Previous Article | Next Article ![]()
Antimicrobial Agents and Chemotherapy, August 2003, p. 2674-2681, Vol. 47, No. 8
0066-4804/03/$08.00+0 DOI: 10.1128/AAC.47.8.2674-2681.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.
Drug Discovery, Ribapharm, Inc., Costa Mesa, California 92626
Received 5 March 2003/ Returned for modification 12 May 2003/ Accepted 23 May 2003
|
|
|---|
|
|
|---|
The objectives for the treatment of chronic hepatitis C are to achieve complete and sustained clearance of HCV RNA in serum and normalization of serum alanine aminotransferase levels. In the absence of a prophylactic vaccine or a highly specific antiviral agent, treatment options for chronically infected individuals are limited. The current treatment options for chronic hepatitis C include (pegalated) alpha interferon (IFN-
) monotherapy and (pegalated) IFN-
-ribavirin combination therapy, with sustained virological response rates of between 10 and 60% (4, 7, 15-17, 20, 21). Clearly, more effective and direct antiviral interventions are necessary for further prevention and treatment of the life-threatening complications caused by HCV infection. Efforts to identify and develop highly specific and potent HCV inhibitors have intensified recently. Investigators have targeted all regions of the HCV genome and virally encoded replication enzymes for potential therapeutic discovery.
HCV is a positive-strand RNA virus belonging to the family Flaviviridae (3). This virus family also contains about 40 flaviviruses that are associated with human diseases, including the dengue fever viruses, yellow fever viruses and Japanese encephalititis virus, as well as pestiviruses, whose infection of domesticated livestock can cause significant economic losses worldwide. Like other RNA viruses, a virally encoded replication enzyme, RNA-dependent RNA polymerase (RdRp), plays a central role in viral RNA replication of HCV and other members of the family Flaviviridae. In the case of HCV, the replication protein is termed "NS5B" (nonstructural protein 5B) (2, 5, 6, 12, 13). RdRp proteins are the key components of the viral replicase complexes and therefore serve as attractive targets for antiviral development (1, 8, 9, 11).
It is generally believed that HCV RNA replication is initiated by NS5B RdRp via a de novo or primer-independent mechanism (10, 14, 18, 19, 25, 28). Our previous observations suggest that dinucleotides can be used more efficiently as initiation molecules by HCV NS5B than mononucleotides in the in vitro RNA replication assay (27). It is thus perceivable that dinucleotide analogues may have potential to inhibit initiation of HCV RNA replication. Because the initiation process is considered to be the rate-limiting step in viral RNA replication, inhibitors that interfere with the initiation process will have advantages in suppressing virus replication. The use of dinucleotide analogues as inhibitor molecules to target the initiation step of viral RNA synthesis represents a novel approach to antiviral interference.
Synthesis of dinucleotide compounds. We initiated this study by synthesizing two dinucleotide analogues (I and II) (Fig. 1) that contained a guanine base at the 5' end and a cytosine base at the 3' end. Modifications were introduced in the sugar moieties (2' or 3'-O-methyl) and the linker (phosphorothioate) for improvement of compound stability. In addition, a 5'-phosphate group was added to further mimic the natural dinucleotides. These compounds were guanine-cytosine analogues (pGpsC).
![]() View larger version (23K): [in a new window] |
FIG. 1. Chemical synthesis schemes of dinucleotide analogues I (A) and II (B).
|
The chemical phosphorylation reagent used for addition of the 5'-phosphate had a terminal dimethoxytrityl (DMTr) group, which was utilized to facilitate reverse-phase high-performance liquid chromatography (HPLC) purification. After completion of the synthesis, dinucleotide I was treated with NH4OH at 60°C for 22 h, while dinucleotide II was treated at 80°C for 8 h. This NH4OH treatment cleaved dinucleotides from solid support and removed protecting groups from nucleobases and internucleoside phosphate linkage. The compounds were then purified with C18 columns (Waters). The dilute trifluoroacetic acid solution was used as a mobile phase to remove the DMTr protection group. The purified dinucleotide compounds were characterized by mass spectrum (I, M+H+ = 683.62; II, M+H+ = 713.42). HPLC purity was more than 85%, with the single largest impurity not more than 3%.
Dinucleotide analogues inhibited de novo initiation of RNA synthesis directed by HCV NS5B.
To determine whether the synthesized dinucleotide compounds (pGpsC analogues) can inhibit initiation of RNA replication by HCV NS5B, a de novo initiation assay was performed, which used a small synthetic 10-mer RNA (5' AAAAAAAAGC 3') as the template. In this assay, GTP (at 100 µM) was used as the initiating nucleotide, and formation of the dinucleotide product pppGpC was monitored by inclusion of [
-33P]CTP as the elongating nucleotide in the reaction. The 3'-terminal bases (GC-3') of the template RNA were chosen based on the following considerations. (i) The 3'-terminal sequence of negative-strand HCV RNA (-GC 3') serves as a template for multiple rounds of positive-strand RNA synthesis. (ii) The dinucleotide compounds synthesized mimick guanine-cytosine analogues. It has been demonstrated previously that GTP is an efficient initiation molecule for HCV NS5B in a number of assays (14, 19, 25, 28). As shown in Fig. 2, both dinucleotide analogues I (lanes 2 to 6) and II (lanes 7 to 11) had an inhibitory effect on NS5B-directed de novo initiation, with 50% inhibitory concentrations (IC50s) of 20 and 65 µM, respectively. This result indicated that the dinucleotide compounds were able to efficiently compete with the initiating nucleotide GTP for binding to the polymerase protein and were able to inhibit the initiation of RNA syn-thesis.
![]() View larger version (36K): [in a new window] |
FIG. 2. Inhibition of HCV NS5B-catalyzed de novo initiation of RNA synthesis by dinucleotide analogues. (A) Structures of dinucleotide analogues I and II. (B) Inhibition of NS5B-directed de novo initiation of RNA synthesis. A short synthetic RNA (5' AAAAAAAAGC 3') was used as a template in the de novo initiation assay. GTP and radiolabeled CTP were used as the initiating and the elongating nucleotides, respectively. A standard reaction mixture in a total volume of 20 µl contained 50 mM HEPES (pH 7.5), 10 mM ß-mercaptoethanol (ß-ME), 10 mM MgCl2, 5 µM RNA template, 100 µM GTP, 100 µM CTP-10 µCi of [ -33P]CTP, 1 µM HCV NS5B, and increasing concentrations (0 to 300 µM) of the analogue compounds. The reaction mixture was incubated at 30°C for 30 min, and the RNA products were resolved on a 25% polyacrylamide gel electrophoresis-6 M urea-Tris-borate-EDTA gel prior to PhosphorImaging analysis. Compounds I and II are shown in lanes 2 to 6 and 7 to 11, respectively. The labeled dinucleotide product (pppGpC) is indicated on the left.
|
The 5'-phosphate group was important for the inhibitory activity of the dinucleotide analogues. Both analogues tested in Fig. 2 contained a phosphate group at the 5' terminus. This was based on the assumption that the phosphate group might contribute to the binding affinity with NS5B polymerase. To demonstrate that such a phosphate group was indeed important for the inhibitory activity of the dinucleotide compounds, a third analogue (III) was tested in the same replication initiation assay. Analogue III possessed a similar structure to analogues I and II, except the 5'-phosphate group was deleted. As shown in Fig. 3, analogue III lost its inhibitory activity on de novo initiation of RNA synthesis, with an IC50 of >200 µM (compare lanes 6 and 7 with lanes 2 to 5). This result suggested that the 5'-phosphate group of the dinucleotide analogues played an important role in the interaction with NS5B protein, the lack of which resulted in loss of binding affinity required for efficient competition with the initiating GTP.
![]() ![]() View larger version (47K): [in a new window] |
FIG. 3. Requirement of the 5'-phosphate group in the dinucleotide compound for inhibitory activity. Inhibitory effects on de novo initiation of RNA synthesis by dinucleotide analogues I (lanes 2 and 3), II (lanes 4 and 5), and III (lanes 6 and 7) were compared (50 and 200 µM). A short synthetic RNA (5' AAAAAAAAGC 3') was used as the template in the de novo initiation assay. GTP and radiolabeled CTP were used as the initiating and elongating nucleotides, respectively (see the legend to Fig. 2 for assay details). The structure of compound III is shown on the right.
|
20 µM) than on ATP-initiated RNA synthesis (IC50 of
90 µM). This result, together with the fact that the dinucleotide compound tested was a guanine-cytosine (GC) analogue, suggested that the base-pairing potential between the compound and the 3'-terminal bases of the template RNA played a positive role in maximizing the inhibitory activity on de novo initiation of RNA synthesis.
![]() View larger version (52K): [in a new window] |
FIG. 4. Base pairing between the dinucleotide analogue and the template 3'-terminal bases is important for optimal activity. De novo initiation of RNA synthesis was tested with two synthetic RNA templates, which differ only in the terminal base (U versus C), in the presence of increasing concentrations of dinucleotide analogue I (see Fig. 2 for structure). ATP and GTP were used as the initiating nucleotides, respectively, based on the terminal base of the template RNA (ATP for 5' AAAAAAAAGU 3' and GTP for 5' AAAAAAAAGC 3').
|
![]() ![]() View larger version (54K): [in a new window] |
FIG. 5. Inhibition of RNA synthesis initiated from a dinucleotide primer (pGpC). The initiation assay reaction mixture contained, in addition to the standard components listed in the legend to Fig. 2, 5 µM template RNA (5' AAAAAAAAGC 3'), 5 µM end-labeled [33P]GpC primer, 100 µM UTP, and 1 µM NS5B enzyme. Extended RNA products were detected from extension of the labeled primer due to multiple UMP incorporations.
|
Despite the successful confirmation of the concept that dinucleotide analogues are effective inhibitors of de novo RNA replication, significant challenges remain before further development of this class of compounds for direct therapeutic application can be considered. Dinucleotide molecules, especially RNA, are generally unstable. The intramolecular nucleophilic attack by the 2'-hydroxyl group in the sugar moiety on the adjacent phosphate bond causes the degradation of the dinucleotides to mononucleotides and even further to nucleosides. Dinucleotides can also be hydrolyzed by water and other nucleophiles present in the biological system. In addition, the 5'-phosphate dinucleotides, which were shown to be more effective in inhibiting de novo initiation of RNA replication, are even less stable because of the presence of the additional phosphate group. Furthermore, the polyanionic nature of the dinucleotide molecules strongly hampers the penetration of such compounds through cell membrane due to the high lipophilicity on the membrane surface. Finally, dinucleotide molecules can also bind to other nucleic acids through hydrogen bonds, which may result in a loss of selectivity for a predetermined target in vivo.
The pGpsC type of dinucleotide analogues used in this study contained a phosphorothioate linker. Such linker is believed to enhance the stability of the dinucleotide compounds over the natural nucleic acid. For the same purpose, the 2'-hydroxyl group of the sugar moiety was further modified by methylation so that the 2'-hydroxyl group is not available for intramolecular hydrolysis. Nucleic acids with 2'-O-methyl modifications have been reported to possess improved stability, lipophilicity, and bioavailability, as well as other pharmacological properties. The physicochemical and pharmacological features of the dinucleotide compounds can be further improved by modifications of the phosphate groups in reducing their anionic character and, in the mean time, increasing lipophilicity. In this case, the various conjugation strategies used for improving properties of oligonucleotide compounds may be applied.
Despite all of these challenges, the demonstration that dinucleotide analogues can effectively inhibit de novo initiation of RNA replication by HCV NS5B confirms the proof of principle for this novel antiviral approach. With future advancements in addressing the outstanding issues related to the dinucleotide compounds, this class of compounds may eventually reach a stage of development for therapy against important viral diseases.
This work was supported by Ribapharm, Inc.
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»