Inhibition of Herpesvirus Replication by Hexadecyloxypropyl Esters of Purine- and Pyrimidine-Based Phosphonomethoxyethyl Nucleoside Phosphonates

Abstract

Patients infected with human immunodeficiency virus (HIV) often suffer from herpesvirus infections as a result of immunosuppression. These infections can occur while patients are receiving antiretroviral therapy, and additional drugs required to treat their infection can adversely affect compliance. It would be useful to have antivirals with a broader spectrum of activity that included both HIV and the herpesviruses. We reported previously that alkoxyalkyl ester prodrugs of cidofovir are up to 3 orders of magnitude more active against herpesvirus replication and may be less toxic than the unmodified drug. To determine if this strategy would be effective for certain phosphonomethoxyethyl nucleoside phosphonates which are also active against HIV infections, the hexadecyloxypropyl (HDP) esters of 1-(phosphonomethoxyethyl)-cytosine, 1-(phosphonomethoxyethyl)-5-bromo-cytosine (PME-5BrC), 1-(phosphonomethoxyethyl)-5-fluoro-cytosine, 9-(phosphonomethoxyethyl)-2,6-diaminopurine (PME-DAP), and 9-(phosphonomethoxyethyl)-2-amino-6-cyclopropylaminopurine (PME-cPrDAP) were evaluated for activity against herpesvirus replication. The HDP esters were substantially more active than the unmodified acyclic nucleoside phosphonates, indicating that esterification with alkoxyalkyl groups increases the antiviral activity of many acyclic nucleoside phosphonates. The most interesting compounds included HDP-PME-cPrDAP and HDP-PME-DAP, which were 12- to 43-fold more active than the parent nucleoside phosphonates against herpes simplex virus and cytomegalovirus, and HDP-PME-cPrDAP and HDP-PME-5BrC which were especially active against Epstein-Barr virus. The results presented here indicate that HDP-esterified acyclic nucleoside phosphonates with antiviral activity against HIV also inhibit the replication of some herpesviruses and can extend the spectrum of activity for these compounds.

Nucleoside phosphonate antiviral agents have been approved for the treatment of human cytomegalovirus (HCMV), hepatitis B virus, and human immunodeficiency virus (HIV) infections and are under development as therapies for hepatitis C virus infections (11). Many members of this versatile class of nucleotide analogs, such as 3-hydroxy-2-(phosphonomethoxy)propyl-cytosine (cidofovir [CDV]), are highly effective against many different viruses, and some are well tolerated in humans (12). However, certain nucleoside phosphonates, such as CDV, concentrate in the proximal tubules of the kidney, resulting in dose-limiting nephrotoxicity (10). Alkoxyalkyl ester prodrugs of CDV, such as hexadecyloxypropyl-CDV (HDP-CDV; CMX001), have been synthesized that are orally bioavailable, do not concentrate in the kidney, and do not appear to cause nephrotoxicity (1, 8). Alkoxyalkyl esterification of CDV also improved the efficacy of the compound by almost 3 orders of magnitude against HCMV in cell culture (30) and was also effective when administered orally to mice infected with HCMV or murine cytomegalovirus (MCMV) (5, 18). Similar improvements in efficacy were also observed with this drug against the other human herpesviruses, orthopoxviruses, adenoviruses, and BK polyomavirus (3, 15, 19, 27, 30).

This approach also proved to be successful when other phosphonomethoxyethyl purines and pyrimidines were converted to HDP, octadecyloxyethyl (ODE), and oleyloxyethyl esters and resulted in significant improvements in their efficacy against HIV, as well as in a thousandfold improvement in the efficacy for HDP 9-(phosphonomethoxyethyl)-2-amino-6-cyclopropylaminopurine(HDP-PME-cPrDAP) and HDP 9-(phosphonomethoxyethyl)-2,6-diaminopurine (HDP-PME-DAP) (28). Both of these compounds were highly active and inhibited the replication of HIV at concentrations of 10 to 16 pM. We hypothesized that the esterification of these compounds may also improve their efficacy against other viruses and evaluated their activity against seven selected human herpesviruses. The results presented here confirmed that the antiviral activities of both HDP-PME-cPrDAP and HDP-PME-DAP were improved significantly against many of the human herpesviruses. The increases in efficacy for these compounds were sufficiently large that they effectively expanded the spectrum of activity for the compounds. Thus, these or other esterified nucleoside phosphonates would be predicted to be useful in therapy for HIV infections and might also provide a measure of protection against opportunistic herpesvirus infections in this population.

MATERIALS AND METHODS

Chemistry methods.

Syntheses of the following compounds and their structures were described previously (28): 1-(phosphonomethoxyethyl)-cytosine (PME-C), HDP-PME-C, 1-(phosphonomethoxyethyl)-5-bromo-cytosine (PME-5BrC), HDP-PME-5BrC, 1-(phosphonomethoxyethyl)-5-fluoro-cytosine (PME-5FC), HDP-5FC, PME-DAP, HDP-PME-DAP, PME-cPrDAP, and HDP-PME-cPrDAP. Briefly, PME-C, PME-5BrC, and PME-5FC were prepared by a reaction of diisopropyl 2-chloroethoxymethylphosphonate with the appropriate cytosine derivative as reported previously by Holy and colleagues (16). After hydrolysis of the diisopropyl esters (trimethylsilyl bromide, H₂O), the phosphonic acids were esterified with 3-hexadecyloxy-1-propanol using the coupling reagent 1,3-dicyclohexylcarbodiimide (25). To obtain the diaminopurine analogs, 2,6-diaminopurine-9-(2-hydroxyethyl)purine and 2-amino-6-cyclopropylaminopurine-9-(2-hydroxyethyl)purine were prepared by a reaction of ethylene carbonate with the appropriate bases and then alkylated with hexadecyloxypropyl or diethyl p-toluenesulfonyloxymethylphosphonate (25).

Cells and viruses.

Human foreskin fibroblast (HFF) cells were prepared by methods reported previously (30). The viruses used to assess antiviral activity were the E-377 strain of herpes simplex virus type 1 (HSV-1), the MS strain of HSV-2, the Ellen strain of varicella-zoster virus (VZV), the Smith strain of MCMV, and the AD169 strain of HCMV. All isolates were obtained from the American Type Culture Collection (ATCC, Manassas, VA). P3HR-1 cells used to produce stocks of Epstein-Barr virus (EBV) and Daudi cells were also obtained from ATCC. HSB-2 cells and the GS strain of human herpesvirus 6 variant A (HHV-6A) were obtained from the NIH AIDS Research and Reference Reagent Program, Division of AIDS, NIAID, NIH. The Z29 strain of HHV-6B and MOLT-3 cells were obtained from Scott Schmid at the Centers for Disease Control and Prevention, Atlanta, GA.

Plaque reduction assay for HSV-1, HSV-2, VZV, HCMV, and MCMV.

Plaque reduction assays were performed with monolayers of HFF cells by methods described previously (30). Briefly, six-well plates containing confluent monolayers of HFF cells were infected to yield approximately 20 to 30 plaques per well. Compound dilutions were prepared in minimal essential medium with 10% fetal bovine serum and standard concentrations of l-glutamine, penicillin, and gentamicin. Growth media containing the compounds were added to duplicate wells, and the plates were incubated at 37°C with 5% CO₂ and 90% humidity. Cell monolayers were stained with a 1.5% solution of neutral red after an incubation of 3 days for HSV-1 and HSV-2, 8 days for HCMV, and 10 days for VZV. Assays for MCMV were performed by similar methods but utilized primary mouse embryo fibroblast cells in 12-well plates, and the cells were stained at 7 days after infection. All plaques were enumerated on a stereomicroscope at 10× magnification, and effective concentrations that were sufficient to reduce viral replication by 50% (EC₅₀) were calculated by standard methods (23).

Determination of efficacy against EBV.

Antiviral activity against EBV was determined in Daudi cells by methods described previously (30). Briefly, Daudi cells were superinfected with the P3HR-1 strain of EBV, which induced a lytic infection in approximately 10% of the cells. Dilutions of compounds were then added to the infected cells, which were incubated for 3 days at 37°C. The infected cells were then placed in 96-well plates and dried to fix the cells to the bottoms of the wells. Lytic viral replication was detected with a monoclonal antibody specific for viral capsid antigen (Chemicon, Tamecula, CA) and an HRP-labeled secondary antibody (Southern Biotech, Birmingham, AL) in an enzyme-linked immunosorbent assay. Bound secondary antibody was detected with O-phenylenediamine dihydrochloride, and the optical density was determined at 492 nm. The interpolation of the EC₅₀s was performed by standard methods.

Antiviral assays for HHV-6A and HHV-6B.

An assessment of the antiviral activity against HHV-6 was performed as described previously (30). Uninfected HSB-2 or MOLT-3 cells were infected by the addition of HHV-6A-infected HSB-2 cells or HHV-6B-infected MOLT-3 cells, respectively, at a ratio of approximately one infected cell for every uninfected cell. Diluted compounds were added to the infected cells in 96-well plates, which were incubated for 7 days at 37°C. The accumulation of the viral DNA was assessed by DNA hybridization, and standard methods were used to interpolate the EC₅₀s.

Evaluation of cytotoxicity.

Cytotoxicity was determined in parallel with antiviral activity in the same cell lines used to determine antiviral activity. In HFF cells, the compound concentration that reduced the uptake of the neutral red vital dye by 50% (CC₅₀) was used as a measure of toxicity. Briefly, 2.5 × 10⁴ low-passage HFF cells were seeded into each well of the 96-well tissue culture plates containing growth medium and incubated for 24 h at 37°C in a CO₂ incubator. Media containing compound dilutions were then added to the plates, which were incubated for an additional 7 days. Cell monolayers were stained with a 0.01% solution of neutral red in phosphate-buffered saline and incubated for 1 h. The cells were washed, the dye internalized by the cells was solubilized in 100 μl of a 50% ethanol solution supplemented with 1% glacial acetic acid, and the optical density was determined at 540 nm. CellTiter-Glo (Promega, Madison, WI) assays were also performed in a similar manner, but after 7 days of incubation, 100 μl of the medium was removed from each well, and 100 μl of CellTiter-Glo reagent was added to each of the wells. The plates were mixed for 2 min and read on a luminometer to determine the number of live cells. In lymphocytes, cytotoxicity was determined with an MTS [3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2H-tetrazolium] assay (Promega). The plates were prepared in the same manner as those for the antiviral assays described above, except that the cells were uninfected or were not induced to undergo a lytic infection. Control wells were also included that contained only growth medium. After incubation, 20 μl of MTS was added to each well, the plates were incubated at 37°C for 4 h, and the optical density was determined at 490 nm. All CC₅₀ values were calculated by standard methods.

RESULTS

Biological evaluation.

A series of compounds were evaluated to determine if the esterification of the parent phosphonomethoxyethyl purine and pyrimidine nucleoside phosphonates with HDP substituents would improve their efficacy against a series of human herpesviruses. Many of the prodrugs exhibited improved efficacy while not generally increasing their toxicity (Table 1) and resulted in improved selectivity indices (Tables 2 to 4). Most notable was the 10- to 40-fold improvement in the efficacy of HDP-PME-5BrC and HDP-PME-cPrDAP against the alphaherpesviruses (Table 2). Improved activity was not accompanied by increased cytotoxicity in HFF cells in the neutral red assay (Table 1) and resulted in increases in the selective indices between 12- and 44-fold. Increased efficacy was also observed with HDP-PME-5BrC against HSV yet did not appear to affect activity against VZV (Table 2). Similar increases in the efficacy of these three compounds were also observed against HCMV (Table 3). The most significant improvement against HCMV was with HDP-PME-cPrDAP that decreased the EC₅₀ by 20-fold and yielded a selective index of 5,500. This compound was also highly effective against MCMV, with a selectivity index of >3,667.

TABLE 1.

Toxicity of alkoxyalkyl esters of nucleoside phosphonates

Compound	CC50 (μM)a
	HFF cells (neutral red)	HFF cells (CellTiter-Glo)	Daudi cells (MTS)	HSB-2 cells (MTS)	MOLT-3 cells (MTS)
PME-C	>300	220 ± 54	>100	>50 ± 0	>50
HDP-PME-C	>300	200 ± 74	>100	>50	>50 ± 0
PME-5BrC	>300	200 ± 6	27	3.9 ± 2.8	84 ± 13
HDP-PME-5BrC	>300	35 ± 5	>100	>41 ± 12	>50 ± 0
PME-5FC	NT	220 ± 58	NT	>41 ± 13	>50 ± 0
HDP-PME-5FC	35 ± 3	33 ± 2	>100	22.6	>50
PME-DAP	>100	170 ± 36	>50	>56 ± 9	39 ± 6
HDP-PME-DAP	120 ± 42	170 ± 59	>50	6.4 ± 2.4	51 ± 16
PME-cPrDAP	>100	150 ± 49	20	3.0 ± 2.4	8.5 ± 1
HDP-PME-cPrDAP	110 ± 26	26 ± 6	>50	0.05 ± 0.01	0.3 ± 0.2

^aCC₅₀, concentration that reduced the number of viable cells by 50% by the methods shown. All values shown are the average of at least two experiments. NT, not tested.

TABLE 2.

Activity of alkoxyalkyl esters of nucleoside phosphonates against HSV-1, HSV-2, and VZV^a

Compound	HSV-1 E-377 strain		HSV-2 MS strain		VZV Ellen strain
	EC50	SI	EC50	SI	EC50	SI
PME-C	>300	1	>300	1	>300	1
HDP-PME-C	260	>1.1	35	>9	>300	1
PME-5BrC	>60	>5	>60	>5	38	>8
HDP-PME-5BrC	1.2 ± 0.8	>250	18 ± 1.2	>17	42	>7
PME-5FC	NT	NT	NT	NT	NT	NT
HDP-PME-5FC	>12	<3	>12	<3	2.6 ± 0.3	13
PME-DAP	>10	>10	>10	>10	2.9 ± 0	>34
HDP-PME-DAP	0.6 ± 0.1	200	0.4 ± 0.2	300	0.3 ± 0.2	390
PME-cPrDAP	4.0 ± 1.5	>25	7.9 ± 0.3	>13	0.8 ± 0	>130
HDP-PME-cPrDAP	0.1 ± 0.08	1100	0.2 ± 0.2	550	0.03 ± 0.004	3,700

^aEC₅₀, effective concentration that reduced plaque formation by 50%. The selectivity index (SI) is the ratio of the EC₅₀ divided by the concentration of the compound that reduced cell viability by 50% (CC₅₀) as determined in HFF cells by the neutral red uptake assay. NT, not tested. All values shown are the average of at least two experiments and are given in μM.

TABLE 4.

Activity of alkoxyalkyl esters of nucleoside phosphonates against EBV, HHV-6A, and HHV-6B^a

Compound	EBV P3HR-1		HHV-6A GS		HHV-6B Z29
	EC50	SI	EC50	SI	EC50	SI
PME-C	>41 ± 1.3	<2.4	>50 ± 0	1	>50 ± 0	1
HDP-PME-C	>50 ± 0	<2	>50 ± 0	1	>50 ± 0	1
PME-5BrC	24 ± 1.5	1.1	>50 ± 0	1	>47 ± 4	<2
HDP-PME-5BrC	0.03 ± 0.03	>3,400	38 ± 13	1	>45 ± 6	1
PME-5FC	>50 ± 0		>50 ± 0	1	>50 ± 0	1
HDP-PME-5FC	1.3 ± 0.2	>77	7.2 ± 0.6	>6	37 ± 4.5	>1
PME-DAP	1.2 ± 0.9	>42	15 ± 6	>4	28 ± 10	1
HDP-PME-DAP	0.8 ± 0.8	>63	2.0 ± 0.1	3	34 ± 2.3	1
PME-cPrDAP	0.1 ± 0.1	200	1.0 ± 0.9	3	6.4 ± 1.6	1
HDP-PME-cPrDAP	0.0007 ± 0	>71,000	0.05 ± 0.01	1	0.05 ± 0.007	6

^aEC₅₀, effective concentration that reduced plaque formation by 50%. The selectivity index (SI) is the ratio of the EC₅₀ divided by the concentration of the compound that reduced cell viability by 50% (CC₅₀). The CC₅₀ values used to calculate the SI values were obtained with MTS assays of Daudi, HSB-2, and MOLT-3 cells for EBV, HHV-6A, and HHV-6B, respectively. All values shown are the average of at least two experiments and are given in μM.

TABLE 3.

Activity of alkoxyalkyl esters of nucleoside phosphonates against HCMV and MCMV^a

Compound	HCMV AD169		MCMV Smith
	EC50	SI	EC50	SI
PME-C	>20 ± 0	<150	63 ± 7	>480
HDP-PME-C	>20 ± 0	<150	>20 ± 0	<150
PME-5BrC	>20 ± 0	<150	>20 ± 0	<150
HDP-PME-5BrC	4.8 ± 3	>63	1.3 ± 1	>230
PME-5FC	>20 ± 0	<11	78 ± 1	2.8
HDP-PME-5FC	2 ± 2	18	0.08 ± 0.01	440
PME-DAP	13 ± 7	>8	0.2 ± 0.1	>500
HDP-PME-DAP	0.5 ± 0.1	240	0.5 ± 0.1	240
PME-cPrDAP	0.4 ± 0.01	250	<0.06 ± 0.05	>1,700
HDP-PME-cPrDAP	0.02 ± 0.01	5,500	<0.03 ± 0	>3700

^aEC₅₀, effective concentration that reduced plaque formation by 50%. The selectivity index (SI) is the ratio of the EC₅₀ divided by the concentration of the compound that reduced cell viability by 50% (CC₅₀) as determined in HFF cells by the neutral red uptake assay. All values shown are the average of at least two experiments and are given in μM.

The efficacy of the esters was also increased against EBV and resulted in 800- and 142-fold improvements in the EC₅₀s for HDP-PME-5BrC and HDP-PME-cPrDAP, respectively (Table 4). Increased efficacy was also observed with HDP-PME-5BrC against this virus. While improvements in antiviral activity against HHV-6 strains were also observed with HDP-PME-cPrDAP, yielding 20- to 128-fold improvements over that with PME-cPrDAP, the increased efficacy was also accompanied by increased cytotoxicity in these cell culture systems and did not result in improved selective indices (Table 4).

DISCUSSION

The in vitro antiviral activity of purine and pyrimidine phosphonomethoxyethyl nucleoside phosphonates against seven selected human herpesviruses was significantly improved through their esterification with the HDP alkoxyalkyl moiety. The synthesis of the molecules tested here was reported previously, and their antiviral activity against HIV was also shown to be dramatically improved by the addition of the HDP esters (28). The most-impressive effects were observed with HDP-PME-DAP and HDP-PME-cPrDAP, which increased their efficacy against HIV by 4 orders of magnitude over the unesterified nucleoside phosphonates. The most potent analog against HIV was HDP-PME-cPrDAP, which was active at concentrations below 10 pM and was also the most-active compound against EBV and HCMV, with EC₅₀s of 0.7 nM and 20 nM, respectively. Observed improvements in activity in cell culture are likely related to the increased uptake of compounds and the increased formation of the active metabolites as reported previously for HDP-CDV (1), as well as (S)-9-[3-hydroxy-(2-phosphonomethoxy)propyl]adenine (21). Thus, esterification with moieties of alkoxyalkyl groups appears to improve cellular uptake with this class of compounds generally and is consistent with the results presented here. This may be particularly important for the nucleoside phosphonates since the large size of the molecules together with the negative charges associated with the phosphonate moiety is thought to limit their uptake via fluid-phase endocytosis (9). Many of the compounds also appeared to be more toxic in the lymphocyte cell lines, particularly the T-cell lines HSB-2 and MOLT-3. The cytotoxicity of this class of compounds is typically greater in rapidly dividing cancer cells than in nonmalignant cells. Subsequent studies in animal models will be required to assess these and other potential toxicities of these compounds.

The human herpesviruses also appeared to differ in their susceptibility to some of the specific phosphonomethoxyethyl nucleoside phosphonates in these studies. For example, VZV, HCMV, and EBV were more sensitive to PME-cPrDAP than HSV-1 or HSV-2 and occurred irrespective of esterification with HDP. Thus, the increased efficacy against these viruses appears to be a characteristic of the nucleotide analog. Other comparisons were difficult to make because of the limited efficacy of the unesterified nucleotides, but it is clear that both the nucleotides and the presence of HDP can impact the activity of the molecules. Additional HDP esters of phosphonomethoxyethyl nucleoside phosphonates have been synthesized and are currently being evaluated against HIV and the herpesviruses. These data, together with those presented here, will be used to select candidate compounds for further testing in animal models. While HDP-PME-cPrDAP appears to be an excellent candidate, it is possible that other related analogs might exhibit even better activity. The planned in vivo studies will be critical because they will identify the most potent analogs that are well tolerated in animals.

Esterification of CDV and cyclic CDV with HDP has also been shown to improve the efficacy of these molecules by 3 orders of magnitude against the herpesviruses (3, 29, 30). Improved activity was not limited to in vitro systems but was also was observed against HCMV and MCMV in experimental animal infections (5, 18). Enhanced in vitro efficacy was observed also against the orthopoxviruses (4, 17, 19), and this increased activity also translated into superior efficacy in animal models for HDP-CDV, as well as hexadecyloxypropyl-[(S)-9-(3-hydroxy-2-phosphonylmethoxypropyl)-adenine] [(S)-HPMPA] and ODE-(S)-HPMPA, which were shown to reduce mortality in lethal models of orthopoxvirus infection (7, 24-26). However, one of the most important attributes of HDP and ODE esters is their oral bioavailability and provides an important advantage over the free nucleoside phosphonates that must be administered parenterally (8) and exhibit nephrotoxicity.

Data presented here provide additional evidence that the esterification of pyrimidine and purine nucleoside phosphonates that are active against HIV with an alkoxyalkyl moiety (HDP) (28) also substantially improves their antiviral activity against the herpesviruses. This finding is significant because the extended spectrum of activity exhibited by these molecules might also be able to suppress opportunistic herpesvirus infections. This is relevant to the treatment of herpesvirus infections in this clinical setting at two critical levels. First, additional therapies are required to treat these infections, and the therapies of choice for opportunistic infections, such as HCMV, possess significant dose-limiting toxicities (2, 6). Second, the development of resistance to existing herpesvirus drugs is a significant problem in the immunosuppressed (13, 20). Thus, therapies for HIV infections that also suppress the replication of opportunistic herpesviruses should reduce virus loads during reactivation events and would be predicted to reduce the development of drug-resistant variants. HSV infections have also been shown to upregulate HIV replication (14), and a recent clinical study has shown that the suppression of HSV replication with valacyclovir can reduce the levels of HIV-1 RNA levels (22). Although the mechanism is likely indirect, it is possible that the inhibition of herpesvirus replication by a drug may actually increase its efficacy against HIV. The additional protection afforded by broad-spectrum agents such as those described here would be of significant value in immunosuppressed individuals, and further development of such agents is warranted.