Concurrent chemotherapy inhibits Herpes simplex virus 1 replication and oncolysis

Cells and viruses

Vero (African Monkey kidney), HT29 and SW620 (human colon carcinoma), Capan2 (human pancreatic cancer) and 293 (human kidney) were obtained from American Type Culture Collection (Rockville, MD). 0–28 [ICP (infected cell protein) 0-transformed Vero cells] and E5 (ICP4-transformed Vero cells) were kindly provided by E. Antonio Chiocca (Massachusetts General Hospital, Boston, MA). V27 (ICP27-transformed Vero cells) were kindly provided by David Knipe (Harvard Medical School, Boston, MA). Cells were propagated in DMEM with 10% fetal bovine serum (FBS), 100 units/ml penicillin, and 100 mg/ml streptomycin. hrR3 (ICP6-defective HSV-1 mutant)²⁴ was kindly provided by Sandra Weller (University of Connecticut, Farmington, CT). 7134 (ICP0-defective HSV-1 mutant)¹⁹ and gal4 (ICP4-defective HSV-1 mutant)²³ were kindly provided by E. Antonio Chiocca (Massachusetts General Hospital). d27 (ICP27-defective HSV-1 mutant)²⁵ were kindly provided by David Knipe. Wild-type HSV-1 strain KOS²⁶ was kindly provided by Donald Coen (Harvard Medical School). R3616 (γ₁34.5- defective HSV-1 mutant)¹³ was kindly provided by Bernard Roizman (University of Chicago, Chicago, IL). Heat-inactivation of viruses was performed as described.²⁷ The drugs used in cell cultures studies were tumor necrosis factor-alpha (TNF-α, SIGMA, St. Louis, MO), methotrexate (MTX, Ben Vanue Laboratories, Inc., Bedford, OH), 5-fluorouracil (5-FU, Pharmacia & Upjohn Co., Kalamazoo, MI) and irinotecan (CPT-11, Pharmacia & Upjohn Co., Kalamazoo, MI).

The recombinant adenovirus vectors used in this study were replication-defective Ad5-based vectors constructed with the transgene expression driven by the CMV early/intermediate promoter/enhancer. All vectors were expanded in 293 cells and purified and titered as described previously²⁸. The vector Ad.CMV.IκBα expresses the super-repressor form of IκBα that is mutated at serine residues 32 and 36 and functions as a potent and specific repressor of NF-κB-mediated events.^{28, 29} The control vector Ad.CMV3, generously provided by J. A. Roth (University of Texas M.D. Anderson Cancer Center, Houston, TX), contains a CMV promoter similar to Ad.CMV.IκBBα but lacks a transgene insert.

In vitro viral cytotoxicity and replication assays

Viral replication assays were performed by infecting 10⁶ cells with HSV-1 for 2 h, at which time unabsorbed virus was removed by washing with a glycine-saline solution (pH = 3.0). At 0, 8, 16, 24, 32 and 40 h after infection, the supernatant and cells were exposed to three freeze/thaw cycles to release virions and titered on Vero cells or 0–28 cells. Viral cytotoxicity was determined as described previously.¹ Briefly, 5000 cells/well were plated onto 96-well plates and grown for 36 h. Cells were then infected with either KOS or HSV-1 mutants using multiplicity of infection (MOI) values ranging from 0.001 to 10. The number of viable cells was determined using a colorimetric 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. All experiments were performed in quadruplicate.

Electrophoretic mobility shift assay for NF -κB activation

Activation of NF-κB in response to infection with HSV-1 or treatment with chemotherapeutic drugs was determined by the electrophoretic mobility shift assay (EMSA). HT29 cells (5×10⁶ cells/ plate) were cultured in 100 mm dishes. Twenty-four hours later, cells were infected with HSV-1 or treated with chemotherapeutic drugs. Cells were then harvested at times 0, 3, 6, 12, 24 and 36 h after treatment. For experiments using the adenoviral-mediated delivery of the IκBα supper-repressor (IκBα-SR) gene, HT29 cells were infected with Ad.CMV.IκBα or Ad.CMV3 (MOI = 20) for 1 h and then washed with PBS. Twelve hours after adenovirus infection, cells were infected with HSV-1 (KOS, 7134 and hrR3) for 1 h and then washed with PBS. Cells were then harvested 0, 8, 16, 24, 32 and 40 h after HSV-1 infections and resuspended in 3 volumes of C.E. Buffer [10 mM Hepes (pH = 7.6), 60 mM KCl, 1 mM EDTA, 10 mM 1,4-Dithio-L-threitol (DTT)] with 0.1% NP-40 and proteinase inhibitors (0.5 mM phenylmethylsulfonyl fluoride, 2.5 µg/ml aprotinin, 2.5 µg/ml pepstain, and 2.5 µg/ml leupeptin). Nuclear pellets were collected at 3000 rpm (4° C) to which 100µl of C.E. buffer was added. After centrifugation at 3000 rpm (4° C), the pellet was resuspended in 2 pellet volumes of N.E. Buffer [20 mM Tris (pH = 8.0), 420 mM NaCl, 1.5 mM MgCl₂, 0.2 mM EDTA, and 25% glycerol] with proteinase inhibitors, and then added 60 µl of 5M NaCl. The soluble protein was released by centrifugation (20 min at 14,000 rpm), and stored at −80° C. The amount of protein in the supernatant was quantified using a BCA protein assay kit (Pierce Chemical Co., Rockford, IL). The DNA probe used contains an NF-κB site (underlined) from the H-2κ^b gene (5’-CAG GGC TGGGGA TTC CCA TCT CCA CAG TTT CAC TTC-3’)²⁸ Ten µg of nuclear extracts were preincubated with 1 µg of poly deoxyinosinic-deoxycytidylic acid [poly (dI-dC)•poly (dI-dC); Amersham Pharmacia Biotech Inc., Piscataway, NJ] in binding buffer (10 mM Tris, 50 mM NaCl, 20% glycerol, 0.5 mM EDTA, and 1 mM DTT) for 10 min at room temperature. Approximately 20,000 cpm of ³²P-labeled DNA probe was then added and allowed to bind for 15 min. The complexes were then separated on a 1% agarose/0.25% Synergel (Diversified Biotech, Boston, MA) in a 0.5× TBE buffer and autoradiographed.³⁰

Purification of bacteria-expressed PKR and eIF-2α

Production of PKR and eIF-2α were performed as described previously.⁶ In brief, Escherichia coli BL21 cells harboring the pGEX-PKR and pQE-eIF-2α expression vectors (kindly provided by Bryan R.G. Williams, Lerner Research Institute, Cleveland, OH) were grown overnight in 50 ml Luria-Bertani (LB) broth containing 50 µg/ml ampicillin. Following 1:10 dilution in fresh LB broth, cells were grown for 3 h to an optical density of 0.8, at which time isopropyl-1-thio-β-D-galactopyranoside (IPTG) was added to a concentration of 1 mM for an additional 4 h. Bacteria were pelleted and used for the fusion protein purification. To purify the His-tagged eIF-2α protein, a His•Bind purification Kit (Novagen Inc., Madison, WI) was used following the manufacturer’s instructions. To purify the GST-tagged PKR, a GST•Bind purification Kit (Novagen Inc.) was used according to the manufacturer’s instructions.

In vitro phosphorylation assays

The PKR autophosphorylation assay was performed as follows. HT29 cells were harvested at 24 h after treatment with 5-FU (100 µM), CPT-11 (1 µM), MTX (25 µM), and TNF-α (10 ng/ml). The cells were rinsed with phosphate-buffered saline (PBS), resuspended in lysis buffer containing 10 mM HEPES (pH = 7.6), 150 mM NaCl, 10 mM MgCl₂, 0.2% Triton X-100, 10% glycerol, 0.5 mM phenylmethylsulfonyl fluoride, and 2 mM benzamide, placed on ice for 30 min and subjected centrifugation to remove nuclei. The cell lysates were reacted with [γ³²P] ATP (100 µCi/sample) for 30 min at 32° C, precleared with protein A-agarose (Roche Diagnostics Corp., Indianapolis, IL) and reacted with 1 µg of antibody to PKR (K-17) (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) or without the PKR-antibody for negative control. These complexes were precipitated using protein-A agarose, rinsed with RIPA buffer (PBS containing 0.25% NP-40, 0.5% sodium deoxycholate, and 0.1% sodium dodecyl sulfate), solubilized in buffer C [20 mM Tris-HCl (pH = 7.0), 50 mM KCl and 2 mM MgCl₂], electrophoretically separated in a 10% SDS-PAGE gels, and subjected to autoradiography. For eIF-2α phosphorylation assay, 2 µl of purified eIF-2α protein (0.32 µg/µl) was added to the kinase reaction [2 µl of cell lysate (0.32 µg/ l), 7 µl of buffer C, and 1 µl of [γ³²P]ATP (10 µCi/µl)]. These reactions were incubated at 32° C for 40 min before the addition of 2 × SDS loading buffer (12 µl) was added to each sample before boiling, separation by 10% SDS-PAGE, and autoradiography. The PKR-³²P or eIF-2α-³²P was quantified by an image analyzer (LabWorks 4.0; UVP Inc., Upland, CA).

In vitro eIF-2α dephosphorylation assays

Confluent 100 mm plate of HT29 cells were harvested 24 h after treatment with chemotherapeutic drugs [5-FU (100 µM), CPT-11 (1 µM), MTX (25 µM), and TNF-α (10 ng/ml)] in the presence or absence of HSV-1 KOS (MOI = 1) infection. Cells were once rinsed with PBS and added to 100 µl of cell lysate buffer and placed on ice for 30 min and then centrifuged. The amount of protein in supernatant was quantified using a BCA protein assay kit (Pierce Chemical Co.). Purified eIF-2α protein (12.5 µl; 0.32 µg/µl) was reacted with GST-PKR protein (12.5 µl; 0.32 µg/µl) in 38 µl of buffer C and 7 µl of [γ³²P]ATP (10 µCi/µl) in a final volume of 70 µl for 40 min at 32° C to yield phosphorylated eIF-2α. After 40 min at 32° C, 8 µl of phosphorylated eIF-2α (eIF-2α-³²P) was mixed with 10 µl of the lysates and 7 µl of buffer C. Following incubation at room temperature for different time intervals, 10 µl aliquots were removed at specific time point, mixed with 2 × SDS loading buffer (10 µl), boiled for 3 min and analyzed by SDS-PAGE. The ³²P remaining in eIF-2α-³²P was quantified by an image analyzer (LabWorks 4.0).

Statistical analysis

The mean values of the cell survival rates from the cytotoxicity assay, viral yields from the viral replication assay and optical density from the EMSA of NF-κB, PKR/eIF-2α phosphorylation assay or eIF-2α dephosphorylation assay were compared by two sided Student’s t test. Statistical analysis and calculation of the regression coefficients were performed using StatView software (SAS Institute Inc., Gary, NC).

Chemotherapy-induced cytotoxicity inhibits viral replication

We initially determined the relationship between concentration of chemotherapy agents and cytoxicity in colon and pancreatic carcinoma cells. We included TNF-α in these studies because of the known effect of TNF-α on viral replication, NF-κB, and PKR. As expected, increasing drug concentrations were associated with increasing cytotoxicity for all cell lines (Fig. 1A). We then infected these cells with HSV-1 in three conditions: no chemotherapy, a low concentration that produced no cytotoxicity alone, and a higher concentration that induced cytotoxicity alone. In all three cell lines, cytotoxicity induced by HSV-1 alone was no different than cyotoxicity induced by HSV-1 combined with a low concentration of chemotherapy or TNF-α (Fig 1B). Exposure to a higher concentration of chemotherapy combined with HSV-1 infection did not produce more cytotoxicity that either HSV-1 alone or drug alone, suggesting that chemotherapy agents antagonized viral oncolysis. We observed identical results with other HSV-1 mutants, including 7134 (ICP0-defective) and hrR3 (ICP6-defective) (data not shown). To investigate the relationship between viral replication and cellular responses to chemotherapy, we used a burst assay to measure HSV-1 replication in cells treated with TNF-α, 5-FU, CPT-11, and MTX. KOS replication is significantly inhibited by all of the drugs (Fig. 1C). Similar results were observed with HSV- strains 7134 and hrR3 (data not shown).

NF-κB activation is required for efficient HSV-1 replication

The NF-κB pathway provides an attractive target to viral pathogens; many viruses exploit the anti-apoptotic properties of NF-κB to evade host defense mechanisms.¹⁶ Beginning at around 3 h after infection, KOS-infected HT29 cells demonstrate inducible NF-κB activation compared to mock-infected cells (Fig. 2A). NF-κB nuclear translocation increases over time, leading a maximum at 24 h. HT29 cells infected with HSV-1 mutants that are defective in ICP0, ICP4, and ICP27 (7134, gal4, and d27, respectively) do not activate NF-κB as strongly as cells infected with KOS. In contrast, the ICP6-defective HSV-1 mutant (hrR3) and γ₁34.5-defective HSV-1 mutant (R3616) induce NF-κB activation as effectively as cells infected with KOS.

To further examine the relationship between HSV-1 replication and NF-κB activation, we used a serine-mutated form of IκBα, which does not undergo phosphorylation at key sites and is therefore not targeted for polyubiquitination and degradation by the proteosomes.^{17, 29} Adenoviral-mediated delivery of a gene encoding IκBαα supper-repressor (IκBα-SR) effectively blocks NF-κB.^{17, 29} We first confirmed that infection of HT29 cells with a replication-defective adenovirus expressing IκBα-SR (Ad.CMV.IκBα) 12 h prior to HSV-1 infection blocks NF-κB activation (Fig. 4A). To determine whether inhibition of NF-κB activation affects virus replication, we compared viral replication in IκBα-SR-expressing HT29 cells relative to control cells. HSV-1 replication in the presence of the control adenovirus (Ad.CMV3) is no different than replication observed in saline-treated control cells. In contrast, HSV-1 replication in the presence of the adenovirus expressing the IκBα-SR (Ad.CMV.IκBα) is significantly attenuated for each HSV-1 virus (Fig. 4B). NF-κB activation appears critical for robust viral replication, because inhibition of NF-κB activation markedly attenuates HSV-1 replication. This effect was apparent in all three cell lines examined, with only minor variability observed between KOS, 7134, and hrR3 (Fig. 4C, D).

Differential effects of chemotherapeutic agents on NF-κB activation

Exposure of colon cancer cells to 5-FU, CPT-11, and TNF-α induces NF-κB activation in a manner that is both time-dependent and dose-dependent (Figs. 3A and 3B). In contrast, exposure to MTX does not activate NF-κB. We examined the effect of chemotherapeutic agents on HSV-1-induced NF-κB activation. Treatment with TNF-α, 5-FU or CPT-11 during KOS infection does not alter NF-κB activation compared to KOS infection alone (Fig. 3C). This finding suggests that the mechanism by which 5-FU and CPT-11 inhibits HSV-1 replication is independent of their effects on NF-κB activation. The mechanism by which MTX inhibits viral replication may be through its ability to inhibit of NF-κB activation.

Chemotherapeutic agents induce PKR/eIF-2α phosphorylation

Eukaryotic viruses and their multicellular hosts have coevolved complex interrelationships to permit virus reproduction without destruction of the host.^{10, 11} A central enzyme involved in antiviral activity is the PKR.^{10, 11} One of the major biological functions of PKR is to regulate protein translation through phosphorylation of eIF-2α¹¹ HSV-1 γ₁34.5 expression results in eIF-2α dephosphorylation, which aids efficient HSV-1 replication,¹³ and we have previously demonstrated the importance of eIF-2α dephosphorylation for viral oncolysis.⁶ Accordingly, we measured eIF-2α phosphorylation induced by PKR in response to TNF-α, 5-FU, CPT-11 or MTX. Protein lysates of HT29 cells treated with these agents were examined for their ability to phosphorylate PKR, phosphorylate eIF-2α and dephosphorylate eIF-2α in the presence or absence of HSV-1 infection (KOS). KOS stimulates PKR phosphorylation, as do each of the agents; however, the PKR phosphorylation by KOS infection is unchanged by TNF-α, 5-FU, CPT-11 or MTX (Fig. 4A). Similarly, eIF-2α phosphorylation observed in response to each of the agents is unchanged by (Fig. 4B). CPT-11 and MTX induce some eIF-2α dephosphorylation activity, but overall dephosphorylation activity in response to HSV-1 infection is not altered by any of the agents (Fig. 4C). Thus, chemotherapy induces PKR and eIF-2α phosphorylation, but HSV-1-mediated dephosphorylation of these regulatory proteins remains active. Overall, these results suggest that HSV-1 induced PKR phosphorylation and eIF-2α dephosphorylation remain robust in chemotherapy treated cells, yet viral replication is attenuated in the presence of some of these agents.