KSHV positive body cavity lymphoma cell lines BC3 was obtained from Prof. Erle S Robertson at University of Pennsylvania, Philadelphia, USA. and GFP-HIF1α plasmid and Supercos-GB22 (KSHV genomic region 85,000-102000bp cloned in Spercos bacmid) were also kind gift from Prof. Erle S Robertson at University of Pennsylvania. The cell lines were maintained in RPMI medium supplemented with 10% fetal bovine serum and appropriate antibiotics at 37°C/5% CO₂ 2 in an humified CO₂ incubator. Hypoxic treatment was performed by incubating cells in hypoxic chamber at the oxygen concentration of 1%. The electroporation of plasmid in cells were carried out using Bio-Rad Gene Pulser Xcell electroporator system. Briefly, 10 million cells were pelleted down followed by washing with phosphate buffer saline followed by pelleting down by centrifugation at 1,500 rpm for 5 min. The cell pellets were resuspended in 0.5 ml PBS and mixed with 10 µg of plasmid. Followed by transferring the whole content in a new electroporation cuvette. Electroporation was performed by providing electric shock at 350 V and 975 µF capacitance. Electroporation cells were transferred in a well of 6-well plate in a total medium volume of 2 ml and transferred to CO2 incubator overnight followed by transferring in a T25 flask next morning. For the lentiviral based transduction, the control or ShRNA encoding plasmids was co-transfected along with helper plasmids into the HEK293T cells. The lentivirus production was stimulated by adding media containing sodium butyrate. The supernatant was collected and the lentiviruses were concentrated by high-speed centrifugation. The lentiviral transduction was performed in the presence of 4 μg/ml polybrene in a 50 ml tube in the total volume of 1 ml. The cells were the pelleted down and cultured in medium containing 2 μg/ml puromycin for selection of stably transduced cells.

RNA from cell lines were isolated using TRIzol reagent by standard chloroform extraction method. Briefly, cells were pelleted down by centrifugation at 1,500 rpm for 5 min. Cells were lysed directly by adding 1 ml of trizol reagent and pipetting several times. The lysed cells along with TRIzol reagent were incubated at room temperature for 10 min 200 µl of chloroform per 1 ml of Trizol used was added followed by vigorous shaking for 10 times. The mixture was centrifuged at 13,200 rpm for 10 min at 4°C. The separated top aqueous phase was transferred to a new tube followed by RNA precipitation by adding 0.5 ml isopropanol. The RNA pellets were washed with 75% ethanol and semidried at room temperature. RNA pellets were dissolved in RNase free water by pipetting up and down followed by incubating at 55°C for 5 min 2 µg of RNA was used to synthesize cDNA using RevertAid First strand synthesis kit according to manufacturer protocol. cDNA was diluted 10 times and 2 µl (10 ng/μl) was used per reaction in the real-time PCR experiment. GAPDH was used endogenous control. The sequence of primers used in this study are as follows a) PDK1 Forward primer: 5′-GGA​TCA​GAA​ACC​GAC​ACA​AT-3’; PDK1 Reverse primer 5′-ACA​TTC​TGG​CTG​GTG​ACA​GG-3’ (product size 100 bp) b) P4HA1 Forward primer 5′-GTT​GGG​CAT​CCA​GTA​AAT​GC-3’; P4HA1 Reverse primer 5′-AGG​ACC​AGA​TTC​TCC​AAC​TC-3’ (product size 77 bp) c) ORF17 Forward primer (set 1) 5′-CGG​ATT​ATC​TCC​CAG​TCA​CA-3’; ORF17 Reverse primer (set 1) 5′-CTT​GAA​ATA​GAC​CCA​GTG​TC-3’ (product size 90 bp) d) ORF17 Forward primer (set 2) 5′-GGG​CGC​CGA​CGC​GGC​ACA​GT-3’; ORF17 Reverse primer (set 2) 5′-GGC​GCG​CTC​CGA​CTT​AGA​TA-3’ (product size 90 bp) e) GAPDH Forward primer 5′-ACC​CAG​AAG​ACT​GTG​GAT​GG-3’; GAPDH Reverse primer 5′- TCA​GCT​CAG​GGA​TGA​CCT​TG-3’ (product size 124 bp).

Cells were lysed in radio-immunoprecipitation buffer (50 mM Tris-HCl, 150 mM NaCl, 1 mM EDTA, 1% (v/v) NP-40) for 1 h with vertexing at regular interval of 10 min. The lysed cells were centrifuged at 4°C for 20 min at 15,000 rpm to remove cell debris. The clear supernatant was transferred to a new micro centrifuge tube. Concentration of proteins were measured by standard Bradford reagent. Equal amounts of proteins were separated on 10% polyacrylamide gel followed by transferring to nitrocellulose membrane. The membrane was stained with ponceau reagent to confirm the transfer of proteins. The membrane was washed witht tris-buffered saline containing tween-20 followed by blocking with 5% skimmed milk for 1 h at room temperature. The membrane was incubated overnight with primary antibody against PDK1 (Santa Cruz Biotechnology, sc-515944) or GAPDH (Santa Cruz Biotechnology, sc-47724) followed by washing with TBST for times, 5 min each. IR -conjugated secondary antibodies were used to probe proteins on Odyssys scanner. Immunofluorescence experiments were performed as described earlier (Kumar Singh et al., 2021). In brief, KSHV-positive BC3 cells were reactivated by treatment with TPA and butyric acid. The reactivated cells were collected by centrifugation, washed and semi-dried in a well of 8-well slides. The cells were fixed by dipping in 4% paraformaldehyde for 15 min at room temperature followed by washing with PBS. The permealization and blocking were performed in a single step by adding 3% bovine serum albumin supplemented with 0.1% triton X-100. The permealization and blocking was performed for 1 h at room temperature. Incubation with primary antibody was performed overnight at 4°C followed by 3 times washing with PBS. Secondary antibody incubation was performed for 1 h at room temperature and slides were washed thrice, 5 min each. Nuclei were counterstained with DAPI for 20 min at room temperature followed by washing with PBS. Excess of PBS was wiped out and mounting were performed using Prolog gold antifade reagent. The images were captured on confocal microscope (Olympus, Lambertville, NJ).

PDB ID 4P3H was used to represent KSHV-encoded protease ORF17. Discovery studio 2.5 (DS, Accelrys Inc., San Diego) was used to remove all the water molecule except from the catalytic region. Addition of hydrogen atoms and optimization of side chain was performed using Prepwizard (Schrodinger suite). Prediction of Ionizable groups at neutral pH was performed by PROPKA. H-bond optimization was performed by ProtAssign. Impref utility module was used for structure minimization allowing restrain on heavy atoms only. The centroid of ligand N-[2-Benzyl-4-(1 h-Tetrazol-5-Yl) phenyl]-6-(Cyclohexylmethyl) pyridine-2-Carboxamide is used as the grid center. Chemical compounds of biological origin were downloaded from ZINC database. A library of 307,814 compounds of biological origin has been used as a screen library. Based on ionization states (ionization states possible between pH 5–9) and specified chirality (possible stereoisomer for specified carbon), the number of total compounds were 481,799. The list of top 10 ligands after the virtual screening is provided in Table 1.

Docking experiments were performed through a virtual screen workflow which combines glide module and high throughput virtual screening (Glide HTVS) with single precision (Glide SP) and extra precision (Glide XP). Extra precision provides high stringency by utilizing extensive conformational sampling. Prism MMGBSA was used to calculate the binding energy of protein-ligand complex derived from the Glide XP docking experiment. The VSGB2 was used as an implicit solvation model. The MMGBSA dG binding score was used to evaluate complexes using the formula dG bind = E_complex (minimized)-E_ligand (minimized)-E_protein (minimized). Desmond molecular dynamics and OPLS force field (version OPLS_2005) were used for molecular dynamics simulation of top scoring complexes. 10 Ǻ orthorhombic box was used as boundary conditions and TIP3P was used as solvation model for the molecular dynamics simulation. Neutralization of charges were made by adding Na⁺ ions. 0.15 M sodium chloride was added as salt. 200 ns time frame was used for simulation run under NPT conditions. Simulations were run at 300 K at 1.01 bars. The total number of atoms in the molecular system is 25,702. The heating and equilibration stages were done utilizing isothermal-isochoric ensemble (NVT) ad isothermal-isobaric ensemble (NPT) with restraints on solute heavy atoms using the desmond default settings. Nose-Hoover chain has been used as thermostat, whereas Martyna-Tobias-Klein has been sed as barostat. The relaxation time for thermostat and barostat was 1 ps and 2 ps, respectively.

KSHV reactivation from naturally infected BC3 cells was performed as described earlier (Kumar Singh et al., 2021). Briefly, cells were grown at a concentration 5*10⁵ cells/ml in RPMI medium. Reactivation was initiated by adding TPA and butyric acid at a concentration of 5 μM/ml and 3 ng/ml respectively. For hypoxic reactivation, cells were grown under 1% oxygen condition for 72–96 h. ORF17 inhibitor ZINC00026154046 was added 12-h post reactivation with hypoxia or TPA/Butyric acid. Control and inhibitor containing cell culture flask were transferred to CO2 incubator and KSHV reactivation was allowed for 72-h. Cells along with medium were collected at the end of reactivation and centrifuged at 2000 rpm for 20 min to pellet down any cell debris. The clear supernatant was transferred into a 50 ml centrifuge tube, and the pellets were resuspended in 5 ml of PBS. The resuspended pellet was freeze-thawed thrice to break the cellular membrane and release intracellular KSHV virions, followed by centrifugation at 3,000 rpm for 30 min. All the supernatants were pooled together and filtered though 0.45 µM syringe filter and centrifuged at 25,000 rpm for 30 min to pellet down virions. Supernatant were aspirated and the pellets from different experiments were resuspended in equal volume of PBS. 200 µl of viral preparation was used for DNA isolation followed by copy number calculation. Briefly, equal volumes of 2X lysis buffer (20 mM Tris-pH 8.0, 2 mM EDTA and 300 mM NaCl) as added to viral preparation. SDS and proteinase K (final concentration 1% each) were added, and the samples were incubated at 55°C for 2 h. Standard phenol-chloroform extraction was used to isolate viral DNA. DNA pellet were dissolved in 200 µl DNase free water and 1 µl of DNA sample was used for KSHV copy number calculation as described earlier (Kumar Singh et al., 2021). KSHV copy number was calculated by the standard curve method using real-time PCR. Briefly, the known number of Cosmid construct containing KSHV genomic region (Co-ordinates: 85,820–100784) was used to generate a standard curve (30–300,000 copies in multiples of 10-fold). Genomic DNA isolated from purified KSHV virions from equal volumes of filtered supernatants were used to calculate the comparative number of KSHV generated in response of hypoxia or chemical chemically induced reactivation.

Experiments were performed at least in triplicate. All the statistical analyses of this study were performed using freely available GraphPad Prism software (GraphPad, San Diego, CA). The mean values with standard error of mean were presented in this study when appropriate. Statistical significance in expression of different sets of experiment was calculated by performing a 2-tailed Student’s t test. The p-value of ≤0.05 was considered statistically significant. *, p-value < 0.05; ns denote non-significant change.

KSHV-encoded protease (ORF17 protein) plays an indispensable role during KSHV assembly and maturation during its reactivation and hence an up-regulated expression of KSHV-encoded ORF17 is anticipated phenomenon during KSHV reactivation. Since hypoxic reactivation of KSHV represents one of the physiological possible modes of KSHV reactivation, we hypothesized that hypoxia could up-regulate expression of KSHV-encoded ORF17. We performed a quantitative real-time PCR for investigating levels of ORF17 transcripts in KSHV positive cells grown under hypoxic conditions as compared to normoxic conditions. In brief, KSHV positive BC3 cells were grown in normoxic or hypoxic conditions for various time points followed by RNA isolation and cDNA synthesis. The confirmation of induction of hypoxia was made by investigating real-time expression of hypoxic markers such as PDK1 and P4HA1 (Figures 1A, B). A real-time PCR based method was used to determine the differential expression of ORF17 in the cells grown in hypoxic conditions as compared to their counterpart normoxic conditions. An up-regulation in the expression of ORF17 of up to approximately 5-fold (±1.5) at longer time periods was observed in BC3 cells (Figures 1C, D). Further, we wanted to confirm if the up-regulated expression ORF17 is a universal characteristic during KSHV reactivation and co-relate with viral production in reactivated conditions. We reactivated KSHV positive BC3 cells by treatment with butyric acid and TPA. The reactivation of BC3 cells was confirmed by immunofluorescence against KSHV-encoded RTA (Figure 1E). A real-time PCR based experiment to investigate expression profile of ORF17 in chemically reactivated BC3 cells showed that expression of ORF17 is even higher compared to the hypoxic reactivation (Figure 1F). We finally investigated if a direct co-relation between ORF17 expression and production of KSHV virions exists in reactivated condition. The results clearly indicated a proportional relationship between ORF17 expression and production of KSHV virions (Figure 1G). Altogether, these results confirmed that up-regulated expression of ORF17 is an associated event during hypoxic/chemically induced reactivation.

HIF1α is a direct effector protein involved in hypoxic conditions. In hypoxic conditions, HIF1α stabilization occurs due to inactivation of vHL protein responsible for proteasomal degradation of HIF1α. In stabilized condition, HIF1α transactivate expression of several genes responsible for cell survival, metabolic reprogramming. These ensure hypoxic cells to withstand the negative impact of hypoxia till normoxic favorable conditions restores. The transactivation of genes by HIF1α depends on the presence of specific DNA consensus sequences [N-ASGT-NN (S=C/G)] called as hypoxia responsive elements. We screened 1,000 bp upstream sequence in the promoter region of ORF17 to find if any HRE is/are present in this sequence. Based on the presence of nine distinct HREs within the promoter region of ORF17 (Figure 2A), we investigated if ectopic expression of HIF1α can up-regulate expression of ORF17. Cells were transduced with either mock (GFP) or GFP-HIF1α expressing lentiviruses and successfully transduced cells were selected in puromycin containing medium. The expression of HIF1α was confirmed by fluorescence microscopy (Figure 2B). A comparative study for real-time expression of ORF17 in mock or HIF1α expressing cells showed that, ectopic expression of HIF1α can lead to up-regulated expression of ORF17 (Figure 2C). Although, the magnitude of elevated expression of ORF17 in this condition was lower as compared to cells grown under hypoxic conditions. To further corroborate these observations, we knocked down HIF1α in KSHV positive cells by ShRNA expressing lentiviruses. The successfully transduced cells were selected in puromycin containing medium and visualized by GFP fluorescence (Figure 2D). The cells were grown under either normoxic or hypoxic conditions followed by investigating expression of KSHV-encoded ORF17 in these cells. As expected, ORF17 expression was lower but statistically non-significant in ShHIF1α expressing cells (Figure 2E; *<0.05, ns = non-significant).

Based on direct role of KSHV- encoded protease, we wanted to screen a natural product which can be utilized as its inhibitor for targeting KSHV-encoded protease and hence the reactivation. In a grid defined experimental setup, docking was performed as mentioned in materials and method section. A library of 307,814 natural compounds of biological origin has been used as a screen library. Based on ionization states (ionization states possible between pH 5–9) and specified chirality (possible stereoisomer for specified carbon), the number of total compounds were 481,799. Several parameters based on scoring functions, such as docking score, energy score, XP Gscore, glide gscore, were used to identify aa dominant inhibitor of KSHV-encoded protease ORF17. Based on the docking score, top 10 compounds are shown in Table 1. ZINC000261524046 (original conformer; 1-oleoyl-2-hydroxy-sn-glycero-3-phospho-(1′-myo-inositol was selected for further studies based on its highest docking score. 1-oleoyl-2-hydroxy-sn-glycero-3-phospho-(1′-myo-inositol), is an easily available lysophosphatidic acid and showed a high binding and hence inhibitory characteristics. A RMSD, radius of gyration, MolSA, SASA and PSA study was performed to investigate ligand properties (Figure 3A). The molecular Interaction of ligand with protease is shown in Figure 3B. Briefly, the bonding and characteristics such as charged (positive/negative), glycine, hydrophobic, polar, water hydration sites, hydrogen bond, halogen bond, metal co-ordinates, Pi-cation, salt bridge and solvent exposure were utilized (Figure 3B). The main hydrogen bond at His88, Thr86 and Arg82 were identified (Figure 3B). We further performed protein RMSD, protein RMSF, interaction fraction and protein contact studies. A protein RMSD plot for 200 nsec is shown in Figure 4A. RMSF values till 200 proteins residues in shown in Figure 4B. The interaction fraction of important residues in the terms of hydrogen bonding, hydrophobic bonding, ionic bonding and water bridge is shown in Figure 4C. The important hydrophobic bonding regions were concentrated near PHE76, LEU79, ALA80, ALA90 and THR109. Similarly, hydrogen bonding’s were maximum observed for THR86, SER87, HIS88 and VAL89 (Figure 4C). A timeline representation for contact and interactions are shown in Figure 4D.

Based on the in-vitro results showing inhibitory effect of KSHV-encoded protease by LPA, we wanted to validate this result in cell culture-based experiments. Frist, the cytotoxicity of LPA on KSHV-positive BC3 cells was investigated by treating these cells with increasing concentration (1–100 µM) for 24–72 h (Figure 5A). The 1 μM and 10 µM concentration showed minimal cytotoxic effect on BC3 cells, with retained viability of 95 ± 3% at 24 h and 85 ± 5% at 72 h. However, 100 µM concentration of LPA showed approximately 85 ± 5% even at 24 h of treatment. Treatment of these cells with 50 µM of LPA showed an intermediate effect on cell viability (Figure 5A). Based on the viability test results, we choose to perform further experiments using 1 µM concentration of LPA. BC3 cells were reactivated by either growing under hypoxic conditions (1%O₂) or by adding TPA/BA in the growing culture. Induction of Hypoxia was confirmed by PDK1 Western blot (Figure 5B). KSHV reactivation in response to TPA/BA was confirmed by immunofluorescence against KSHV-encoded RTA (Figure 5C). 12-h post reactivation, LPA at a final concentration of 1 µM was added to the cells. For KSHV copy number calculation from extra cellular medium of reactivated cells, a standard curve with known number of KSHV genomic region cloned in cosmid vector was generated (Figure 5D). In brief, clone of KSHV genomic region encompassing 85,000-102000 bp in cosmid was serially diluted to obtained 30 to 30⁵ copies of cloned region per µl of solution. 1µl sample from each of the dilution was used to generate a standard curve for the known number of KSHV specific DNA. The extra-cellular medium from the reactivated cells were collected and excreted KSHV virions were collected by centrifugation followed by isolation of viral DNA. Equal DNA was used for copy number calculation. The superimposed image for the standard curve and the copy number calculation is showed in Figure 5D. The results showed a statistically significant reduction in the production of KSHV virions in the cells treated with LPA. Interestingly, the results also suggested that hypoxic reactivation is very less efficient as compared to TPA/BA mediated reactivation. Nevertheless, LPA inhibited production of KSHV virions in hypoxia reactivated cells in the same manner as that of TPA/BA activated cells.