MK-5172, a selective inhibitor of hepatitis C virus NS3/4a protease with broad activity across genotypes and resistant variants.

HCV NS3/4a protease inhibitors are proven therapeutic agents against chronic hepatitis C virus infection, with boceprevir and telaprevir having recently received regulatory approval as add-on therapy to pegylated interferon/ribavirin for patients harboring genotype 1 infections. Overcoming antiviral resistance, broad genotype coverage, and a convenient dosing regimen are important attributes for future agents to be used in combinations without interferon. In this communication, we report the preclinical profile of MK-5172, a novel P2-P4 quinoxaline macrocyclic NS3/4a protease inhibitor currently in clinical development. The compound demonstrates subnanomolar activity against a broad enzyme panel encompassing major hepatitis C virus (HCV) genotypes as well as variants resistant to earlier protease inhibitors. In replicon selections, MK-5172 exerted high selective pressure, which yielded few resistant colonies. In both rat and dog, MK-5172 demonstrates good plasma and liver exposures, with 24-h liver levels suggestive of once-daily dosing. When administered to HCV-infected chimpanzees harboring chronic gt1a or gt1b infections, MK-5172 suppressed viral load between 4 to 5 logs at a dose of 1 mg/kg of body weight twice daily (b.i.d.) for 7 days. Based on its preclinical profile, MK-5172 is anticipated to be broadly active against multiple HCV genotypes and clinically important resistance variants and highly suited for incorporation into newer all-oral regimens.

A limited group of class I histone deacetylases acts to repress human immunodeficiency virus type 1 expression.

Silencing of the integrated human immunodeficiency virus type 1 (HIV-1) genome in resting CD4(+) T cells is a significant contributor to the persistence of infection, allowing the virus to evade both immune detection and pharmaceutical attack. Nonselective histone deacetylase (HDAC) inhibitors are capable of inducing expression of quiescent HIV-1 in latently infected cells. However, potent global HDAC inhibition can induce host toxicity. To determine the specific HDACs that regulate HIV-1 transcription, we evaluated HDAC1 to HDAC11 RNA expression and protein expression and compartmentalization in the resting CD4(+) T cells of HIV-1-positive, aviremic patients. HDAC1, -3, and -7 had the highest mRNA expression levels in these cells. Although all HDACs were detected in resting CD4(+) T cells by Western blot analysis, HDAC5, -8, and -11 were primarily sequestered in the cytoplasm. Using chromatin immunoprecipitation assays, we detected HDAC1, -2, and -3 at the HIV-1 promoter in Jurkat J89GFP cells. Targeted inhibition of HDACs by small interfering RNA demonstrated that HDAC2 and HDAC3 contribute to repression of HIV-1 long terminal repeat expression in the HeLa P4/R5 cell line model of latency. Together, these results suggest that HDAC inhibitors specific for a limited number of class I HDACs may offer a targeted approach to the disruption of persistent HIV-1 infection.

First demonstration of cerebrospinal fluid and plasma A beta lowering with oral administration of a beta-site amyloid precursor protein-cleaving enzyme 1 inhibitor in nonhuman primates.

beta-Site amyloid precursor protein (APP)-cleaving enzyme (BACE) 1 cleavage of amyloid precursor protein is an essential step in the generation of the potentially neurotoxic and amyloidogenic A beta 42 peptides in Alzheimer's disease. Although previous mouse studies have shown brain A beta lowering after BACE1 inhibition, extension of such studies to nonhuman primates or man was precluded by poor potency, brain penetration, and pharmacokinetics of available inhibitors. In this study, a novel tertiary carbinamine BACE1 inhibitor, tertiary carbinamine (TC)-1, was assessed in a unique cisterna magna ported rhesus monkey model, where the temporal dynamics of A beta in cerebrospinal fluid (CSF) and plasma could be evaluated. TC-1, a potent inhibitor (IC(50) approximately 0.4 nM), has excellent passive membrane permeability, low susceptibility to P-glycoprotein transport, and lowered brain A beta levels in a mouse model. Intravenous infusion of TC-1 led to a significant but transient lowering of CSF and plasma A beta levels in conscious rhesus monkeys because it underwent CYP3A4-mediated metabolism. Oral codosing of TC-1 with ritonavir, a potent CYP3A4 inhibitor, twice daily over 3.5 days in rhesus monkeys led to sustained plasma TC-1 exposure and a significant and sustained reduction in CSF sAPP beta, A beta 40, A beta 42, and plasma A beta 40 levels. CSF A beta 42 lowering showed an EC(50) of approximately 20 nM with respect to the CSF [TC-1] levels, demonstrating excellent concordance with its potency in a cell-based assay. These results demonstrate the first in vivo proof of concept of CSF A beta lowering after oral administration of a BACE1 inhibitor in a nonhuman primate.

Genome-scale RNAi screen for host factors required for HIV replication.

Human immunodeficiency virus (HIV)-1 depends on the host cell machinery to support its replication. To discover cellular factors associated with HIV-1 replication, we conducted a genome-scale siRNA screen, revealing more than 311 host factors, including 267 that were not previously linked to HIV. Surprisingly, there was little overlap between these genes and the HIV dependency factors described recently. However, an analysis of the genes identified in both screens revealed overlaps in several of the associated pathways or protein complexes, including the SP1/mediator complex and the NF-kappaB signaling pathway. cDNAs for a subset of the identified genes were used to rescue HIV replication following knockdown of the cellular mRNA providing strong evidence that the following six genes are previously uncharacterized host factors for HIV: AKT1, PRKAA1, CD97, NEIL3, BMP2K, and SERPINB6. This study highlights both the power and shortcomings of large scale loss-of-function screens in discovering host-pathogen interactions.

Hit selection with false discovery rate control in genome-scale RNAi screens.

RNA interference (RNAi) is a modality in which small double-stranded RNA molecules (siRNAs) designed to lead to the degradation of specific mRNAs are introduced into cells or organisms. siRNA libraries have been developed in which siRNAs targeting virtually every gene in the human genome are designed, synthesized and are presented for introduction into cells by transfection in a microtiter plate array. These siRNAs can then be transfected into cells using high-throughput screening (HTS) methodologies. The goal of RNAi HTS is to identify a set of siRNAs that inhibit or activate defined cellular phenotypes. The commonly used analysis methods including median +/- kMAD have issues about error rates in multiple hypothesis testing and plate-wise versus experiment-wise analysis. We propose a methodology based on a Bayesian framework to address these issues. Our approach allows for sharing of information across plates in a plate-wise analysis, which obviates the need for choosing either a plate-wise or experimental-wise analysis. The proposed approach incorporates information from reliable controls to achieve a higher power and a balance between the contribution from the samples and control wells. Our approach provides false discovery rate (FDR) control to address multiple testing issues and it is robust to outliers.

ROCK controls matrix synthesis in vascular smooth muscle cells: coupling vasoconstriction to vascular remodeling.

Tenascin-C (TN-C) is an extracellular matrix (ECM) protein expressed within remodeling systemic and pulmonary arteries (PAs), where it supports vascular smooth muscle cell (SMC) proliferation. Previously, we showed that A10 SMCs cultivated on native type I collagen possess a spindle-shaped morphology and do not express TN-C, whereas those on denatured collagen possess a well-defined F-actin stress fiber network, a spread morphology, and they do express TN-C. To determine whether changes in cytoskeletal architecture control TN-C, SMCs on denatured collagen were treated with cytochalasin D, which decreased SMC spreading and activation of extracellular signal-regulated kinase 1/2 (ERK1/2), signaling effectors required for TN-C transcription. Next, to determine whether cell shape, dictated by the F-actin cytoskeleton, regulates TN-C, different geometries of SMCs (ranging from spread to round) were engineered on denatured collagen: as SMCs progressively rounded, ERK1/2 activity and TN-C transcription declined. Because RhoA and Rho kinase (ROCK) regulate cell morphology by controlling cytoskeletal architecture, we reasoned that these factors might also regulate TN-C. Indeed, SMCs on denatured collagen possessed higher levels of RhoA activity than those on native collagen, and blocking RhoA or ROCK activities attenuated SMC spreading, ERK1/2 activity, and TN-C expression in SMCs on denatured collagen. Thus, ROCK controls the configuration of the F-actin cytoskeleton and SMC shape in a manner that is permissive for ERK1/2-dependent production of TN-C. Finally, we showed that inhibition of ROCK activity suppresses SMC TN-C expression and disease progression in hypertensive rat PAs. Thus, in addition to its role in regulating vasoconstriction, ROCK also controls matrix production.

The RNA-unwinding activity of hepatitis C virus non-structural protein 3 (NS3) is positively modulated by its protease domain.

The nonstructural protein 3 (NS3) of hepatitis C virus contains a protease domain at its amino terminus and RNA helicase domain at its carboxyl terminus. To identify optimal NS3 protein for developing screening assays, we expressed full-length NS3 protease/helicase and helicase domains from both HCV type 1a (H77 strain) and 1b (Con1 strain), using either E. coli or baculovirus expression systems. Our studies showed that the full-length NS3 proteins, either with or without the presence of the NS4A domain, from either strains were at least 10-fold more efficient than the corresponding helicase domains in unwinding partial duplex RNA substrates. These findings provide a rationale for the use of full-length NS3 in high throughput screening assays to identify potent small molecule inhibitors of this important target of HCV.

Sequence requirements for the development of a chimeric HCV replicon system.

The hepatitis C virus (HCV) 3'nontranslated region (3'NTR) is important for virus infection and replicon replication. Here, we constructed a panel of chimera replicons containing non-structural (NS) and 3'NTR sequences from different HCV strains or types, and examined the requirements for stable replication. A subgenomic replicon chimera comprising the polymerase and 3'NTR from HCV strain Con1, and other non-structural genes from type 1a strain H77, supported stable colony formation and replication in Huh7 cells. However, extending the type 1a sequence to include 132 amino acids of NS5B resulted in a defective HCV replicon. In contrast, a similar chimera containing HCV strain J4 sequences linked in cis to Con1 NS5B and 3'NTR supported stable replication suggesting that the interaction between the NS proteins and the 3'NTR may represent a critical determinant. Lastly, the type 1a 3'NTR from pCV-J4L6S was unable to confer replication when paired with non-structural coding sequences from BB7 or J4 and the 3'NTR from Con1 was unable to confer replication when paired with J4 or H77 sequences. These results highlighted the importance of sequence specific interaction among 3'NTR and two distinct subdomains of the NS coding region as a determinant in supporting stable replication of subgenomic replicons. The results underscore the importance of directly cloning 3'NTR sequences from relevant clinical samples.

Resistance profile of a hepatitis C virus RNA-dependent RNA polymerase benzothiadiazine inhibitor.

Recently, a benzo-1,2,4-thiadiazine antiviral agent (C(21)H(21)N(3)O(4)S; compound 4) was shown to be a potent, highly specific inhibitor of the primary catalytic enzyme of the hepatitis C virus (HCV) replicase complex. In this study, we selected for resistance to confirm the mechanism of action for compound 4 in HCV replicon cells. As expected, spontaneous mutations or fluidity in the HCV polymerase (NS5B) coding sequence occurred upon routine passage of the HCV replicon cells in the absence of compound 4. After 1 month of culture in the presence of 10 microM compound 4, or 20 times the 50% inhibitory concentration of the replicon, replicon cells were almost 20-fold less susceptible to compound 4. Twenty-one NS5B cDNA clones were generated from the resistant replicon cells. Five mutations in the 21 NS5B clones were present at frequencies higher than that of control replicon cells, and no clone contained more than a single mutation within the polymerase gene. RNA-dependent RNA polymerase studies using purified recombinant NS5B containing these single point mutations allowed the identification of residue 414 as sufficient for biochemical resistance to compound 4. Further, the contribution of this residue to confer cell-based resistance to compound 4 was validated using a stable recombinant mutant replicon cell line which harbors a methionine-to-threonine change at residue 414. The potential for additional mutations in other nonstructural genes of HCV to contribute to the resistance profile of compound 4 is discussed.

Replication studies using genotype 1a subgenomic hepatitis C virus replicons.

Recently, cell-based replicon systems for hepatitis C virus (HCV), in which the nonstructural proteins stably replicate subgenomic viral RNA in Huh7 cells, were developed. To date, one limitation of using these replicon systems to advance drug discovery is the inability of other genotypic derivatives, beyond those of two distinct strains of genotype 1b (HCV-N and Con1), to stably replicate in Huh7 cells. In this report, we evaluated a series of replicon genotype 1a-1b chimeras, as well as a complete genotype 1a replicon clone. A subgenomic replicon construct containing only type 1a sequences failed to generate stable colonies in Huh7 cells even after repeated attempts. Furthermore, addition of an NS5A adaptive mutation (S2204I) which enhances type 1b replicon efficiency was insufficient to confer replication to the wild-type 1a replicon. This subgenomic replicon was subsequently found to be inefficiently translated in Huh7 cells compared to a type 1b replicon, and the attenuation of translation mapped to the N-terminal region of NS3. Therefore, to ensure efficient translation and thereby support replication of the 1a genome, the coding sequence for first 75 residues from type 1a were replaced with the type 1b (strain Con 1) NS3 coding sequence. Although nonstructural proteins were expressed at lower levels with this replicon than with type 1b and although the amount of viral RNA was also severalfold lower (150 copies of positive-strand RNA per cell), the replicon stably replicated in Huh7 cells. Notwithstanding this difference, the ratio of positive- to negative-strand RNA of 26 was similar to that found with the type 1b replicon. Similar results were found for a 1b replicon expressing the type 1a RNA-dependent RNA polymerase. These 1a hybrid replicons maintained sensitivity to alpha interferon (IFN-alpha), albeit with an eightfold-higher 50% inhibitory concentration than type 1b replicons. Evidence is provided herein to confirm that this differential response to IFN-alpha may be attributed directly to the type 1a polymerase.

Identification and biological characterization of heterocyclic inhibitors of the hepatitis C virus RNA-dependent RNA polymerase.

The hepatitis C virus (HCV) NS5B protein encodes an RNA-dependent RNA polymerase (RdRp), the primary catalytic enzyme of the HCV replicase complex. We established a biochemical RNA synthesis assay, using purified recombinant NS5B lacking the C-terminal 21 amino acid residues, to identify potential polymerase inhibitors from a high throughput screen of the GlaxoSmithKline proprietary compound collection. The benzo-1,2,4-thiadiazine compound 1 was found to be a potent, highly specific inhibitor of NS5B. This agent interacts directly with the viral polymerase and inhibits RNA synthesis in a manner noncompetitive with respect to GTP. Furthermore, in the absence of an in vitro-reconstituted HCV replicase assay employing viral and host proteins, the ability of compound 1 to inhibit NS5B-directed viral RNA replication was determined using the Huh7 cell-based HCV replicon system. Compound 1 reduced viral RNA in replicon cells with an IC(50) of approximately 0.5 microm, suggesting that the inhibitor was able to access the perinuclear membrane and inhibit the polymerase activity in the context of a replicase complex. Preliminary structure-activity studies on compound 1 led to the identification of a modified inhibitor, compound 4, showing an improvement in both biochemical and cell-based potency. Lastly, data are presented suggesting that these compounds interfere with the formation of negative and positive strand progeny RNA by a similar mode of action. Investigations are ongoing to assess the potential utility of such agents in the treatment of chronic HCV disease.