Base-modified thymidine and thymine analogs with low cytotoxicity effectively obstruct DNA replication in papovaviridae

Background: Current chemotherapeutic antimetabolites often exhibit severe side effects that limit their use as drugs; therefore, we designed nucleoside compounds with mechanisms of action focusing on inhibiting DNA replication rather than targeting multiple pathways. We previously discovered cytotoxic basemodified thymidine and thymine analogs that show higher selectivity against cancerous versus normal cells compared to the current antimetabolites used in cancer chemotherapy. We anticipated these antimetabolites have the potential to effectively inhibit viral DNA replication while showing low cytotoxicity. Methods: Base-modified thymidine and thymine analogs were synthesized and their anti-viral activity was evaluated in human cells infected with human pappiloma, John Cunningham, and BK viruses using quantitative DNA polymerase chain reaction assay. In addition, their toxicity toward host cells was determined using CellTiter-Glo assay, and compared to cytotoxicity toward human breast cancer cells. Results: Novel lead compounds with high activity against human papilloma (HPV) and John Cunningham (JCV) viruses have been identified. Their EC50 values lie in low micromolar range (1-2 μM), which is significantly less than that of cidofovir (9-10 μM), a current drug used against DNA viruses. Cytotoxicity of the leads toward the host cells was found to be in 200-300 μM range, which is generally higher than that observed toward MCF-7 human breast cancer cells. None of the tested compounds significantly inhibited BK viral DNA replication. Conclusion: The lead compounds affect the viruses substantially more selectively than the host cells, which makes them a novel class of bioactive compounds with the potential to become effective anti-viral drugs.


Introduction
Papovaviridae is a family of DNA viruses that are associated with serious diseases in patients with compromised immune system. Of those, human papillomavirus (HPV) is the most common sexually transmitted infection worldwide [1]. The high risk types HPV (e.g., 16,18) cause cervical cancer [2][3][4], while the low risk types HPV (e.g., 6,11) incur respiratory diseases, sometimes fatal, in spite of the current treatment [5]. John Cunningham virus (JCV) is known to cause progressive multifocal leukoencephalopathy [6], which is usually deadly. BK virus, a close analog of JCV, is implicated in nephropathy [7] in renal transplant patients. Currently, there are no effective drugs that inhibit or cure these viral infections without off-target toxicity. Therefore, developing novel therapeutic agents against these viruses with low cytotoxicity can be lifesaving.
Recently, we have discovered base-modified thymidine analogs that show higher cytotoxicity toward cancerous cells compared to benign [27]. Although the modifying moiety attached to the nucleobase obstructs further DNA synthesis upon incorporation of the 5'-triphosphate of such nucleoside into a partial double helix DNA primer [28], the preliminary insight into the mechanism of action strongly suggested inhibition of a DNA polymerase instead [27]. However, it was still evident that termination of DNA synthesis occurs only in the presence of a certain bulky group attached at the 5-methyl group of the thymine nucleobase. Considering the high selectivity of the novel species toward cancer cells versus normal cells, we have decided to investigate the possibility of selective targeting of viral DNA replication versus killing the host cell by the newly developed nucleoside analogs. In this paper, we report the evaluation of 16 thymidine and 15 thymine analogs (Figure 1) with modifications at the 5-methyl group in their ability to obstruct DNA replication of HPV and JC viruses, and compare the antiviral activity to the toxicity toward the host cells. The lead compounds identified from these structure-activity relationship studies for both viruses exhibit diminutive cytotoxicity, which opens the new passage to further drug development. Our contribution to the field of anti-viral drug discovery is significant as it gives the promise to access new agents for these viral diseases that currently have no non-toxic drugs available for treatment.

Synthesis
All chemicals, reagents, and solvents were purchased from Sigma-Aldrich Inc., TCI, and Fisher Scientific, Inc., and used as received unless stated otherwise. All reactions were carried out under an atmosphere of dry argon in oven-dried glassware. Indicated reaction temperatures refer to those of the reaction bath, while room temperature (rt) is noted as 25°C. Pure reaction products were typically dried under high vacuum in the presence of phosphorus(V) oxide. Analytical thin layer chromatography (TLC) was performed using glass backed silica plates (5x20 cm, 60 Å, 250 μm). Visualization was accomplished using a 254 nm UV lamp. 1 H and 13 C NMR spectra were recorded on either a Bruker Avance 400 MHz or Bruker DPX 500 MHz spectrometer using solutions of samples in either of the deuterated solvents: DMSO, methanol, acetonitrile. Chemical shifts are reported in ppm with tetramethylsilane as standard. Data are reported as follows: chemical shift, number of protons, multiplicity (s=singlet, d=doublet, dd=doublet of doublet, t=triplet, q=quartet, b=broad, m=multiplet, abq=ab quartet), and coupling constants. High resolution mass spectral data were recorded on a Shimadzu Q-TOF 6500 instrument. All novel compounds were characterized by 1 H, 13 C, DEPT 13 C NMR spectroscopy and high resolution mass spectrometry. The identity of previously made nucleoside derivatives was confirmed by comparison of their 1 H NMR to the published data (reference provided). HPLC analysis of final products was performed on an Agilent 1200 HPLC with UV detection. Compounds biologically tested were at least 95% pure as judged by 1 H NMR and HPLC. (1) [28] and appropriate alcohol (4-20 eq.) were heated neat at 110-120 o C for the period between 15 minutes to 2 hours under argon atmosphere. The mixture was cooled down to room temperature, dissolved in tetrahydrofuran (ca 5 ml), and to this solution chilled at 0 o C tetra-n-butylammonium fluoride trihydrate (TBAF) was added (ca 2.5 eq.). The reaction mixture was stirred for 2 hours while gradually warming up to room temperature. The solvent was removed under reduced pressure and the residue was purified by silica gel (chloroform/methanol=1:0 to 10:1) and then by C8 reverse-phase column chromatography (water/methanol=19:1 to 1:4) to yield the product as a waxy solid. Compounds 3a-11a and 14a-18a were obtained as reported previously [27].

Quantitative PCR analysis
Human embryonic kidney cells HEK 293 were infected with human papilloma (strain: HPV 11 ), human foreskin fibroblast HFF cells with BK (strain: Gardner), and human embryonic kidney 293TT cells with John Cunningham (strain: MAD -1 ) viruses. The infected cells were treated with DMSO solutions of test compounds and cidofovir, a known anti-viral drug (®Vistide), as positive control at concentrations 0, 0.1, 1, 10, 50, 100, 200, and 300 μM. The cells were incubated for 72 hours. After incubation, the cells were harvested and their DNA was isolated followed by quantitative real-time PCR analyses using FastStart Universal Probe Master (ROX) (Roche Applied Science). PCR doi: 10.7243/2052-9341-4-3 was performed with an initial denaturation reaction at 95 o C for 1 minute and then amplified with 40 cycles of 95°C for 30 seconds, 60°C for 30 seconds, and 72°C for 30 seconds. The amplification was monitored on Step One Plus (Applied Biosystems Inc.). A 5 μL aliquot of the resulting solution was loaded into each well containing a 12% denaturing PAGE gel, which was subsequently run at constant 18 Watts for 35 minutes. The gel was visualized using the Odyssey Infrared Imaging System (LiCor) with 169 μm resolution and the 700-channel laser source which has a solid-state laser diode at 680 nm and ImageQuant 5.0 software was used to determine density measurements. Experiments were performed in triplicate. EC 50 and EC 90 values were determined by comparison to the vehicle (negative control). The concentrations of compounds causing 50% and 90% inhibition of HPV and JCC viral DNA replication are summarized in Tables 1 and 2, respectively.

Cell viability
The CellTiter-Glo® Luminescent Cell Viability Assay was used. Upon incubation, cells were treated with solution of beetle luciferin in the presence of ATP. Luminescence was recorded 10 minutes after reagent addition using a GloMax®-Multi+Detection System. The luminescent signal from the host cells was compared to the background signal from serum-supplemented medium without cells. The CC 50 curves were determined by plotting viability versus compound concentration. Kaleidagraph software was used to calculate the R value for each logarithmic curve fitting. The results are outlined in Tables  Cidofovir (control) 9.49 >300 >300 >32 1

Results and discussion
The bioactive compounds were obtained by heating 5-bromomethyl-3-N-(tert-butyloxy)carbobyl-3' ,5'-bis-(tert-butyl) dimethylsilyl-O-2'-deoxyuridine (1) [28] or 5-chloromethyluracil (2) with an appropriate alcohol under neat, anhydrous conditions, as reported previously [27]. This transformation yielded nucleobases 3b-18b; removal of the residual TBS groups using tetra-n-butylammonium fluoride yielded nucleoside derivatives 3a-18a (Figure 1). The PCR DNA and CellTiter-Glo assays were performed using standard protocols. The data reflecting quantitative inhibition of human papilloma virus replication is summarized in Table 1. The trends revealed by structure-activity relationship are quite different from those observed for cytotoxicity toward cancer cells [27]. First, the α-tert-butyl and 2-nitrobenzyl groups were not the best substituents, as evidenced by the higher EC 50 of 3a compared to 10a, as well as 4a versus 11a, and 4a versus 5a. Furthermore, when the methoxy group in the benzyl substituent in 5a was moved across the ring, the activity of the meta-substituted analog (6a) was decreased, while restored for the para-substituted analog 7a, which is the opposite of the anti-cancer activity trend [27]. Second, the presence of an aryl group as either R 1 or R 2 was not critical, as evidenced by the high activity of 16a. And third, modified nucleobases demonstrated superior activity compared to the corresponding base-modified nucleosides in greater number of instances, particularly, 6b compared to 6a, 12b to 12a, and 14b to 14a. The most impressive potency, however, was demonstrated by the nucleoside derivative 10a, 5-(α-cyclohexylbenzyloxy) methyl-2'-deoxyuridine, with the 90% efficiency (EC 90 )value of 2.4 µM and the 50% cytotoxicity (CC 50 ) value exceeding 300 μM. The outstanding SI 90 value of over 125 makes 10a a definite hit compound for further drug development, given significant side effects of cidofovir [30,31] whose potency is lower. Expectedly, cytotoxicity of almost all the examined antimetabolites toward host cells is significantly lower than that toward breast cancer cells [29].
The trend discovered for inhibition of JC viral DNA replication was somewhat different ( Table 2). The three most active compounds turned out to be 5-(α-tert-butylortho-bromobenzyloxy) methyl-2'-deoxyuridine 9a and 5-(α-methylbenzyloxy)methyluracil 18b, showing, respectively, moderate and low activity against HPV DNA replication (Table 1). Notably, nucleoside 18a showed 14 fold lower potency compared to the nucleobase, which is unprecedented in the anti-cancer activity of this type of species [27,29]. Surprisingly, the chloro-analog of 9a, compound 8a, was only somewhat active, and the activity of 10a, the lead compound for HPV, was rather modest against JCV. Neither of these nucleoside and nucleobase derivatives significantly inhibited DNA replication of the BK virus (see supporting information).

Conclusions
We have examined the activity of 5-substituted thymidine and thymine derivatives with respect to inhibition of DNA replication in human papilloma, John Cunningham, and BK viruses. Studies of the structure-activity relationship revealed hits for HPV and JCV, but none against BKV. Importantly, these bioactive species with high anti-viral activity have low cytotoxicity, which makes them novel lead compounds with the