In vitro resistance development for RO-0335, a novel diphenylether nonnucleoside reverse transcriptase inhibitor.

Nonnucleoside reverse transcriptase inhibitors (NNRTIs) are important components of current combination therapies for the treatment of human immunodeficiency virus type 1 (HIV-1) infection. However, their low genetic barriers against resistance development, cross-resistance and serious side effects can compromise the benefits of the first generation compounds in this class (efavirenz and nevirapine). To study potential pathways leading to resistance against the novel diphenylether NNRTI, RO-0335, sequential passage experiments at low multiplicity of infection (MOI) were performed to solicit a stepwise selection of resistance mutations. Two pathways to loss of susceptibility to RO-0335 were observed, containing patterns of amino acid changes at either V106I/A plus F227C (with additional contributions from A98G, V108I, E138K, M230L and P236L) or V106I/Y188L (with a potential contribution from L100I, E138K and Y181C). Characterization of the observed mutations by site-directed mutagenesis in the isogenic HXB2D background demonstrated that a minimum of two or more mutations were required for significant loss of susceptibility, with the exception of Y188L, which requires a two-nucleotide change. Patterns containing F227C or quadruple mutations selected by RO-0335 showed a low relative fitness value when compared to wild-type HXB2D.

Synthesis and enantioselectivity of the antiviral effects of (R,Z)-,(S,Z)-methylenecyclopropane analogues of purine nucleosides and phosphoralaninate prodrugs: influence of heterocyclic base, type of virus and host cells.

A series of R and S enantiomers of 2-aminopurine methylenecyclopropane analogues of nucleosides was synthesized. Two diastereoisomeric lipophilic phosphate prodrugs derived from R and S enantiomers of 2,6-diaminopurine analogue were also prepared. Enantioselectivity (diastereoselectivity in case of prodrugs) of in vitro antiviral effects was investigated with human and murine cytomegalovirus (HCMV and MCMV, respectively), herpes simplex virus types 1 and 2 (HSV-1 and HSV-2, respectively), human immunodeficiency virus type 1 (HIV-1), hepatitis B virus (HBV), Epstein-Barr virus (EBV) and varicella zoster virus (VZV). Strong differences in enantioselectivity were found between the R and S enantiomers of adenine analogue and enantiomeric 2-aminopurine analogues. Thus, the enantiomers of adenine analogue were equipotent against HCMV but not MCMV, where the S enantiomer is strongly preferred. The same S preference was found throughout the 2-aminopurine series for both HCMV and MCMV. In contrast, R-synadenol in HIV-1 assays was the best agent, whereas the S enantiomers of moderately effective 2-amino-6-cyclopropylamino and 2-amino-6-methoxypurine analogues were preferred. Little enantiomeric preference was found for R and S enantiomers of synadenol and the corresponding enantiomers of 2,6-diaminopurine analogue against HBV. A mixed pattern of enantioselectivity was observed for EBV depending on the type of host cells and assay. Against VZV, the R and S enantiomers of adenine analogue were equipotent or almost equipotent, but throughout the series of 2-aminopurine analogues a distinct preference for the S enantiomers was found. The stereoselectivity pattern of both diastereoisomeric prodrugs mostly followed enantioselectivity of the parent analogues. The varying enantioselectivities in the series of purine methylenecyclopropane analogues are probably a consequence of differences in the mechanisms of action in different virus/host cell systems.

Deoxy sugar analogues of triciribine: correlation of antiviral and antiproliferative activity with intracellular phosphorylation.

Triciribine (TCN) and triciribine monophosphate (TCN-P) have antiviral and antineoplastic activity at low micromolar or submicromolar concentrations. In an effort to improve and better understand this activity, we have conducted a structure-activity relationship study to explore requirements for the number of hydroxyl groups on the ribosyl moiety for biological activity. 2'-Deoxytriciribine (2'-dTCN), 3'-deoxytriciribine (3'-dTCN), 2', 3'-epoxytriciribine (2',3'-epoxyTCN), 2',3'-dideoxy-2', 3'-didehydrotriciribine (2',3'-d4TCN), and 2',3'-dideoxytriciribine (2',3'-ddTCN) were synthesized and evaluated for activity against human immunodeficiency virus (HIV-1), herpes simplex virus type 1 (HSV-1), and human cytomegalovirus (HCMV). Antiproliferative activity of the compounds also was tested in murine L1210 cells and three human tumor cell lines. All compounds were either less active than TCN and TCN-P or inactive at the highest concentration tested (100 microM) in both antiviral and antiproliferative assays. Reverse-phase HPLC of extracts from uninfected cells treated with the deoxytriciribine analogues only detected the conversion of 3'-dTCN and 2',3'-ddTCN to their respective monophosphates. Therefore, either the deoxytriciribine analogues were not transported across the cell membrane or, more likely, they were not substrates for a nucleoside kinase or phosphotransferase. We have concluded that the hydroxyl groups on the ribosyl ring system of TCN and TCN-P must be intact in order to obtain significant antiviral and antineoplastic activity.

6-N-Acyltriciribine analogues: structure-activity relationship between acyl carbon chain length and activity against HIV-1.

Triciribine (TCN) and triciribine-5'-monophosphate (TCN-P) are active against HIV-1 at submicromolar concentrations. In an effort to improve and better understand this activity, we have conducted a structure-activity relationship study to explore the tolerance of TCN to structural modifications at the 6-position. A number of 6-N-acyltriciribine analogues were synthesized and evaluated for antiviral activity and cytotoxicity. The cytotoxicity of these compounds was minimal in three human cell lines (KB, CEM-SS cells, and human foreskin fibroblasts (HFF)). The compounds were marginally active or inactive against herpes simplex virus type 1 (HSV-1) and human cytomegalovirus (HCMV). In contrast, most of the compounds exhibited moderate to high activity against human immunodeficiency virus type 1 (HIV-1), IC(50)'s = 0.03 to 1 microM. This structure-activity relationship study identified the N-heptanoyl group as having the optimal carbon chain length. This compound was as active against HIV-1 as TCN and TCN-P. Reverse phase HPLC of extracts from uninfected cells treated with 6-N-acyltriciribines detected sufficient TCN-P to account for anti-HIV activity thereby suggesting a prodrug effect. Studies in an adenosine kinase deficient cell line showed that the 6-N-acyl derivative was not phosphorylated directly but first was metabolized to triciribine which then was converted to TCN-P.

Synthesis and antiviral evaluation of halogenated beta-D- and -L-erythrofuranosylbenzimidazoles.

A series of 2-substituted benzimidazole D- and L-erythrofuranosyl nucleosides were synthesized and tested for activity against herpesviruses and for cytotoxicity. The D-nucleosides 2,5, 6-trichloro-1-(beta-D-erythrofuranosyl)benzimidazole (8a) and 2-bromo-5,6-dichloro-1-(beta-D-erythrofuranosyl)benzimidazole (8b) were prepared by coupling 1,2,3-tri-O-acetyl-beta-D-erythrofuranose (D-6) with the appropriate benzimidazole, followed by removal of the acetyl protecting groups. The 2-isopropylamino (9), 2-cyclopropylamino (10), and 2-mercaptobenzyl (11) derivatives were synthesized by nucleophilic displacements of the C-2 chlorine in the benzimidazole moiety of 8a. The D-nucleoside 4-bromo-5, 6-dichloro-2-isopropylamino-1-(beta-D-erythrofuranosyl)benzimid azo le (17) was prepared by coupling D-6 with the appropriate benzimidazole. The L-erythrofuranosyl derivatives, 5, 6-dichloro-2-isopropylamino-1-(beta-L-erythrofuranosyl)benzimid azo le (21a), its 2-cyclopropylamino analogue (21b), and the 2-isopropylamino analogue (25), were prepared by coupling L-6 with the appropriate benzimidazole. Several of these new derivatives had very good activity against HCMV in plaque and yield reduction assays (IC(50) = 0.05-19 microM against the Towne strain of HCMV) and DNA hybridization assays. Very little activity was observed against other herpesviruses. This pattern is similar to the antiviral activity profile observed for the corresponding ribofuranosides 2,5, 6-trichloro-1-(beta-D-ribofuranosyl)benzimidazole (4a), its 2-bromo analogue (4b), and the 2-cyclopropylamino analogue (4c). In comparison, 8a was 15-fold more active against HCMV than 4a, and 8b was 4-fold more active against HCMV than 4b. The 5, 6-dichloro-2-isopropylamino-1-(beta-L-erythrofuranosyl)benzimid azo le (21a) was less active than 4c, which is now in clinical trials for HCMV infection. Both 8a,b had comparable HCMV activity to 4c. Mode of action studies with the D-erythrose analogues established that 8b acted by inhibition of viral DNA processing whereas 9 and 10 may act via a different mechanism. The lack of a 5'-hydroxymethyl group in all members of this series established that antiviral activity occurred without 5'-phosphorylation, a feature required for the activity of most nucleoside analogues.

Differences in DNA packaging genes and sensitivity to benzimidazole ribonucleosides between human cytomegalovirus strains AD169 and Towne.

The AD169 strain of human cytomegalovirus was approximately twofold more sensitive to polyhalogenated benzimidazole ribonucleosides than Towne strain. Sequence differences between the two strains have been identified in genes UL51, UL52, UL56, UL77, UL89 and UL104. Because these genes are involved in cleavage and packaging of viral DNA and the benzimidazole ribonucleosides inhibit this process, these sequence differences may be involved in the difference in drug sensitivity.

Synthesis and antiviral activity of phosphoralaninate derivatives of methylenecyclopropane analogues of nucleosides.

Phenylmethylphosphoro-L-alaninate prodrugs of antiviral Z-methylenecyclopropane nucleoside analogues and their inactive E-isomers were synthesized and evaluated for their antiviral activity against HCMV, HSV-1, HSV-2, HHV-6, EBV, VZV, HIV-1 and HBV. The adenine Z-analogue was a potent inhibitor of all these viruses but it displayed cellular toxicity. The guanine Z-derivative was active against HCMV, HBV, EBV and VZV and it was not cytotoxic. The 2,6-diaminopurine analogue was the most potent against HIV-1 and HBV and somewhat less against HHV-6, HCMV, EBV and VZV in a non-cytotoxic concentration range. The 2-amino-6-cyclopropylamino and 2-amino-6-methoxypurine prodrugs were also more active than parent analogues against several viruses but with a less favorable cytotoxicity profile. In the E-series of analogues, adenine derivative was active against HIV-1, HBV and EBV, and it was non-cytotoxic. The guanine analogue exhibited a significant effect only against HBV. The 2,6-diaminopurine E-analogue was inactive with the exception of a single EBV assay. The 2-amino-6-methoxypurine Z-methylenecyclopropane nucleoside analogue was an effective inhibitor of HCMV, MCMV and EBV. The 2,6-diaminopurine Z-prodrug seems to be the best candidate for further development.

Methylenecyclopropane analogues of nucleosides: synthesis, absolute configuration, and enantioselectivity of antiviral effect of (R)-(-)- and (S)-(+)-synadenol.

Synthesis, absolute configuration and antiviral activity of enantiomeric antiviral agents (R)-(-)- and (S)-(+)-synadenol (2 and 3a) are described.

Acyclic sugar analogs of triciribine: lack of antiviral and antiproliferative activity correlate with low intracellular phosphorylation.

Triciribine and triciribine monophosphate have antiviral and antiproliferative activity at low or submicromolar concentrations. In an effort to improve and better understand this activity, we have synthesized a series of acyclic analogs and evaluated them for activity against select viruses and cancer cell lines. We conclude that the rigid ribosyl ring system of triciribine must be intact in order to be phosphorylated and to obtain significant antiviral and antiproliferative activity.

(R)-(-)- and (S)-(+)-Synadenol: synthesis, absolute configuration, and enantioselectivity of antiviral effect.

Synthesis of (R)-(-)- and (S)-(+)-synadenol (1a and 2a, 95-96% ee) is described. Racemic synadenol (1a + 2a) was deaminated with adenosine deaminase to give (R)-(-)-synadenol (1a) and (S)-(+)-hypoxanthine derivative 5. Acetylation of the latter compound gave acetate 6. Reaction with N, N-dimethylchloromethyleneammonium chloride led to 6-chloropurine derivative 7. Ammonolysis furnished (S)-(+)-synadenol (2a). Absolute configuration of 1a was established by two methods: (i) synthesis from (R)-methylenecyclopropanecarboxylic acid (8) and (ii) X-ray diffraction of a single crystal of (-)-synadenol hydrochloride. Racemic methylenecyclopropanecarboxylic acid (10) was resolved by a modification of the described procedure. The R-enantiomer 8 was converted to ethyl ester 13 which was brominated to give vicinal dibromides 14. Reduction with diisobutylaluminum hydride then furnished alcohol 15 which was acetylated to the corresponding acetate 16. Alkylation-elimination procedure of adenine with 16 yielded acetates 17 and 18. Deprotection with ammonia afforded a mixture of Z- and E-isomers 1a and 19 of the R-configuration. Comparison with products 1a and 2a by chiral HPLC established the R-configuration of (-)-synadenol (1a). These results were confirmed by X-ray diffraction of a single crystal of (-)-synadenol hydrochloride. The latter forms a pseudosymmetric dimer with adenine-adenine base pairing in the lattice with the nucleobase in an anti-like conformation. Enantiomers 1a and 2a exhibit varied enantioselectivity toward different viruses. Both enantiomers are equipotent against human cytomegalovirus (HCMV) and varicella zoster virus (VZV). The S-enantiomer 2a is somewhat more effective than R-enantiomer 1a in herpes simplex virus 1 and 2 (HSV-1 and HSV-2) assays. By contrast, enantioselectivity of antiviral effect is reversed in Epstein-Barr virus (EBV) and human immunodeficiency virus type 1 (HIV-1) assays where the R-enantiomer 1a is preferred. In these assays, the S-enantiomer 2a is less effective (EBV) or devoid of activity (HIV-1).

Phosphorylation of triciribine is necessary for activity against HIV type 1.

Triciribine (TCN) is a tricyclic nucleoside with known antineoplastic and antiviral activity. It is a potent and selective inhibitor of HIV-1 and HIV-2, including strains known to be resistant to AZT or TIBO. TCN is phosphorylated to its 5'-monophosphate (TCN-P) by intracellular adenosine kinase (AK), but is not converted to di- or triphosphates. We now report that 5'-phosphorylation is requisite for the activity of TCN against HIV-1. CEM cells incubated with TCN at concentrations ranging from 0.1 to 330 microM gave intracellular TCN-P concentrations from 27 to 775 microM, respectively. There was no difference in the amount of intracellular TCN-P detected in uninfected compared with HIV-1-infected CEM cells. The antiviral effect of TCN against HIV-1 was strongly antagonized by the AK inhibitor 5-iodotubercidin (ITu). In contrast, TCN and ITu only exhibited additive cytotoxicity. The 5'-deoxy analog of TCN, which cannot be phosphorylated, had no antiviral effect against HIV-1 at a concentration more than 100 times higher than the IC50 of TCN. Similarly, TCN was not active against HIV-1 in an AK-deficient cell line (AA-2) at concentrations shown to inhibit the virus by >95% in CEM cells. Consistent with its AK-deficient phenotype, this cell line phosphorylated TCN to only 3% of the extent observed in CEM cells. We conclude that TCN must be phosphorylated to TCN-P for activity against HIV-1.

Resistance of human cytomegalovirus to benzimidazole ribonucleosides maps to two open reading frames: UL89 and UL56.

2,5,6-Trichloro-1-beta-D-ribofuranosyl benzimidazole (TCRB) is a potent and selective inhibitor of human cytomegalovirus (HCMV) replication. TCRB acts via a novel mechanism involving inhibition of viral DNA processing and packaging. Resistance to the 2-bromo analog (BDCRB) has been mapped to the UL89 open reading frame (ORF), and this gene product was proposed as the viral target of the benzimidazole nucleosides. In this study, we report the independent isolation of virus that is 20- to 30-fold resistant to TCRB (isolate C4) and the characterization of the virus. The six ORFs known to be essential for viral DNA cleavage and packaging (UL51, UL52, UL56, UL77, UL89, and UL104) were sequenced from wild-type HCMV, strain Towne, and from isolate C4. Mutations were identified in UL89 (D344E) and in UL56 (Q204R). The mutation in UL89 was identical to that previously reported for virus resistant to BDCRB, but the mutation in UL56 is novel. Marker transfer analysis demonstrated that each of these mutations individually caused approximately 10-fold resistance to the benzimidazoles and that the combination of both mutations caused approximately 30-fold resistance. The rate and extent of replication of the mutants was the same as for wild-type virus, but the viruses were less sensitive to inhibition of DNA cleavage by TCRB. Mapping of resistance to UL56 supports and extends recent work showing that UL56 codes for a packaging motif binding protein which also has specific nuclease activity (E. Bogner et al., J. Virol. 72:2259-2264, 1998). Resistance which maps to two different genes suggests that their putative proteins interact and/or that either or both have a benzimidazole ribonucleoside binding site. The results also suggest that the gene products of UL89 and UL56 may be antiviral drug targets.

(Z)- and (E)-2-((hydroxymethyl)cyclopropylidene)methyladenine and -guanine. New nucleoside analogues with a broad-spectrum antiviral activity.

New nucleoside analogues 14-17 based on a methylenecyclopropane structure were synthesized and evaluated for antiviral activity. Reaction of 2,3-dibromopropene (19) with adenine (18) led to bromoalkene 20, which was benzoylated to give N6,N6-dibenzoyl derivative 23. Attempts to convert 20 or 23 to bromocyclopropanes 21 and 22 by reaction with ethyl diazoacetate catalyzed by Rh2(OAc)4 were futile. By contrast, 2,3-dibromopropene (19) afforded smoothly (E)- and (Z)-dibromocyclopropane carboxylic esters 24 + 25. Alkylation of adenine (18) with 24 + 25 gave (E)- and (Z)-bromo derivatives 21 + 22. Base-catalyzed elimination of HBr resulted in the formation of (Z)- and (E)-methylenecyclopropanecarboxylic esters 26 + 27. More convenient one-pot alkylation-elimination of adenine (18) or 2-amino-6-chloropurine (30) with 24 + 25 afforded (Z)- and (E)-methylenecyclopropane derivatives 26 + 27 and 31 + 32. The Z-isomers were always predominant in these mixtures (Z/E approximately 2/1). Reduction of 26 + 27 and 31 + 32 with DIBALH afforded (Z)- and (E)-methylenecyclopropane alcohols 14 + 16 and 33 + 34. The latter were resolved directly by chromatography. Compounds 14 + 16 were converted to N6-(dimethylamino)methylene derivatives 28 and 29 which were separated and deprotected to give 14 and 16. Reaction of 33 and 34 with HCO2H led to guanine analogues 15 and 17. The 1H NMR spectra of the Z-analogues 14 and 15 are consistent with an anti-like conformation of the nucleobases. By contrast, 1H NMR and IR spectra of bromo ester 21 are indicative of syn-conformation of adenine. Several Z-(hydroxymethyl)methylenecyclopropanes exhibited in vitro antiviral activity in micromolar or submicromolar range against human and murine cytomegalovirus (HCMV and MCMV), Epstein-Barr virus (EBV), human herpes virus 6 (HHV-6), varicella zoster virus (VZV), and hepatitis B virus (HBV). Analogues 14, 15, and 33 were the most effective agents against HCMV (IC50 1-2.1, 0.04-2.1, and 0.8-5.6 microM), MCMV (IC50 2.1, 0.3, and 0.3 microM) and EBV in H-1 (IC50 0.2, 0.3, and 0.7 microM) and Daudi cells (IC50 3.2, 5.6, and 1.2 microM). Adenine analogue 14 was active against HBV (IC50 2 microM), VZV (IC50 2.5 microM), and HHV-6 (IC50 14 microM). Synadenol (14) and the E-isomer (16) were substrates of moderate efficiency for adenosine deaminase from calf intestine. The E-isomer 16 was more reactive than Z-isomer 14. The deamination of 14 effectively stopped at 50% conversion. Synadenol (14) was also deaminated by AMP deaminase from aspergillus sp.

(Z)- and (E)-2-(hydroxymethylcyclopropylidene)-methylpurines and pyrimidines as antiviral agents.

Several Z- and E-methylenecyclopropane nucleoside analogues were synthesized and tested for antiviral activity in vitro against human and murine cytomegalovirus (HCMV, MCMV), Epstein-Barr virus (EBV), varicella zoster virus (VZV), hepatitis B virus (HBV), herpes simplex virus types 1 and 2 (HSV-1, HSV-2), human herpesvirus 6 (HHV-6) and human immunodeficiency virus type 1 (HIV-1). The Z-2-amino-6-cyclopropylaminopurine analogue was the most effective agent against HCMV (EC50 or EC90 0.4-2 microM) followed by syncytol and the Z-2,6-diaminopurine analogues (EC50 or EC90 3.4-29 and 11-24 microM, respectively). The latter compound was also a strong inhibitor of MCMV (EC50 0.6 microM). Syncytol was the most potent against EBV (EC50 < 0.41 and 2.5 microM) followed by the Z-2,6-diaminopurine (EC50 1.5 and 6.9 microM) and the Z-2-amino-6-cyclopropyl-aminopurine derivative (EC50 11.8 microM). Syncytol was also most effective against VZV (EC50 3.6 microM). Activity against HSV-1, HSV-2 and HHV-6 was generally lower; synthymol had an EC50 of 2 microM against HSV-1 (ELISA) and 1.3 microM against EBV in Daudi cells but was inactive in other assays. The 2-amino-6-cyclopropylamino analogue displayed EC50 values between 215 and > 74 microM in HSV-1 and HSV-2 assays. 2-Amino-6-cyclopropylaminopurine and 2,6-diaminopurine derivatives were effective against HBV (EC50 2 and 10 microM, respectively), whereas none of the analogues inhibited HIV-1 at a higher virus load. Syncytol and the E isomer were equipotent against EBV in Daudi cells but the E isomer was much less effective in DNA hybridization assays. The E-2,6-diaminopurine analogue and E isomer of synthymol were devoid of antiviral activity.

Infection and replication of human cytomegalovirus in bone marrow stromal cells: effects on the production of IL-6, MIP-1alpha, and TGF-beta1.

We have investigated the mechanisms by which hematopoiesis is suppressed in patients suffering from human cytomegalovirus (HCMV) infections. Mixed populations of human bone marrow stromal and hematopoietic progenitor cells were inoculated with the Towne strain of HCMV to determine whether these populations could be infected and support HCMV replication. We found that the Towne strain of HCMV was capable of infecting and replicating in a mixed population of bone marrow stromal cells. We observed no significant alterations in bone marrow stromal cell proliferation or the production of IL-6, GM-CSF, soluble c-kit ligand and TNF-alpha following HCMV replication in either stimulated lipopolysaccharide (LPS) or unstimulated conditions. In samples of culture supernatants from LPS-stimulated HCMV-infected stromal cells, significant elevations in MIP-1alpha were observed. TGF-beta1 levels on the other hand exhibited two patterns following HCMV exposure; either TGF-beta1 levels decreased regardless of LPS stimulation or there was no effect. In addition, we observed that exposure to the Towne strain of HCMV resulted in significant inhibition of both granulocytic and erythrocytic colony formation in methylcellulose progenitor assays. Thus, both the direct effect of HCMV on hematopoietic progenitors as well as altered cytokine production by bone marrow stromal cells (including MIP-1alpha and TGF-beta1, but not IL-6) could contribute to hematopoietic failure during HCMV infection.

Synthesis of non-nucleoside analogs of toyocamycin, sangivamycin, and thiosangivamycin: the effect of certain 4- and 4,6-substituents on the antiviral activity of pyrrolo[2,3-d]pyrimidines.

A number of 4-substituted 7-(ethoxymethyl)- and 7-[(2-methoxyethoxy)methyl]pyrrolo[2,3-d]-pyrimidine-5-carbonitrile and -5-thiocarboxamide derivatives and several 7-substituted 4,6-diaminopyrrolo[2,3-d]pyrimidine-5-carbonitrile, -5-carboxamide, and -5-thiocarboxamide analogs related to the nucleoside antibiotics toyocamycin and sangivamycin were prepared and tested for activity against human cytomegalovirus (HCMV) and herpes simplex virus type 1 (HSV-1). Biologically, modifications at the 4-position were not well tolerated in cell culture, and in almost all cases no activity against HCMV or HSV-1 was observed. Furthermore, none of the compounds inhibited the growth of L1210 murine leukemic cells in vitro. In sharp contrast to the 4-substituted compounds, all of the 4,6-diamino 5-nitrile and the 5-thioamide analogs were active against HCMV, whereas the 5-carboxamides were inactive. The corresponding 4-amino 6-methylamino and 6-dimethylamino 5-nitrile analogs were inactive against HCMV, establishing that an amino group at both C-4 and C-6 is a likely requirement for antiviral activity. Overall, our results demonstrate that an amino group at C-4 and a thioamide moiety at C-5 of a 7-substituted pyrrolo[2,3-d]pyrimidine are essential for activity against HCMV, whereas a 4,6-diamino analog does not necessarily require a thioamide group at C-5 for activity against HCMV.

Synthesis and antiproliferative and antiviral activity of 2'-deoxy-2'-fluoroarabinofuranosyl analogs of the nucleoside antibiotics toyocamycin and sangivamycin.

The glycosylation of 3,4-dicyano-2-[(ethoxymethylene)amino]pyrrole (7) with 2-deoxy-2-fluoro-alpha-D-erythro-pentofuranosyl bromide (2) furnished an anomeric mixture of nucleosides (8a,b). This mixture was separated, and the individual anomers were treated with methanolic ammonia to effect a concomitant deblocking and ring closure. This furnished both anomers of 2'-deoxy-2'-fluoro-ara-toyocamycin (9a,b). The cyano moiety of 9b was converted to the carboxamide moiety to furnish 2'-deoxy-2'-fluoro-ara-sangivamycin (10) and to the thiocarboxamide moiety to furnish 2'-deoxy-2'-fluoro-ara-thiosangivamycin (11). The target compounds 10 and 11 showed similar antiproliferative activity against L1210 cells in vitro, with IC50's of 3 and 5 microM. Antiviral evaluation revealed a somewhat different pattern of activity. All analogs, both alpha and beta anomers, were active against human cytomegalovirus (HCMV), albeit the beta anomers were most active. The beta anomers also were active against herpes simplex virus type 1 (HSV-1) and human immunodeficiency virus (HIV). Compound 10 was most active in the series, ca. 10-fold more potent than 11; IC50's for 10 ranged from 4 to 25 nM for HCMV, HIV, and varicella zoster virus (VZV) and from 30 to 500 nM for HSV-1. Even though compound 10 was cytotoxic, which will probably preclude its use as an antiviral drug (IC50's = 0.2-5.5 microM), the difference between cytotoxicity and activity against HCMV, HIV, and VZV was sufficient to indicate specific activity against a viral target.

Inhibition of human cytomegalovirus in culture by alkenyl guanine analogs of the thiazolo[4,5-d]pyrimidine ring system.

A series of alkyl and alkenyl guanine analogs containing a thiazolo[4,5-d]pyrimidine ring system were prepared by reaction of the appropriate alkyl halide with the sodium salt of the heterocycle. In preliminary antiviral efficacy evaluations against laboratory strains of both human cytomegalovirus (HCMV) and herpes simplex virus types 1 and 2, it was determined that two of the compounds (T70072 and T01132) were more active and less toxic in stationary-phase cell monolayers than were the other derivatives tested. T01132 and T70072, which have 2-pentenyl and 3-methyl-2-butenyl moieties attached to position 3 of the 5-aminothiazolo[4,5-d]pyrimidine-2,7-dione, respectively, were then more extensively evaluated for anti-HCMV activity. The concentrations of T01132 and T70072 required to inhibit HCMV by 50% in plaque reduction assays were approximately 0.5 and 6.8 microM, respectively. These two compounds inhibited the growth of KB, MRC-5, or Vero cells at concentrations of 75 to 150 microM, depending upon the cell line. In bone marrow progenitor cells T01132 was slightly less toxic than ganciclovir (DHPG). The 50% inhibitory concentrations of T01132 against clinical isolates and DHPG-resistant strains of HCMV were approximately the same as those obtained for laboratory strains of HCMV (approximately 0.5 microM). When tested in combination with DHPG, the resultant antiviral activity was determined to be additive but not synergistic. Experiments performed using variations of the viral multiplicity of infection (MOI) demonstrated that T01132 was more active than DHPG at a low MOI (0.002 or 0.02). However, when a higher MOI (0.2 or 2.0) was used, DHPG was more efficacious than T01132. In experiments in which drug was added at various times post-viral infection, T01132 was most effective when added within the first 24 h post-HCMV infection while DHPG was able to protect cells in this assay system when added up to 48 h postinfection, indicating that T01132 is exerting its antiviral effect on events leading up to and possibly including viral DNA synthesis. The data presented in this report suggest that the antiviral activity of alkenyl-substituted thiazolopyrimidine derivatives may represent a mechanism of action against herpesviruses alternative to that of classical nucleoside analogs such as acyclovir or DHPG.