Oncolytic virus expressing PD-1 inhibitors activates a collaborative intratumoral immune response to control tumor and synergizes with CTLA-4 or TIM-3 blockade

Background Oncolytic viruses (OVs) are capable to inflame the tumor microenvironment (TME) and elicit infiltrating tumor-specific T cell responses. However, OV treatment negatively alters the cancer-immune set point in tumors to attenuate the antitumor immune response, which suggests the necessity of dissecting the immune landscape of the virus-treated tumors and developing novel strategies to maximize the potential of OVs. The aim of this study is to investigate the effect of the single-chain variable fragment (scFv)-armed OVs targeting PD-1 on the TME, and ultimately overcome localized immunosuppression to sensitize tumors to immunotherapies. Methods A tumor-selective oncolytic herpes simplex virus vector was engineered to encode a humanized scFv against human PD-1 (hPD-1scFv) (YST-OVH). The antitumor efficacy of YST-OVH was explored in multiple therapeutic mouse models. The neurotoxicity and safety of YST-OVH were evaluated in nonhuman primates. The precise dynamics in the TME involved in YST-OVH treatment were dissected using cytometry by time-of-flight (CyTOF). Results The identified hPD-1scFv showed superior T-cell activating activity. Localized delivery of hPD-1scFv by YST-OVH promotes systemic antitumor immunity in humanized PD-1 mouse models of established cancer. Immune profiling of tumors using CyTOF revealed the enhanced antitumor effect of YST-OVH, which largely relied on CD8+ T cell activity by augmenting the tumor infiltration of effector CD8+ T cells and establishment of memory CD8+ T cells and reducing associated CD8+ T cell exhaustion. Furthermore, YST-OVH treatment modified the cancer-immune set point of tumors coupled to coexpression of CTLA-4 and TIM-3 on exhausted CD8+ T cells and high levels of CTLA-4+ Treg cells. A combination approach incorporating anti-CTLA-4 or anti-TIM-3 further improved efficacy by increasing tumor immunogenicity and activating antitumor adaptive immune responses. Moreover, this therapeutic strategy showed no neurotoxicity and was well tolerated in nonhuman primates. The benefit of intratumoral hPD-1scFv expression was also observed in humanized mice bearing human cancer cells. Conclusion Localized delivery of PD-1 inhibitors by engineered YST-OVH was a highly effective and safe strategy for cancer immunotherapy. YST-OVH also synergized with CTLA-4 or TIM-3 blockade to enhance the immune response to cancer. These data provide a strong rationale for further clinical evaluation of this novel therapeutic approach.


PD-1 antibody production
Six-week-old female BALB/c mice were maintained in individually ventilated cages under specific pathogen free condition and immunized with 1×10 7 293T/hPD-1 cells in 500 µl of PBS at multiple sites by standard vaccination procedures.The hybridoma generation procedure followed the protocol reported previously. 3The positive clones were selected by an indirect ELISA assay coated with recombinant human PD-1 protein, and subjected to colonized culture for obtaining stable single clones.MAbs were prepared by ascites production and purified by using protein G chromatography according to the manufacturer's instructions.

Antibody evaluation
The binding ability of full-length anti-PD-1 antibodies and hPD-1scFv was determined by an indirect chemiluminescence immunoassay (CEIA).Briefly, 96-well plates were coated with 10 ng/well of hPD-1-His protein (Sino Biological), and nonspecific binding was blocked with PBS containing 5% BSA.Purified anti-PD-1 antibodies, hPD-1scFv, supernatants from YST-OVHinfected or non-infected cells, or serum from YST-OVH-treated or non-treated mouse were added to the wells for 1 h incubation, followed by washing and reaction with anti-His-HRP antibody (Proteintech).After the addition of Luminol substrates (Wantai BioPharm) for 5 min, the plates were measured with a chemiluminescence reader (Berthod).For detection of hPD-1scFv in tumors, the tumors were homogenized with PBS by gentleMACS tissue dissociator (Miltenyi BioTec) according to the manufacturer's instructions, and the lysates were assayed using the abovementioned method.For quantitative detection of hPD-1scFv, a serial dilution of purified hPD-1scFv was used to generate a standard curve.
The blocking activity of full-length anti-PD-1 antibodies and hPD-1scFv was determined using a blocking CEIA assay.Briefly, 96-well plates were coated with 100 ng/well of hPD-1-His protein, and nonspecific binding was blocked with PBS containing 5% BSA.Purified hPD-1scFv or anti-PD-1 antibodies were first diluted from 10 µg/ml in PBS containing 20% CBS, followed by twofold serial dilutions to eight gradients.Then, each dilution together with 100 ng/well biotinylated hPD-L1-Fc (Sino Biological) were added to the wells for 1 h incubation, followed by washing, reaction and detection.The inhibitory ratio was calculated as follows: %inhibitory = 100× (1− (average value for each dilution/average value for control)).The results were analyzed by nonlinear, dose-response regression analysis using GraphPad Prism 7 software.

RNA isolation and gene amplification
Total RNA was extracted with TRIzol Reagent (Invitrogen) and reverse-transcribed into cDNA with random primers by SuperScript® Ⅲ Reverse Transcriptase (Invitrogen).The cDNA products were then used as templates for PCR amplification of the variable regions of the heavy chain and light chain with KOD-Plus polymerase (TOYOBO) using an Ig-Prime kit (Novagen).The variable sequences of the heavy chain and light chain were sequenced and cloned into pFUSE antibody cassettes (InvivoGen) for further development.

Humanization of the anti-PD-1 antibody 17D5
Humanization of an anti-PD-1 antibody by CDR grafting and combinatorial library screening was performed as previously described. 4Humanized antibodies were first evaluated by testing their binding reactivity with a recombinant hPD-1 protein and ability to block the interaction between PD-1 and PD-L1 and then examined for their humanness score.293T cells were used to express humanized anti-PD1 antibodies by transfection of recombinant humanized antibody cassettes in FreeStyle 293 Expression Medium, and hu17D5 was purified by using protein G chromatography according to the manufacturer's instructions.

hPD-1 scFv construction and expression
Humanized anti-PD1 antibody 17D5 (hu17D5) was selected for scFv construction, the fragment encoding scFv against hPD-1 consisted of a secretion signal peptide, light chain, 3×G4S, heavy chain and 8×His tag, which were first amplified from hu17D5 cDNA and sequentially assembled into pTT5 vector (CNRC) under the control of the human cytomegalovirus promoter.293T cells were used to express hPD-1scFv by transfection of PTT5-hPD-1scFv in FreeStyle 293 Expression Medium, hPD-1scFv was purified by using Ni-NTA chromatography (GE Healthcare) according to the manufacturer's instructions.hPD-1scFv was eluted using 250 mM Imidazole and dialyzed in PBS.

T cell activation assays
The T cell-activating activity of full length anti-PD-1 antibodies and hPD-1scFv was determined using a PD-L1 overexpressing cells and PBMCs co-culture assay.Human PBMCs of a healthy donor (5×10 4 per well) (Lonza) were stimulated with anti-CD3/CD28-coated beads (Miltenyi) or cultured without stimulation for 2 days in a 96-well plate.Then, 1×10 4 293T/hPD-L1 cells and 0.001 to 10 μg/ml of anti-PD-1 antibodies and hPD-1scFv were added and cultured at 37℃ for 3 days.The IL-2 levels in the supernatant were assayed by Human IL-2 Quantikine ELISA Kit (R&D Systems).Co-culture of tumor cells with Jurkat cells 1×10 4 irradiated MDA-MB-231 cells and 5x10 3 Jurkat-Lucia™ NFAT cells were seeded into wells.

Cytotoxicity assay
Cells were seeded in 96-well plates at 1×10 4 cells per well in 100 µl of complete medium and infected with the indicated virus or mock-infected.After treatment for 72 h, 10 µl of CCK-8 (MCE) was added to the cells, and cells could grow at 37°C for another 4 h.The optical absorbance was determined at 450 nm using a microplate reader (Thermo Scientific).

Western blot analysis
Cells were lysed in RIPA lysis buffer (Thermo Scientific) containing protease inhibitor cocktail (MCE).Cell lysates were separated by SDS-PAGE and transferred onto a nitrocellulose membrane.
After membranes were blocked with blocking buffer containing 5% BSA for 1 h, they were probed with the indicated primary antibodies overnight at 4℃, followed by incubation with HRPconjugated secondary antibodies for 1 h at room temperature, and finally visualized with the Lumi-Light PLUS Western blotting Substrate (Roche).

Tumor-infiltrating lymphocytes (TILs) isolation and analysis
For isolation of TILs, mice were sacrificed and the tumors were harvested on days 7 after two doses of the indicated treatment.Tumors were dissected and dissociated with lysis buffer (1 mg/ml Collagenase D (Sigma) and 100 μg/ml DNase I (Sigma) in RPMI1640 medium with 2% FBS) for 2 h with continuous agitation.Digestion mixture were homogenized by repeated pipetting and filtered through a 70-μm filter.Cell suspensions were first stained with Zombie Aqua™ Fixable Viability Kit to eliminate dead cells, stained with the corresponding antibodies, incubated for 30 minutes at 4°C, and then subjected to FCM analysis.

RNA-seq analysis of tumors
Tumor cells (5×10 6 ) in 100 μl of PBS were subcutaneously inoculated into the right flank of PD-1-HU mice and allowed to establish for 6 days.Once the tumors reached the indicated volume, 10 7 PFU of virus and/or 200 µg anti-CTLA-4 were injected and then repeatedly injected on days 3 and 6 after the initial treatment.On day 3 after the last treatment, the tumors were harvested and stored at -80℃.Biological triplicates for each group were used to extract total RNA using an RNeasy kit (Qiagen), and RNA quality was measured using an Agilent 2100 Bioanalyzer (Agilent Technologies).One microgram of total RNA with an RIN value above 6.5 was used for subsequent library preparation.Then, libraries with different indices were multiplexed and loaded on an Illumina HiSeq instrument according to the manufacturer's instructions (Illumina).Sequencing was carried out using a 2×150 bp paired-end (PE) configuration; image analysis and base calling were conducted by HiSeq Control Software (HCS) + OLB + GAPipeline-1.6(Illumina) on a HiSeq instrument.RNA-seq analysis was performed by GeneWiz.The change in gene expression was calculated for each gene in relation to an untreated control.GO enrichment analysis was performed using ClusterProfiler, and the top 10 most enriched pathways are shown in a plot.Heatmaps of normalized expression values of the enriched gene clusters identified by GO analysis were generated using the R package pheatmap.

Histology
For hematoxylin and eosin (H&E) staining, tissues were fixed in 10% formalin, embedded in paraffin, and stained with H&E.Then the sections were washed, and mounted with cover slips.
Images were captured with a research-level upright microscope (Olympus).

Cell profiling using CyTOF
Single-cell suspensions of TILs were prepared as described previously.Tumors were mechanically disrupted and digested for 2 h at 37°C in lysis buffer.The released cells were pelleted, resuspended BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) J Immunother Cancer doi: 10.1136/jitc-2022-004762 :e004762.10 2024; J Immunother Cancer , et al.Ju F in 3 ml of ACK Lysing Buffer (Thermo Fisher Scientific) for 2 min to lyse red blood cells at room temperature and then filtered through a 70-μm cell strainer.TILs and tumor cells were separated using Percoll gradient centrifugation.Immune cells were enriched from the 40%/70% Percoll interface.For the CyTOF assay, unconjugated antibodies (Table S5) were obtained from Fluidigm and conjugated in-house using the Maxpar R X8 Multimetal Labeling Kit (Fluidigm) according to the manufacturer's instructions.Briefly, single-cell suspensions were stained with 1 μM cisplatin (Fluidigm) for 15 min and then blocked with Fc receptor blocking buffer (Biolegend) for 10 min at room temperature.Cells from each sample were incubated with a metal-conjugated surface antibody cocktail on ice for 30 min and barcoded with a unique combination of palladium metal barcodes according to the manufacturer's instructions (Fluidigm).Next, the cells were pooled together, followed by fixation and permeabilization using the Nuclear Staining Buffer Set (Fluidigm).The cells were subsequently stained with a metal-conjugated intracellular antibody cocktail for 30 min at 4℃.After washing, the cells were incubated in 1 ml of intercalator buffer (0.125 nM MaxPar Intercalator-Ir in 1 ml of Fixation & Permeabilization Buffer).Prior to acquisition, the cells were diluted to 8×10 5 cells/ml in deionized water containing 10% EQ Four Element Calibration Beads (Fluidigm) and filtered through a 70-μm nylon filter.Events were acquired on a CyTOF 2 Helios upgraded mass cytometer at an event rate of 200-300 cells/second at the Flow Cytometry and Cellular Imaging Facility of Xiamen University..fcsfiles were normalized to the EQ 4-element bead signal (Lot P15K0802, Passport EQ 4_P13H2302) in 100-s interval windows using normalization software (version 6.7.1014,Fluidigm).Mass tag barcodes were resolved with a doublet filtering scheme using Debarcoder (Fluidigm).Live immune cells were manually gated with FlowJo by event length, live/dead discrimination, and the expression status of CD45.The data were then exported for downstream analysis and transformed with a coefficient of 5 with the cytofAsinh method.For downstream analyses, individual sample data were subsampled to 10000 events of the CD45 + population.Contour plots were used to specifically define the T cell clusters in manual gates with FlowJo and exported as .fcsfiles.t-SNE dimensionality reduction and PhenoGraph clustering analyses were performed using the tool cytofkit run in R package software. 5Partial markers were used during the t-SNE 6 and PhenoGraph analyses. 7For the generation of heatmap displays, marker expression was normalized by dividing by the range of all markers (expression range from the 1 st to 99 th percentile).Data displays were generated using the ggplot2 R package. 8J Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s)

Safety study
For the evaluation of the neurovirulence of YST-OVH, twelve 3-to 4.5-year-old male or female rhesus macaques were obtained from the Beijing Prima Biotech (Beijing, China) before the beginning of the study.Animals were housed and maintained at the animal facility of Beijing JOINN Laboratories according to guidelines provided by the Institutional Animal Care and Use Committee for animal welfare.Two animals were randomly selected for inclusion in the control group, and the other ten animals received one left thalamic injection (i.c.) and one right i.c.injection of virus at a dose of 1.65×10 7 PFU/injection.The i.c.inoculations were performed with a stereotaxic instrument (Stoelting).Animals were monitored for 22 days following inoculation.
For the evaluation of the toxicity study of YST-OVH, twenty 3-to 5-year-old male or female cynomolgus monkeys were obtained from Guangxi Xiongsen Non-human Primates Laboratory (Guangxi, China) before the beginning of the study.Animals were housed and maintained at the animal facility of Beijing JOINN Laboratories.Animals were randomized into two groups (n=10 for each group), and each group included 5 male and 5 female cynomolgus monkeys.YST-OVH was administered by i.v.injection at a dose of 2.4×10 8 PFU in 3 ml of saline for three cycles of a total of 15 injections on days 1-5, days 13-17, and days 25-29 relative to the start date.Six monkeys from the YST-OVH-treated or untreated group were euthanized with pentobarbital (200 mg/kg, i.p.) on day 30, followed by histological examinations.The remaining animals were monitored for 58 days following inoculation.Animals were bled on days -3, 6, 25, 30 and 58 relative to the start date for a total of five blood draws.Clinical symptoms, body weight and body temperature were monitored on days -14, -3, 5, 12, 19, 26, 30, 33, 40, 47, 54, 57 and 58 relative to the start date.

Determination of serum biochemical parameters was performed on a TOSHIBA TBA-120FR
Chemistry Analyzer and EasyLyte ® electrolyte analyzer.Determination of hematological parameters were performed on a Siemens Advia 2120 Hematology Analyzer.Determination of cytokines was performed on a BD FACSCalibur using a CBA nonhuman primate Th1/Th2 cytokine kit (BD), according to the manufacturer's instructions.The reference values (95% CIs) BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s)    BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s)

A B
BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s)  BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s)

Fig. S5 .Fig. S6 .Fig. S7 .
Fig. S5.Assessment of hPD-1scFv expression in the serum (A) and tumors (B) of OV-treated immunocompetent mice at 24 h post-treatment under different dosing schemes.All values are presented as the mean ± SEM; unpaired two-tailed Student's t test (A, B).

Fig. S8 .
Fig. S8.OV treatment modulates Hepa1-6 tumor-infiltrating immune populations.(A) t-SNE plot of CD45 + Hepa1-6 tumor immune infiltrates overlaid with the expression of selected markers.(B) t-SNE plot derived from CyTOF analysis of tumor immune infiltrates obtained from each treatment group.Cells are colored by the clusters identified by PhenoGraph.(C) Density t-SNE plots of equal numbers of tumor immune infiltrates from each treatment group.(D) Heatmap displaying the normalized marker expression of each CD45 + cell cluster.(E) Quantitative analysis of the T cell clusters as a percentage of CD45 + cells.(F) Representative flow cytometry plots show characteristic markers of selected intratumoral T-cell clusters.HSV-1-specific H-2 Kb-HSV-1 gB dextramer, and SIINFEKL-specific H-2Kb-OVA Tetramer.(G) Quantification of antigen-specific and virus-specific memory CD8 + T cells in tumors.Quantitative data are presented as the mean ± SEM and were analyzed by an unpaired two-sided Student's t test.

Fig. S9 .
Fig. S9.Heatmap showing the expression of differentially expressed cytotoxicity genes (A) and cell cycle checkpoint genes (B).Gene expression was normalized to values obtained for the untreated control.
on this supplemental material which has been supplied by the author(s) placed J Immunother Cancer doi: 10.1136/jitc-2022-004762 :e004762.10 2024; J Immunother Cancer , et al.Ju F

Table S5 . Antibodies used for time-of-flight mass cytometry analysis.
BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s) BMJ Publishing Group Limited (BMJ) disclaims all liability and responsibility arising from any reliance Supplemental material placed on this supplemental material which has been supplied by the author(s)