PPP2R5D Promotes Hepatitis C Virus Infection Through Binding to NS5B Protein

Muhammad Ikram Anwar Guangdong Provincial Key Laboratory of Gastroenterology,Department of Gastroenterology and Hepatology Unit,Nanfang Hospital,Southern Medical University Ni Li Institute of Human Virology,Zhongshan School of Medicine,and Key Laboratory of Tropical Disease Control of Ministry of Education,Sun Yat-sen University Qing Zhou Institute of Human Virology,Zhongshan School of Medicine,and Key Laboratory of Tropical Disease Control of Ministry of Education,Sun Yat-sen University Mingxiao Chen Institute of Human Virology,Zhongshan School of Medicine,and Key Laboratory of Tropical Disease Control of Ministry of Education,Sun Yat-sen University Chengguang Hu Guangdong Provincial Key Laboratory of Gastroenterology,Department of Gastroenterology and Hepatology Unit,Nanfang Hospital,Southern Medical University Tao Wu Department of Infectious Diseases,Hainan General Hospital,Hainan A liated Hospital of Hainan Medical University Haihang Chen Institute of Human Virology,Zhongshan School of Medicine,and Key Laboratory of Tropical Disease Control of Ministry of Education,Sun Yat-sen University Yi-Ping Li Institute of Human Virology,Zhongshan School of Medicine,and Key Laboratory of Tropical Disease Control of Ministry of Education,Sun Yat-sen University https://orcid.org/0000-0001-6011-3101 Yuanping Zhou (  yuanpingzhou@163.com ) Guangdong Provincial Key Laboratory of Gastroenterology, Department of Gastroenterology and Hepatology Unit, Nanfang Hospital, Southern Medical University https://orcid.org/0000-0002-2493-396X


Introduction
Most hepatitis C virus (HCV) infections develop chronic hepatitis C, which increases the risk of developing brosis, cirrhosis, and hepatocellular carcinoma (HCC). To date, ~71 million people are affected by HCV, with ~1.75 million new infections and ~0.7 million deaths annually (1). In recent years, treatment of hepatitis C has been revolutionized from interferon (IFN)-based to IFN-free direct-acting antivirals (DAAs) regimens, which signi cantly increases the cure rate up to 95%. Although great success was achieved in HCV therapy, challenges exist in HCV infection, which has included, but not limited to, the lack of an HCV vaccine, global low access to DAA therapy, unceasing progression to HCC and reinfection after the cure, the emergence of drug resistance, etc. Besides, the pathogenesis of HCV infection is incompletely understood.
HCV is a member of the Hepacivirus genus in the Flaviviride family. The positive RNA genome of 9.6 kb encodes a single polyprotein of ~3,010 amino acids, anked by 5' and 3' untranslated regions (UTRs), necessary for viral RNA replication, translation, and stability (2). 5'UTR contains a highly structured internal ribosomal entry site (IRES), which helps to initiate polyprotein translation. The polyprotein is further cleaved by both host and viral proteases to produce three structural (core, E1 and E2) and seven non-structural proteins (p7, NS2, NS3, NS4A, NS4B, NS5A and NS5B). Structural proteins form HCV virions, while non-structural proteins are primarily responsible for RNA replication, translation, polyprotein processing, virus assembly, and release. (2).
HCV NS5B is an RNA dependent RNA polymerase (RdRp) required for viral RNA replication (2). Many host factors have been identi ed to be involved in and regulate HCV RNA replication by different mechanisms, such as disrupting viral proteins, RNA, or the replication complex (3,4). In general, virus propagation in the cell displays a landscape of complex virus-host interaction, and uncovering host factors involved in the interactions is critical to understand the mechanism of HCV RNA replication and viral pathogenesis.
In developing infectious HCV recombinants and full-length clones, many cell culture adaptive mutations have been identi ed and demonstrated to initiate or enhance the viral RNA replication or virus production by accelerating the complete HCV life cycle (5, 6). We previously identi ed three key mutations, F1464L/A1672S/D2979G (LSG), that permitted the replication of full-length HCV clones of genotype 2a and 2b isolates (5-7) and the infectious clones of genotypes 1a, 2c, 3a and 6a (5,6,(8)(9)(10). These studies demonstrate that the LSG mutations enhance HCV RNA replication and virus production. However, the mechanism of LSG function is still elusive. The "G" mutation is located in NS5B thumb domain and corresponding to the D559G of NS5B protein. The D559G change has been predicted to reduce the e cacy of HCV antivirals in a bioinformatics study (11). In this study, we investigated the functional role of NS5B D559G in the virus-host interaction and found that HCV NS5B interacted with cellular PPP2R5D protein, while D559G enhanced the interaction. Further, we demonstrated that PPP2R5D is required for HCV infection in cultured cells.
PPP2R5D gene was ampli ed from human hepatoma cell line Huh7.5 cells (generously provided by Dr. Charles Rice, Apath L.L.C. and Rockefeller University), and HA-tag (NH 2 -YPYDVPDYA-COOH) was added at N-or C-terminus of PPP2R5D using fusion PCR, cloned into the pEGFP-C1 (Addgene) replacing the eGFP sequence, and designated HA/n-PPP2R5D or PPP2R5D-HA/c, respectively. Untagged PPP2R5D was also constructed. All of the constructs were con rmed by DNA sequencing (Sangon Biotech Company, China).
Cell culture and transfection HCV RNA transfection, virus production and focus forming unit (FFU) assay HCV RNA transfection, virus production and focus forming unit (FFU) assay were performed as previously described. (5,6,12). To determine the HCV titer, FFU assay was performed as previously described (6).
Brie y, 6× 10 3 Huh7.5 cells were seeded in polylysine-coated 96-well plates for 24 h and then infected with serial dilutions of HCV for another 48 h. The cells were xed and immunostained with the anti-HCV Core antibody C7-50 (Santa Cruz Biotechnology, USA). The percentage of HCV Core positive cells was enumerated under microscope.

HCV infection
Wild-type and PPP2R5D-KO Huh7.5 cells were seeded into 6-well plates (3×10 5 cells per well) and allowed to grow ~16 h before transfection of pC1-PPP2R5D for HCV infection validation. After 16 h of transfection, cells were infected with HCV at a MOI of 0.01 and allowed to grow for the next 48 h. HCV Core protein was detected by western blotting. Total RNA was extracted from the infected cells, and the level of HCV RNA was determined by qRT-PCR.
Immunoprecipitations HEK 293T cells were co-transfected with D559G-NS5B and pC1-PPP2R5D, WT-NS5B and pC1-PPP2R5D, or each of constructs D559G-NS5B, WT-NS5B, and pC1-PPP2R5D using Lipofectamine 2000 (Life Technologies). For the immunoprecipitation (IP) experiments, 293T cells transfected with plasmids were lysed with 500 μl of IP lysis buffer [50 mM Tris-HCl (pH=7.4), 1% NP-40, 0.25% Na-deoxycholate, 150 mM NaCl, 1 mM EDTA, 1 mM Na 3 VO 4 , 1 mM NaF, and 1% cocktail protein inhibitors] and incubated on ice for 60 min. The cell lysates were clari ed by centrifugation at 4°C and 12000 g for 10 min. The total protein concentration was determined using the BCA Protein Assay kit (GenStar, China). 450 μl was used for IP with the mouse anti-Flag or anti-HA antibody. Brie y, the 450 μl lysates were incubated with 5 μg of antibody with gentle rotation at 4°C for 2 h. For each IP experiment, 25 μl of protein A-agarose beads (Santa Cruz Biotechnology) were washed 3 times in IP lysis buffer and incubated with 5 μg of the anti-Flag antibody and lysates with rotation at 4°C overnight. The beads were washed 5 times (5 min/time with rotation) in IP lysis buffer, and nally, 60 μl of IP buffer was added before protein was eluted in 5× protein loading buffer for SDS-PAGE analysis.

Western blot
Total protein was loaded on 10% PAGE with 60-100 μg/lane, separated by SDS-PAGE procedure, and then transferred to a PVDF membrane (0.2 μm) (Bio-Rad, USA). The transferred membrane was blocked with 5% milk at room temperature for 1 h and incubated with primary antibodies at 4°C overnight. A secondary antibody was applied at room temperature for 2 h. Proteins were visualized with an ECL chemiluminescence kit (Proteintech, China) and OPTIMAX X-Ray Film Processor (PROTEC GmbH, Germany).

qRT-PCR
Total viral or cellular RNAs were extracted and quanti ed using Nanodrop-2000 (Thermo Fisher Scienti c, USA), and 1 µg of RNA was reverse transcribed using the SuperScript VILO cDNA Synthesis Kit (Thermo Fisher Scienti c). Real-time PCR was performed using TaqMan Universal master mix II (4440038, Applied Biosystems) or an ABI 7500 real-time PCR system (Applied Biosystems). Each reaction was performed in triplicates with TaqMan Gene Expression Master Mix (TAKARA, Japan). Levels of cellular glyceraldehyde-3-phosphate dehydrogenase (GAPDH) served as an internal control. The mRNA of interest and HCV RNA were calculated by the comparative threshold cycle (C T ) method (ΔΔC T ) and normalized to the level of GAPDH. The expression pattern was also analyzed by western blot using the HCV anti-Core antibody.

Statistical analysis
All the experiments were performed in triplicates and data are presented either from a representative experiment or as means standard errors of the means (SEM). Comparisons between groups were analyzed by Student's t test as indicated. GraphPad Prism software (version 8) was used to make graphs for different HCV infection experiments (GraphPad Software).

Identi cation of cellular proteins binding to the mutated HCV NS5B
We previously identi ed the NS5B mutation D559G (D2979G in polyprotein, the "G" of LSG mutations) critical for the initiation of HCV replication and virus productions (5, 6) and here, we hypothesized that presence of D559G may have changed the interaction of NS5B with host proteins, which may facilitate HCV replication. To this end, we constructed wild-type (WT) and mutant D559G NS5B individually, with Flag tag at N-terminus (n-Flag), and con rmed its expression in 293T cells (Fig. 1A). As human hepatoma Huh7.5 cells are primarily used for HCV infections, n-Flag-NS5B-based immunoprecipitation (IP) was performed with the cell lysate of transfected Huh7.5 cells (Fig. 1B). After silver staining of PAGEseparated IP proteins, an extra protein band was visualized in the D559G-NS5B IP pellet mixture (Fig. 1B). The additional band of D559-NS5B, in parallel with the WT-NS5B staining gel at same position, were sliced and sent for mass spectrometry analysis. Several proteins were identi ed, which theoretically interacted with HCV NS5B. We selected to investigate the role of serine/threonine protein phosphatase 2A regulatory subunit delta isoform (PPP2R5D) in the HCV life cycle, since it was hit in D559G-NS5B only and had a relatively high score in mass spectrometry analysis. In addition, PPP2R5D was previously reported to suppress tumor development (14). PPP2R5D interacted with NS5B To con rm the interaction of PPP2R5D with NS5B in the cells, we performed co-immunoprecipitation (Co-IP) for HA-tagged PPP2R5D and Flag-tagged WT-NS5B, as well as HA-tagged PPP2R5D and Flag-tagged D559G-NS5B. The CO-IP experiment was performed with either anti-Flag or anti-HA antibodies to assure the authenticity of PPP2R5D-NS5B interactions (Fig. 2). The results of Co-IP experiment clearly showed that PPP2R5D interacted with WT-NS5B and D559G-NS5B. However, the interaction of PPP2R5D and D559G-NS5B was slightly stronger than that with WT-NS5B ( Fig. 2A and B), consistent with the IP experiment, in which PPP2R5D was identi ed from D559G-NS5B-bound mixture (Fig. 1B). These results con rmed that HCV NS5B interacted with PPP2R5D, while D559G enhanced the interaction. As D559G is necessary for HCV replication (6, 15), we hypothesized that PPP2R5D might play an essential role in HCV infection.
To further validate the supportive role of PPP2R5D in HCV life cycle, we generated PPP2R5D-knockout Huh7.5 cells (PPP2R5D-KO) using CRISPR/Cas9 and con rmed the disruption of PPP2R5D gene by DNA sequencing; one nucleotide insertion was identi ed at the sgRNA-guide breakage region in PPP2R5D gene, which creates a frameshift mutation (Fig. 3C). The PPP2R5D-KO Huh7.5 cells were infected with HCV (MOI=0.01), and infection titer, as well as immuno uorescence, was performed to check HCV infection (Fig. 3D). The infection rate was reduced more than 95% in PPP2R5D-KO cells as compared with the wild-type cells (Fig. 3D).
To further investigate the effect of PPP2R5D in HCV infection, PPP2R5D was overexpressed in wild-type and PPP2R5D-KO Huh 7.5 cells (Fig. 3E). As expected, a very low HCV infection was detected in PPP2R5D-KO cells, while overexpression PPP2R5D restored HCV infection in PPP2R5D-KO cells to ~95% of that infection in wild-type and eGFP plasmid-transfected control Huh7.5 cells (Fig. 3E and F). Noting that overexpression of PPP2R5D or control eGFP in wild-type Huh7.5 cells resulted in ~20% reduction of HCV infection, indicating that plasmid transfection procedures affected partially HCV infection. The expression of PPP2R5D and HCV Core proteins was validated by western blotting, and HCV Core levels mirrored the immuno uorescence observed in the cultures (Fig. 3E and G).

PPP2R5D-KO cells complimentary expressing wild-type PPP2R5D rescued HCV infections
To further con rm the requirement of PPP2R5D for HCV infection in vitro, we selected a PPP2R5D-KO cell line stably expressing PPP2R5D. We constructed PPP2R5D-expressingg plasmid and transfected into PPP2R5D-KO cells (Fig. 4A). After puromycin selection, we obtained a stable PPP2R5D-KO cell line complementally expressing PPP2R5D and designated PPP2R5D-Compl (Fig. 4B). The complementation of PPP2R5D rescued HCV infection, as PPP2R5D-Compl produced HCV infectivity titers close to that of wild-type Huh7.5 cells (Fig. 4B). Together with the results of PPP2R5D knockout and transient transfections, these results strongly demonstrate that PPP2R5D was required for HCV infection in cultured cells.

Discussion
Here, we identi ed PPP2R5D as a host factor required HCV infection in Huh7.5 cells. PPP2R5D was rst found as it differentially present in immunoprecipitation assay with D559G-mutant and WT NS5B protein of HCV. Then, we demonstrated that knockout of PPP2R5D eliminated HCV infection in Huh7.5 cells, while transiently and stably complemented PPP2R5D expression rescued HCV infections ( Fig. 3 and   Fig. 4). CO-IP experiments revealed that PPP2R5D directly interacted with HCV NS5B, and such interaction was slightly enhanced by D559G (Fig. 2). Although the mechanism of PPP2R5D in HCV infection had not been elucidated in this study, it is most likely that PPP2R5D promoted HCV infection through binding to NS5B. D559G mutation enhanced the binding of PPP2R5D with NS5B. These results may explain, at least in part, the enhancement effect of D559G in NS5B for HCV replication (6, 15).
PPP2R5D is a subunit of PP2A enzymes, which is involved in many regulating pathways and turn on or off gene expression by removing the phosphate group from proteins (16). Protein phosphatase 2A (PP2A) is a major and multifunctional serine/threonine-speci c phosphatase consisting of structural subunit A, a regulatory subunit B, and a catalytic subunit C (17). PPP2R5D plays a major role in negative regulation of the PI3K/AKT signaling pathway, autism or other brain related disorders (17). PP2A have been evaluated in cancer progression and these group of proteins have shown the interaction with different HCV proteins (16,(18)(19)(20). A recent study demonstrated that PPP2R5 proteins interacted with HIV Vif protein and affected HIV pathogenesis (21). Here, we identi ed an important role of PPP2R5D in HCV infection warrants future studies to explore the underlying mechanism of PPP2R5D in the complete HCV life cycle, which will facilitate our understanding of HCV pathogenesis.

Conclusions
We have demonstrated that cellular PPP2R5D protein was required for HCV infection in hepatoma cells. PPP2R5D and HCV NS5B had a direct interaction, to which D559G showed an enhancement effect. This study discovers new host factor important for the HCV life cycle, facilitating the studies of virus-host interaction and pathogenesis of HCV.   Co-immunoprecipitation (Co-IP) of HCV wild-type and D559G mutant NS5B with PPP2R5D. Each Flagtagged WT-NS5B and D559G-NS5B was co-transfected with HA-tagged PPP2R5D into 293T cells, and the cells were lysed 48 h post-transfection and used for immunoprecipitation experiments. (A) Co-IP was performed using anti-HA antibody and immunoblotted with anti-HA or anti-Flag antibodies; precipitation of D559G-NS5B was more than WT-NS5B. (B) Co-IP was performed using anti-Flag antibody and immunoblotted with anti-HA or anti-Flag antibodies.