Viruses traverse the human proteome through peptide interfaces that can be biomimetically leveraged for drug discovery.

We present a drug design strategy based on structural knowledge of protein-protein interfaces selected through virus-host coevolution and translated into highly potential small molecules. This approach is grounded on Vinland, the most comprehensive atlas of virus-human protein-protein interactions with annotation of interacting domains. From this inspiration, we identified small viral protein domains responsible for interaction with human proteins. These peptides form a library of new chemical entities used to screen for replication modulators of several pathogens. As a proof of concept, a peptide from a KSHV protein, identified as an inhibitor of influenza virus replication, was translated into a small molecule series with low nanomolar antiviral activity. By targeting the NEET proteins, these molecules turn out to be of therapeutic interest in a nonalcoholic steatohepatitis mouse model with kidney lesions. This study provides a biomimetic framework to design original chemistries targeting cellular proteins, with indications going far beyond infectious diseases.

Tethering of hexokinase 2 to mitochondria promotes resistance of liver cancer cells to natural killer cell cytotoxicity.

Hexokinases (HKs) control the first step of glucose catabolism. A switch of expression from liver HK (glucokinase, GCK) to the tumor isoenzyme HK2 is observed in hepatocellular carcinoma progression. Our prior work revealed that HK isoenzyme switch in hepatocytes not only regulates hepatic metabolic functions but also modulates innate immunity and sensitivity to Natural Killer (NK) cell cytotoxicity. This study investigates the impact of HK2 expression and its mitochondrial binding on the resistance of human liver cancer cells to NK-cell-induced cytolysis. We have shown that HK2 expression induces resistance to NK cell cytotoxicity in a process requiring mitochondrial binding of HK2. Neither HK2 nor GCK expression affects target cells' ability to activate NK cells. In contrast, mitochondrial binding of HK2 reduces effector caspase 3/7 activity both at baseline and upon NK-cell activation. Furthermore, HK2 tethering to mitochondria enhances their resistance to cytochrome c release triggered by tBID. These findings indicate that HK2 mitochondrial binding in liver cancer cells is an intrinsic resistance factor to cytolysis and an escape mechanism from immune surveillance.

Domain 2 of Hepatitis C Virus Protein NS5A Activates Glucokinase and Induces Lipogenesis in Hepatocytes.

Hepatitis C virus (HCV) relies on cellular lipid metabolism for its replication, and actively modulates lipogenesis and lipid trafficking in infected hepatocytes. This translates into an intracellular accumulation of triglycerides leading to liver steatosis, cirrhosis and hepatocellular carcinoma, which are hallmarks of HCV pathogenesis. While the interaction of HCV with hepatocyte metabolic pathways is patent, how viral proteins are able to redirect central carbon metabolism towards lipogenesis is unclear. Here, we report that the HCV protein NS5A activates the glucokinase (GCK) isoenzyme of hexokinases through its D2 domain (NS5A-D2). GCK is the first rate-limiting enzyme of glycolysis in normal hepatocytes whose expression is replaced by the hexokinase 2 (HK2) isoenzyme in hepatocellular carcinoma cell lines. We took advantage of a unique cellular model specifically engineered to re-express GCK instead of HK2 in the Huh7 cell line to evaluate the consequences of NS5A-D2 expression on central carbon and lipid metabolism. NS5A-D2 increased glucose consumption but decreased glycogen storage. This was accompanied by an altered mitochondrial respiration, an accumulation of intracellular triglycerides and an increased production of very-low density lipoproteins. Altogether, our results show that NS5A-D2 can reprogram central carbon metabolism towards a more energetic and glycolytic phenotype compatible with HCV needs for replication.

Toll-like Receptor 4-Induced Glycolytic Burst in Human Monocyte-Derived Dendritic Cells Results from p38-Dependent Stabilization of HIF-1α and Increased Hexokinase II Expression.

Cell metabolism now appears as an essential regulator of immune cells activation. In particular, TLR stimulation triggers metabolic reprogramming of dendritic cells (DCs) with an increased glycolytic flux, whereas inhibition of glycolysis alters their functional activation. The molecular mechanisms involved in the control of glycolysis upon TLR stimulation are poorly understood for human DCs. TLR4 activation of human monocyte-derived DCs (MoDCs) stimulated glycolysis with an increased glucose consumption and lactate production. Global hexokinase (HK) activity, controlling the initial rate-limiting step of glycolysis, was also increased. TLR4-induced glycolytic burst correlated with a differential modulation of HK isoenzymes. LPS strongly enhanced the expression of HK2, whereas HK3 was reduced, HK1 remained unchanged, and HK4 was not expressed. Expression of the other rate-limiting glycolytic enzymes was not significantly increased. Exploring the signaling pathways involved in LPS-induced glycolysis with various specific inhibitors, we observed that only the inhibitors of p38-MAPK (SB203580) and of HIF-1α DNA binding (echinomycin) reduced both the glycolytic activity and production of cytokines triggered by TLR4 stimulation. In addition, LPS-induced HK2 expression required p38-MAPK-dependent HIF-1α accumulation and transcriptional activity. TLR1/2 and TLR2/6 stimulation increased glucose consumption by MoDCs through alternate mechanisms that are independent of p38-MAPK activation. TBK1 contributed to glycolysis regulation when DCs were stimulated via TLR2/6. Therefore, our results indicate that TLR4-dependent upregulation of glycolysis in human MoDCs involves a p38-MAPK-dependent HIF-1α accumulation, leading to an increased HK activity supported by enhanced HK2 expression.

Early Steps of Jaagsiekte Sheep Retrovirus-Mediated Cell Transformation Involve the Interaction between Env and the RALBP1 Cellular Protein.

UNLABELLED: Ovine pulmonary adenocarcinoma is a naturally occurring lung cancer in sheep induced by the Jaagsiekte sheep retrovirus (JSRV). Its envelope glycoprotein (Env) carries oncogenic properties, and its expression is sufficient to induce in vitro cell transformation and in vivo lung adenocarcinoma. The identification of cellular partners of the JSRV envelope remains crucial for deciphering mechanisms leading to cell transformation. We initially identified RALBP1 (RalA binding protein 1; also known as RLIP76 or RIP), a cellular protein implicated in the ras pathway, as a partner of JSRV Env by yeast two-hybrid screening and confirmed formation of RALBP1/Env complexes in mammalian cells. Expression of the RALBP1 protein was repressed in tumoral lungs and in tumor-derived alveolar type II cells. Through its inhibition using specific small interfering RNA (siRNA), we showed that RALBP1 was involved in envelope-induced cell transformation and in modulation of the mTOR (mammalian target of rapamycin)/p70S6K pathway by the retroviral envelope. IMPORTANCE: JSRV-induced lung adenocarcinoma is of importance for the sheep industry. While the envelope has been reported as the oncogenic determinant of the virus, the cellular proteins directly interacting with Env are still not known. Our report on the formation of RALBP/Env complexes and the role of this interaction in cell transformation opens up a new hypothesis for the dysregulation observed upon virus infection in sheep.

The interactomes of influenza virus NS1 and NS2 proteins identify new host factors and provide insights for ADAR1 playing a supportive role in virus replication.

Influenza A NS1 and NS2 proteins are encoded by the RNA segment 8 of the viral genome. NS1 is a multifunctional protein and a virulence factor while NS2 is involved in nuclear export of viral ribonucleoprotein complexes. A yeast two-hybrid screening strategy was used to identify host factors supporting NS1 and NS2 functions. More than 560 interactions between 79 cellular proteins and NS1 and NS2 proteins from 9 different influenza virus strains have been identified. These interacting proteins are potentially involved in each step of the infectious process and their contribution to viral replication was tested by RNA interference. Validation of the relevance of these host cell proteins for the viral replication cycle revealed that 7 of the 79 NS1 and/or NS2-interacting proteins positively or negatively controlled virus replication. One of the main factors targeted by NS1 of all virus strains was double-stranded RNA binding domain protein family. In particular, adenosine deaminase acting on RNA 1 (ADAR1) appeared as a pro-viral host factor whose expression is necessary for optimal viral protein synthesis and replication. Surprisingly, ADAR1 also appeared as a pro-viral host factor for dengue virus replication and directly interacted with the viral NS3 protein. ADAR1 editing activity was enhanced by both viruses through dengue virus NS3 and influenza virus NS1 proteins, suggesting a similar virus-host co-evolution.

Structure homology and interaction redundancy for discovering virus-host protein interactions.

Virus-host interactomes are instrumental to understand global perturbations of cellular functions induced by infection and discover new therapies. The construction of such interactomes is, however, technically challenging and time consuming. Here we describe an original method for the prediction of high-confidence interactions between viral and human proteins through a combination of structure and high-quality interactome data. Validation was performed for the NS1 protein of the influenza virus, which led to the identification of new host factors that control viral replication.

Dengue Virus dependence on glucokinase activity and glycolysis Confers Sensitivity to NAD(H) biosynthesis inhibitors.

Viruses have developed sophisticated strategies to control metabolic activity of infected cells in order to supply replication machinery with energy and metabolites. Dengue virus (DENV), a mosquito-borne flavivirus responsible for dengue fever, is no exception. Previous reports have documented DENV interactions with metabolic pathways and shown in particular that glycolysis is increased in DENV-infected cells. However, underlying molecular mechanisms are still poorly characterized and dependence of DENV on this pathway has not been investigated in details yet. Here, we identified an interaction between the non-structural protein 3 (NS3) of DENV and glucokinase regulator protein (GCKR), a host protein that inhibits the liver-specific hexokinase GCK. NS3 expression was found to increase glucose consumption and lactate secretion in hepatic cell line expressing GCK. Interestingly, we observed that GCKR interaction with GCK decreases DENV replication, indicating the dependence of DENV to GCK activity and supporting the role of NS3 as an inhibitor of GCKR function. Accordingly, in the same cells, DENV replication both induces and depends on glycolysis. By targeting NAD(H) biosynthesis with the antimetabolite 6-Amino-Nicotinamide (6-AN), we decreased cellular glycolytic activity and inhibited DENV replication in hepatic cells. Infection of primary organotypic liver cultures (OLiC) from hamsters was also inhibited by 6-AN. Altogether, our results show that DENV has evolved strategies to control glycolysis in the liver, which could account for hepatic dysfunctions associated to infection. Besides, our findings suggest that lowering intracellular availability of NAD(H) could be a valuable therapeutic strategy to control glycolysis and inhibit DENV replication in the liver.

A hexokinase isoenzyme switch in human liver cancer cells promotes lipogenesis and enhances innate immunity.

During the cancerous transformation of normal hepatocytes into hepatocellular carcinoma (HCC), the enzyme catalyzing the first rate-limiting step of glycolysis, namely the glucokinase (GCK), is replaced by the higher affinity isoenzyme, hexokinase 2 (HK2). Here, we show that in HCC tumors the highest expression level of HK2 is inversely correlated to GCK expression, and is associated to poor prognosis for patient survival. To further explore functional consequences of the GCK-to-HK2 isoenzyme switch occurring during carcinogenesis, HK2 was knocked-out in the HCC cell line Huh7 and replaced by GCK, to generate the Huh7-GCK+/HK2- cell line. HK2 knockdown and GCK expression rewired central carbon metabolism, stimulated mitochondrial respiration and restored essential metabolic functions of normal hepatocytes such as lipogenesis, VLDL secretion, glycogen storage. It also reactivated innate immune responses and sensitivity to natural killer cells, showing that consequences of the HK switch extend beyond metabolic reprogramming.

TLR4 antagonist FP7 inhibits LPS-induced cytokine production and glycolytic reprogramming in dendritic cells, and protects mice from lethal influenza infection.

Dysregulated Toll-like receptor (TLR)-4 activation is involved in acute systemic sepsis, chronic inflammatory diseases, such as atherosclerosis and diabetes, and in viral infections, such as influenza infection. Thus, therapeutic control of the TLR4 signalling pathway is of major interest. Here we tested the activity of the small-molecule synthetic TLR4 antagonist, FP7, in vitro on human monocytes and monocyte-derived dendritic cells (DCs) and in vivo during influenza virus infection of mice. Our results indicate that FP7 antagonized the secretion of proinflammatory cytokines (IL-6, IL-8, and MIP-1ß) by monocytes and DCs (IC50 < 1 µM) and prevented DC maturation upon TLR4 activation by ultrapure lipopolysaccharide (LPS). FP7 selectively blocked TLR4 stimulation, but not TLR1/2, TLR2/6, or TLR3 activation. TLR4 stimulation of human DCs resulted in increased glycolytic activity that was also antagonized by FP7. FP7 protected mice from influenza virus-induced lethality and reduced both proinflammatory cytokine gene expression in the lungs and acute lung injury (ALI). Therefore, FP7 can antagonize TLR4 activation in vitro and protect mice from severe influenza infection, most likely by reducing TLR4-dependent cytokine storm mediated by damage-associated molecular patterns (DAMPs) like HMGB1.

New horizons for antiviral drug discovery from virus-host protein interaction networks.

Viruses are recurrent socio economical and health problems each year worldwide. Current drugs are mainly directed against viral components and select resistant strains that urge the need to develop new antiviral therapeutics. High-throughput screening technologies now allow to draw comprehensive genome-wide maps of physical and genetic virus-host interactions. This has been done recently for several viruses such as HIV, HCV, DENV and FLUAV and revealed a wealth of potential antiviral cellular targets. Systems-level analysis of virus-host protein networks and subnetworks begins to uncover several specific points of intervention for a human centered drug development. We present here this new paradigm in antiviral drug discovery together with the first promising antiviral molecules.

Genetic screens for the control of influenza virus replication: from meta-analysis to drug discovery.

Current anti-influenza virus drugs target two viral proteins and induce a selective pressure for the generation of drug resistant variants. This stresses the need for additional therapeutic strategies including drug targeting of cellular factors that are essential for viral replication. Reverse genetics approaches can be used to identify these factors and recently six independent genomic initiatives have led to the identification of 925 host factors that are essential for the replication of influenza viruses. Here we report a meta-analysis of this dataset, first revealing that these screens are poorly overlapping at the gene level. However, a strong convergence was observed at the level of biological processes which was further supported by an interactomic analysis showing a high interconnectivity of the essential host factors in the human protein network. Plugging virus-host protein interaction data on this dataset reveals a significant targeting of these factors by viral proteins, further validating the cellular targets. Combining this information, the first drug-influenza virus target network was constructed by retrieving from DrugBank 298 molecules interacting with 100 essential host factors. Of these, 204 are FDA-approved offering interesting potential for rapid drug repositioning in the treatment of flu.