV40 DNA replication in vitro (50). The IC₅₀ of 64.2 µM for the inhibition of SV40 Tag ATPase activity by HDP-CDV is more than 3 times lower than that of Mal2-11B. On the other hand, the IC₅₀ for the inhibition of SV40 Tag ATPase activity is nearly 5 times higher than the IC₅₀ (13.3 µM) for the inhibition of SV40 DNA replication in vitro by HDV-CDV, which was statistically significant with a P value of <0.001. Importantly, the inhibition of SV40 DNA replication is most likely caused by the inhibition of SV40 Tag’s origin dsDNA binding activity and helicase activity (see model in Fig. 9 and the discussion below). Recently, 2nd-generation SV40 Tag ATPase inhibitors such as bithionol and hexachlorophene have been described which inhibit the ATPase activity with IC₅₀ values of about 3 and 2 µM, respectively (49). Interestingly the IC₅₀ of bithionol for inhibition of SV40 DNA replication was much higher than these values (IC₅₀ = ∼20 µM [49]). Considering that the 50% cytotoxicity concentration (CC₅₀) of biothionol is ∼20 µM in human cells, which is in the same range as its IC₅₀ for SV40 DNA replication inhibition (49, 68), and that HDP-CDV has a CC₅₀ value of ∼31 µM (69) and an IC₅₀ of ∼13 µM for SV40 DNA replication as described here, HDP-CDV has an advantage as a polyomavirus replication inhibitor compared to bithionol. These data suggest that fatty ester-linked nucleotide analogs themselves, such as HDP-CDV, are efficient polyomavirus DNA replication inhibitors that directly interact with the viral replication machinery by targeting the polyomavirus Tag helicase and origin-binding activity.

FIG 9.

HDP-CDV inhibits DNA binding activities of SV40 Tags at the origin of DNA replication and the replication forks. The model presents the stages of polyomavirus DNA replication inhibited by fatty ester nucleotide analogs (lipid-ANPs). The lipid-ANPs block the binding of Tag to the viral origin DNA (1), thereby inhibiting establishment of replication forks and the unwinding of duplex DNA by Tag. In addition, lipid-ANPs also diminish the extension of established forks by directly inhibiting the Tag helicase activity on the ssDNA-binding and ATPase side (2) and consequently inhibiting the viral DNA replication.

Proposed model for inhibition of polyomavirus DNA replication by fatty ester-linked nucleotide analogs.

The data presented here suggest that the fatty ester-linked nucleotide analog HDP-CDV bind to the Tag possibly at two sites, the ATPase domain and the origin binding domain (OBD) (for a review of the Tag domains see references 52 and 66). The ATPase domain is essential for dsDNA unwinding but is not as efficiently inhibited as polyomaviral DNA replication in vitro. In contrast, the binding of HDP-CDV to the OBD may block polyomaviral Tags’ competency to access dsDNA and ssDNA, which is essential for Tag to form double hexamers at the viral origin of replication and is critical for the initiation of viral DNA replication and establishment of helicase complexes at replication forks during the elongation steps of viral DNA replication (Fig. 9). Alternatively, since the helicase domain and the OBD are both involved in Tag’s binding to DNA, HDP-CDV may interfere with the cooperation of these two domains in their binding to DNA. It is important to note that the conjugation of the analogs with alkoxyalkyl esters is required for this inhibition because neither the fatty ester moiety alone nor the unconjugated nucleotide analog component are sufficient to inhibit polyomavirus DNA replication in vitro at all concentrations tested. In addition, no further maturation of fatty ester-linked nucleotide analogs to an active compound is required to establish their antiviral activities.

MATERIALS AND METHODS

Nucleotide analogs.

Chimerix Ltd. (USA) supplied CDV, CDV-P, CDV-PP, HDP-CDV, and HDP-OH (Table 2). Additionally, samples of CDV, aCDV, acCDV, HDE-OH, and HDE-acCDV (Table 2) were synthesized as previously described (14, 25–27). Samples of CDV and HDP-CDV were also purchased from Sigma and Insight Biotechnology Limited (UK), respectively. Nucleotide analogs were dissolved in phosphate-buffered-saline (1× PBS; 137 mM NaCl, 2.7 mM KCl, 10 mM Na₂HPO₄, 1.8 mM KH₂PO₄), while alkoxyalkanol moieties alone were dissolved in 100% methanol. All protein purifications, extract productions, and dialysis procedures were performed on ice or at 4°C unless otherwise stated.

TABLE 2.

The structures, names, and sources of compounds^a

^aCidofovir (CDV) is the parent acyclic nucleotide phosphonate, which is a dCMP analog. Cidofovir-monophosphate (CDV-P) and cidofovir-diphosphate (CDV-PP) are, respectively, the putative intermediate and active metabolites of CDV and are analogs of dCDP and dCTP. 5-Aza-cidofovir (aCDV) is the derivative of CDV, in which nitrogen is substituted for carbon at the 5 position of the pyrimidine ring. 5-Aza-cyclic-cidofovir (acCDV) is the cyclic derivative of aCDV. All the above-listed nucleotide analogs are described as non-fatty ester derivatives of CDV. The alkoxyalkyl derivatives are hexadecyloxypropyl cidofovir, with the trade name Brincidofovir (HDP-CDV), derived by linking hexadecyloxypropanol to CDV. Hexadecyloxyethyl -5-aza-cyclic-cidofovir (HDE-acCDV) is also an alkoxyalkyl derivative, which is composed of acCDV linked with hexadecyloxyethanol. Hexadecyloxypropanol (HDP-OH) and hexadecyloxyethanol (HDE-OH) are the alkoxyalkanol moieties. 4-[1,1′-Biphenyl]-4-yl-3,4-dihydro-6-methyl-2-oxo-5-[(phenylmethoxy)carbonyl]-1(2H)-pyrimidinehexanoic acid (Mal2-11) is a known inhibitor of ATP hydrolysis activities of polyomavirus large tumor antigens (50).

Preparation of HeLa S100 extract.

Extracts were prepared according to Stadlbauer et al. (70) with slight modifications. Logarithmically growing HeLa cells were harvested, washed twice with ice-cold PBS and once with ice-cold S100 buffer (20 mM HEPES/KOH, pH 7.8, 5 mM KCl, 100 mM NaCl, 1.5 mM MgCl₂, 0.1 mM DTT) without protease inhibitors. The cells were then incubated in S100 buffer supplemented with protease inhibitor for 5 to 10 min. Subsequently, the swollen cells were gently scraped and Dounce-homogenized with 20 to 25 strokes of a loose pestle. Cell debris were removed by centrifugation at 75,000 × g for 30 min. The supernatant was adjusted to 100 mM NaCl and spun at 100,000 × g for 45 min and then dialyzed against dialysis buffer (20 mM HEPES-HOH, pH 7.8, 100 mM NaCl, 1 mM DTT, 10% glycerol, 1× protease inhibitor cocktail [4-(2-aminoethyl)benzenesulfonyl fluoride hydrochloride (AEBSF), aprotinin, bestatin, E-64, EDTA, leupeptin; purchased from Merck] for 1 h and flash-frozen in drops in liquid nitrogen for storage at −80°C.

Purification of SV40, JCV, and BKV large tumor antigens.

Expression and purification of SV40, JCV, or BKV Tag was performed as previously described with slight modifications (71, 72). Logarithmically growing High Five insect cells were infected with 10 PFU of the recombinant baculovirus expressing SV40, JCV, or BKV Tag and incubated at 27°C as previously described (71, 72). Cells were harvested 44 h postinfection, and the Tag was purified according to Nesper et al. and Tikhanovich et al. (72, 73) with slight modifications. Cell pellets were resuspended in lysis buffer (100 mM Tris-HCl, pH 7.5, 100 mM NaCl, 5 mM KCl, 0.5 mM MgCl₂, 0.05% NP-40, 1× SIGMAFAST protease inhibitor cocktail [AEBSF, bestatin, E-64, pepstatin A, phosphoramidon, leupeptin, aprotinin from Merck]) and subsequently Dounce-homogenized with 20 strokes of a tight pestle. The homogenate was cleared at 30,000 × g for 30 min. The supernatant was incubated for 1.5 h with 1 ml bed volume of either nickel or TALON affinity resin that was preequilibrated in wash buffer 1 (20 mM Tris/HCl, pH 8.0, 100 mM NaCl, 3.5 mM 2-mercaptoethanol). The resin was then washed once with 10 volumes of wash buffer 1, twice with 5 volumes of wash buffer 2 (wash buffer 1 plus 0.5% Thesit), and twice with 10 volumes of wash buffer 3 (wash buffer 1 plus 5 mM imidazole). Protein was eluted in elution buffer (wash buffer 1 plus 500 mM imidazole) and dialyzed overnight against dialysis buffer (10 mM HEPES/KOH, pH 7.8, 5 mM NaCl, 1 mM dithiothreitol [DTT], 0.1 mM EDTA, 30% glycerol).

DNA polymerase α-primase expression and purification.

Logarithmically growing High Five cells were infected with 10 PFU of each baculovirus stock expressing one of the subunits of human DNA polymerase α-primase (Pol-prim) (p180, p70, p58, and p48) and harvested 48 h postinfection (74, 75). The Pol-prim was purified as previously described (74–76) with slight modifications. Cell pellets were resuspended in extraction buffer (50 mM Tris/HCl, pH 7.5, 100 mM NaCl, 5 mM KCl, 0.5 mM MgCl₂, 0.1% NP-40, 2 mM EDTA, 1 mM DTT, 0.5 mM phenylmethylsulfonyl fluoride [PMSF]) and Dounce-homogenized with 20 strokes of a tight pestle, and cell debris were cleared by centrifugation at 20,000 × g. The supernatant was incubated for 1 h with a 10-ml bed volume of phosphocellulose (PC11) preequilibrated with PC-wash buffer (100 mM KP_i, pH 7.2; 1 mM EDTA, 2 mM 2-mercaptoethanol). The resin was washed three times with PC-wash buffer and eluted in elution buffer (500 mM KP_i, pH 7.8, 1 mM EDTA, 1 mM Na₂S₂O₅, 2 mM 2-mercaptoethanol). Following 2 h of dialysis in PC-dialysis buffer (50 mM KP_i, pH 7.5, 1 mM EDTA, 1 mM Na₂S₂O₅, 2 mM 2-mercaptoethanol), the eluate was incubated for 1 h with antibody (SJK 237-71)-coupled Sepharose 4B that was preequilibrated in immunoaffinity (IA) wash buffer A (150 mM KP_i, pH 7.5, 0.1 mM EDTA). The resin was washed twice each with IA wash buffer A and IA wash buffer B (20 mM KP_i, pH 7.5, 0.1 mM EDTA). Protein was eluted in IA elution buffer (10 mM triethylamine, 20% ethylene glycol, 1 M NaCl) into tubes containing titration buffer (0.5 M KH₂PO₄, 0.1 M DTT) and dialyzed overnight in storage buffer (20 mM HEPES/KOH, pH 7.8, 1 mM EDTA, 1 mM DTT, 50% glycerol). Specific activity was determined using a DNA synthesis assay containing activated DNA as the substrate in an assay performed as previously described (74, 76, 77).

His-Tagged topoisomerase I expression and purification.

The protein was expressed with recombinant baculovirus in High Five cells and harvested 48 h postinfection (78). The protein was purified according to Soe et al. (78) with slight modifications. The pellets were resuspended in 4 packed-cell volumes (PCV) of lysis buffer (10 mM Tris/HCl, pH 8.0, 1.0 mM EDTA, 5 mM DTT, 1 mM PMSF, 5 mM leupeptin, 1% aprotinin) and Dounce-homogenized (30 strokes with a tight pestle). Then, 4 PCV of sucrose buffer (50 mM Tris/HCl, pH 8.0, 10 mM MgCl₂, 50% [vol/vol] glycerol, 25% [wt/vol] sucrose, 2 mM DTT) was added, followed by addition of 1 PCV saturated ammonium sulfate solution in drops. The mixture was then incubated on ice for 30 min and centrifuged at 35,000 × g for 3 h. Subsequently, 4× supernatant volumes of saturated ammonium sulfate were added to the supernatant in drops with continuous gentle stirring and incubated further for 30 min. Precipitate was collected at 20,000 × g for 30 min and redissolved in wash buffer (30 mM HEPES/KOH, pH 7.8, 150 mM NaCl, 20 mM imidazole, 10% [vol/vol] glycerol, 1 mM 2-mercaptoethanol, 1 mM PMSF). Subsequently, the redissolved precipitate was incubated for 2 h with a 1-ml bed volume of nickel resin preequilibrated with wash buffer. The resin was washed twice with 10 volumes of wash buffer and eluted in elution buffer (30 mM HEPES-KOH, pH 7.8, 150 mM NaCl, 250 mM imidazole, 10% [vol/vol] glycerol, 1 mM 2-mercaptoethanol, 1 mM PMSF). Fractions were dialyzed overnight in dialysis buffer (15 mM HEPES/KOH, pH 7.8, 170 mM NaCl, 50% [vol/vol] glycerol, 0.5 mM DTT, 0.5 mM EDTA), and protein activity was determined with a DNA relaxation assay performed according to references 78 and 79.

Expresssion and purification of human replication protein A.

Replication protein A (RPA) was expressed in BL21(DE3) pLysS cells using the vector pET15-p14-p32-p70 in 1 liter of lysogeny broth (LB) supplemented with 100 µg/ml of ampicillin by inoculating the culture with 5 to 8 colonies of freshly transformed Escherichia coli and incubated overnight without shaking at room temperature as previously described in reference 80. On day 2, the overnight culture was supplemented with an additional 100 µg/ml ampicillin and grown with shaking at 37°C. At an optical density (OD) 0.5, RPA expression was induced with 1 mM IPTG for 3 h. Purification was done as previously described with slight modifications (80–82). The pellets were resuspended in lysis buffer (30 mM HEPES/KOH, pH 6.8, 0.5% NP-40, 0.5% inositol; 150 mM NaCl, 3 mM 2-mercaptoethanol) and sonicated at 50% amplitude for 3 min with 20-s pulses and incubated on ice for 30 min. EDTA-free protease inhibitor (SIGMAFAST protease inhibitor cocktail [AEBSF, bestatin, E-64, pepstatin A, phosphoramidon, leupeptin, aprotinin from Merck]) was added, and the cell debris was cleared by centrifugation at 30,000 × g for 20 min. The supernatant was incubated for 1.5 h with a 1-ml bed volume of nickel affinity resin preequilibrated with wash buffer 1 (30 mM HEPES/KOH, pH 6.8, 0.5% NP-40, 0.5% inositol, 250 mM NaCl, 3 mM 2-mercaptoethanol). The resin was washed twice each with wash buffer 1 and 10 column volumes of wash buffer 2 [100 mM (NH₄)₂SO₄, pH 8.0, 1% Triton X-100, 5 mM imidazole, 3 mM 2-mercaptoethanol]. The resin was further washed with 3 column volumes of wash buffer 3i [100 mM (NH₄)₂SO₄ pH 8.0, 3 mM 2-mercaptoethanol, 7.5 mM imidazole] and 1.5 column volumes of wash buffer 3ii (3i plus 10 mM imidazole) and eluted in elution buffer (3i plus 100 mM imidazole). The fractions were dialyzed overnight in dialysis buffer (30 mM HEPES/KOH, pH 7.8, 20 mM KCl, 30% [vol/vol] glycerol) and further purified using the ÄKTA protein purifier and Mono-Q column (both GE Healthcare). Precisely, the dialyzed fractions were centrifuged and loaded onto a Mono-Q column preequilibrated with the column equilibration buffer (30 mM HEPES/KOH, pH 7.8, 1 mM DTT, 0.25% inositol, 0.01% NP-40, 100 mM KCl). After this, the column was washed with 3 column volumes of wash buffer (equilibration buffer plus 150 mM KCl), and protein was eluted in elution buffer (equilibration buffer plus 400 mM KCl). The fractions were concentrated with Vivaspin (5 kDa molecular weight cutoff [MWCO]) and stored in 30 mM HEPES/KOH, pH 7.8, 20 mM KCl, and 30% (vol/vol) glycerol. Functionality was determined by DNA binding activity in an electrophoretic mobility shift assay (EMSA) as previously described (51, 81).

Polyomavirus DNA replication assay.

The replication of polyomavirus DNA was performed in vitro as previously described Mahon et al. (71) in 40 µl reaction volume containing 20 mM HEPES/KOH, pH 7.8; 7 mM MgAc; 1 mM EGTA; 1 mM DTT; 4 mM ATP; 200 µM each CTP, UTP, and GTP; 50 µM dCTP; 100 µM each dATP, dTTP, and dGTP; 40 mM creatine phosphate, pH 7.8; 40 µg/ml creatine kinase; 0.083 µCi/μl [α-³²P]-dCTP, equivalent to 27.8 nM [α-³²P]-dCTP (Perkin Elmer); 100 ng pUC-18 (plasmid lacking viral origin sequences) or pUC-Ori plasmid containing the corresponding replication origin as indicated; 75 to 150 µg total protein of HeLa replication extract; and 15 or 10 µg/ml Tag. The reaction mixture was incubated for 60 min at 37°C, and products were precipitated for 10 min with ice-cold 10% (wt/vol) trichloroacetic acid (TCA) containing 0.2% (wt/vol) sodium pyrophosphate on microfiber glass discs and washed three times for 5 min each with 1 M ice-cold HCl. The products were analyzed by scintillation counting. When inhibitors/drugs were used, they were added before the addition of Tag.

Monopolymerase asssay.

The monopolymerase assay was performed according to Mahon et al. (71) in a 40-µl reaction volume that contained 30 mM HEPES-KOH, pH 7.8; 7 mM MgAc; 0.1 mM EGTA; 0.5 mM DTT; 200 μM each UTP, GTP, and CTP; 4 mM ATP; 100 μM each dATP, dGTP, and dTTP; 10 or 50 μM dCTP; 40 mM creatine phosphate; 25 μg/ml creatine kinase; 100 μg/ml bovine serum albumin (BSA); 0.1 μCi/μl [α-³²P]-dCTP, equivalent to 33.3 nM [α-³²P]-dCTP (Perkin-Elmer); 12.5 μg/ml pUC-18 or pUC-origin DNA as indicated; 1 μg/ml topoisomerase I; 0.25 units of Pol-prim; 12.5 μg/ml RPA; and 15 μg/ml Tag. The reaction mixture was incubated at 37°C for 60 min, after which the DNA was precipitated on 47-mm circular microfiber filters in ice-cold 10% (wt/vol) TCA plus 0.2% (wt/vol) sodium pyrophosphate. Subsequently, the filters were washed three times for 5 min each in ice-cold 1 M HCl and dried, and incorporation of the radioactive nucleotide was determined by scintillation-counting using TriCarb scintillation counter (Perkin Elmer).

ATPase assay for polyomavirus T antigens.

The ATPase assay was performed with slight modifications according to Scheffner et al. (83) in a 20-μl reaction volume that comprised 1× ATPase buffer (50 mM Tris/HCl, pH 7.5, 7 mM MgCl₂, 0.5 mM DTT 0.05% NP-40, 10 mM NaCl) plus 0.5 mM ATP and 6.6 nM [γ-³²P]-ATP and 30 µg/ml of the relevant Tag. The reaction was carried out at 37°C for 30 min, stopped by spotting 1 or 1.5 μl of the reaction on a polyethyleneimine (PEI) cellulose thin-layer chromatographic plate (Merck), and resolved in a chromatographic tank with 0.75 mM NaH₂PO4 as the mobile phase. The plate was dried and exposed to phosphor imager screens and subsequently analyzed using a Fujifilm FLA-5100 imager. The bands were quantitated using ImageQuant software (Fujifilm).

Preparation of DNA substrate for helicase and DNA binding activity assays.

For the helicase assay, a 22-mer oligonucleotide primer (5′-GTA ACT TTT CCC AGC CTC AAT C-3′) complementary to the ΦX174 virion DNA sequence was end-labeled using T4 polynucleotide kinase (New England Biolabs) in a 20-μl reaction volume that contained T4 polynucleotide kinase reaction buffer (70 mM Tris/HCl, pH 7.6, 10 mM MgCl₂, 5 mM DTT), 0.5 μM DNA, 2.5 μCi/μl [γ-³²P]-ATP equivalent to 0.83 μM [γ-³²P]-ATP (Perkin Elmer), and 1 U of polynucleotide kinase and incubated at 37°C for 10 min as previously described (51). The end-labeled primer was annealed to ΦX174 virion DNA in a 20-μl reaction volume that contained equimolar concentrations of the virion DNA and end-labeled primer in 1× New England Biolabs (NEB) buffer 2 (50 mM NaCl, 10 mM Tris/HCl, pH 7.9, 10 mM MgCl₂, 1 mM DTT) by heating the mixture to 80°C for 5 min and was cooled slowly to room temperature. Afterward, the unincorporated radioactive nucleotide and unannealed primer were removed using Sephadex G-50 spin columns (GE Healthcare) following the manufacturer’s instructions. For the DNA binding substrates, the SV40 origin sites II- and I-derived primers (51) were similarly labeled, annealed in case of dsDNA substrates, and cleaned up as described above.

DNA helicase assay using polyomavirus T antigens.

The DNA helicase assay protocol was adapted from Scheffner et al. (83) with slight modifications and was performed in a 20-µl reaction volume that contained 25 mM HEPES/KOH, pH 7.8, 7 mM MgAc, 20 mM NaCl, 1 mM EDTA, 0.5 mM DTT, 250 µg/ml BSA, 40 µg/ml creatine kinase, 40 mM creatine-phosphate, 4 mM ATP, 2.5 ng helicase substrate, and 30 µg/ml of BKV, SV40, or JCV Tag. The reaction was carried out at 37°C for 30 min and stopped by adding DNA loading buffer. The products were separated on 12% native polyacrylamide gel in 1× Tris-borate-EDTA (TBE) buffer (89 mM Tris-borate, 2 mM EDTA, pH 8.0). The gel was exposed for autoradiography to Fuji imager screens at −20°C and subsequently developed using a Fujifilm FLA-5100 imager. The band signals were quantitated using ImageQuant software (Fujifilm).

Electric mobility shift assay.

The DNA binding was performed for the Tags and RPA as previously described with slight modifications (51, 81). Precisely, the radiolabeled ssDNA or dsDNA was incubated at room temperature in HEPES binding buffer (30 mM HEPES-KOH, pH 7.8, 200 mM NaCl, 0.5 mM DTT, 4 µg/ml BSA) for 30 min in a 20-µl reaction volume, followed by the addition of DNA loading buffer, and the protein-nucleic acid complexes were analyzed in an electrophoretic mobility shift assay (EMSA) using a 6.5% native polyacrylamide gel in 1× TBE. The gel was then exposed overnight to imager screens (Fujifilm), and the image was captured using a FLA-5100 phosphor imager (Fujifilm) for autoradiography detection. Multi Gauge Image Analyzer (Fujifilm) was used to quantitatively determine band intensities.

Data analysis and models used for fitting of dose response data.

All data analyses that involved calculations were performed with Prism GraphPad and SigmaPlot. Models which fitted the best and gave similar IC₅₀ with GraphPad and SigmaPlot were selected. The model “log[inhibitor] versus normalized response” and the equation

$$Y=\frac{100}{1+{10}^{X-\text{LogIC}50}}$$

were used to fit the data unless otherwise stated. DNA binding data were analyzed using the model “one site-specific binding with Hill slope” and the equation

$$Y=\text{Bmax}*X^h/({\text{Kd}}^h+X^h)$$

Statistical comparisons were performed using paired Student’s t test. A P value of <0.05 was considered statistically significant.