An oncolytic vaccinia virus encoding CD47 nanobody potentiates antitumor immunity in multiple myeloma

TL;DR

A virus that targets cancer cells and boosts the immune system shows promise in treating multiple myeloma.

Contribution

A new oncolytic virus with an anti-CD47 nanobody is developed to enhance antitumor immunity and overcome drug resistance.

Findings

OVV-αCD47nb suppressed tumor growth and improved survival in mouse models of multiple myeloma.

The virus reprogrammed macrophages and increased CD8+ T cell activity in the tumor microenvironment.

OVV-αCD47nb synergized with bortezomib to overcome resistance and improve tumor control.

Abstract

Multiple myeloma (MM) is an incurable malignancy exhibiting immune evasion and resistance to proteasome inhibitors like bortezomib. We engineered an oncolytic vaccinia virus encoding an anti-mouse CD47 nanobody (OVV-αCD47nb) that combines direct oncolysis with localized CD47-SIRPα axis blockade. OVV-αCD47nb maintained infectivity and secreted anti-CD47 nanobodies that enhanced macrophage phagocytosis of tumor cells. In murine MM models, OVV-αCD47nb suppressed tumor growth, extended survival, and induced durable responses without hematologic toxicity. Mechanistically, OVV-αCD47nb remodeled the tumor microenvironment by polarizing macrophages to M1-like phenotypes and enhancing CD8+ T cell infiltration and function. Transcriptomics revealed enriched pro-inflammatory and phagocytic pathways with downregulated autophagy genes. OVV-αCD47nb synergized with bortezomib to overcome resistance and improve tumor control over monotherapies. This multifunctional viro-immunotherapy strategy, which integrates oncolysis, immune reprogramming, and chemosensitization, offers a promising therapeutic approach for CD47-expressing malignancies.

•An oncolytic virus encoding an anti-CD47 nanobody induces multiple myeloma tumor regression•The therapy reprograms the tumor microenvironment and enhances immune cell activity•This strategy overcomes bortezomib resistance by inhibiting autophagy•Localized viral delivery increases treatment efficacy and limits systemic toxicity

An oncolytic virus encoding an anti-CD47 nanobody induces multiple myeloma tumor regression

The therapy reprograms the tumor microenvironment and enhances immune cell activity

This strategy overcomes bortezomib resistance by inhibiting autophagy

Localized viral delivery increases treatment efficacy and limits systemic toxicity

Immunology; Medical biotechnology; Cancer

Figures

Construction and characterization of OVV-αCD47nb *in vitro*(A) Schematic representation of OVV-αCD47nb. The anti-CD47 VHH domain was inserted into the thymidine kinase (TK) gene of vaccinia virus under the control of the Pse/L promoter.(B) Oncolytic activity of OVV and OVV-αCD47nb against murine tumor cells. NS-1 and SP2/0 cells were infected with serial dilutions of OVV or OVV-αCD47nb. Cell viability was assessed using the CCK-8 assay at 72 h post-infection.(C) Infectivity analysis of OVV-αCD47nb. NS-1 and SP2/0 cells infected with OVV or OVV-αCD47nb at an MOI of 20 were harvested at indicate

Antitumor activity and systemic immune responses of OVV-αCD47nb *in vivo*(A) Experimental design. Balb/c mice (*n* = 5–6 per group) were subcutaneously injected with 5×10^6^ NS-1 cells in the right groin. When tumors reached 50–100 mm^3^, mice received intratumoral injections of PBS, OVV (2 × 10^7^ PFU), or OVV-αCD47nb (2 × 10^7^ PFU) every other day for three doses.(B) Tumor volume progression schematic. Mice were euthanized when tumor volume exceeded 2,000 mm^3^. (C) Survival analysis. Kaplan-Meier curves of mice treated with PBS, OVV, or OVV-αCD47nb.(D) Tumor growth curves for each mouse.(E

OVV-αCD47nb enhances immune infiltration and reprograms tumor-associated macrophages(A) BALB/c mice bearing subcutaneous NS-1 tumors received intratumoral injections of PBS, OVV, or OVV-αCD47nb (2 × 10^7^ PFU/mouse). Tumors were harvested within 24 h after the final injection for single-cell suspension preparation and flow cytometric analysis of CD8^+^ T cell proportions. Parallel experiments assessed TNF production, CD107a expression, and LAG-3 surface levels on tumor-infiltrating CD8^+^ T cells.(B) Representative diagram of flow cytometric analysis of CD8^+^ T cells.(C) Representative diagra

Mechanism of OVV-αCD47nb in remodeling the multiple myeloma microenvironment(A) Experimental workflow. BALB/c mice bearing subcutaneous NS-1 tumors were treated with PBS, OVV, or OVV-αCD47nb (2 × 10^7^ PFU/mouse). Two days post-treatment, tumors were excised and processed into single-cell suspensions. CD45^+^ cells were isolated via fluorescence-activated cell sorting (FACS), followed by total RNA extraction and purification for sequencing analysis.(B) KEGG pathway enrichment analysis (bubble plot) of differentially expressed genes.(C) Heatmap depicting M1-associated gene expression in macroph

Enhanced antitumor efficacy of OVV-αCD47nb combined with proteasome inhibition in NS-1 tumors(A) Treatment scheme of OVV-αCD47nb and proteasome inhibitor combination therapy in NS-1 subcutaneous tumor model. Balb/c mice (*n* = 4–5 per group) bearing subcutaneous NS-1 tumors (5×10^6^ cells) received PBS, OVV-αCD47nb monotherapy (1 × 10^7^ PFU/mouse), bortezomib monotherapy (1 mg/kg), or combination therapy for three times.(B) Schematic representation of tumor volume progression in mice. Once the tumor volume exceeded 2,000 mm3, the mouse was considered dead.(C) Tumor growth curves for each mous

Full text

Introduction

Multiple myeloma (MM) is a highly aggressive hematological malignancy that accounts for approximately 10% of all hematological tumors and is characterized by an abnormal proliferation of plasma cells in the bone marrow.1^,^2 Protease inhibitors, immunomodulators, and chemotherapy with dexamethasone induction are the mainstays of the current treatment for multiple myeloma.3 Immunotherapies such monoclonal antibodies (mAbs), bispecific antibody therapy, and CAR-T therapy have emerged in recent years, progressively overcoming the high toxicity of conventional chemotherapeutic medications. However, immune escape and disease recurrence remain problems.4^,^5^,^6 Therefore, investigating effective treatment targets is crucial.

CD47, also known as integrin-associated protein, is a ubiquitously expressed cell surface protein that is abnormally highly expressed in a wide range of solid tumors and hematologic malignancies, and its level is significantly associated with poor cancer prognosis.7^,^8^,^9 By attaching to SIRPα and sending the “don’t eat me” signal, CD47 mediates the immune escape mechanism in tumor cells and prevents macrophages from phagocytosing cancer cells.10^,^11 MM represents a particularly relevant model for CD47-targeted therapy, as bioinformatic analyses reveal significant CD47 upregulation in MM cells. This overexpression, coupled with evidence that CD47 blockade potently induces anti-MM phagocytosis, provides a strong rationale for targeting this pathway in MM.12^,^13 Research data show that there are more than 40 clinical trials targeting CD47,14 and in addition, several anti-CD47 mAbs have been used to treat a variety of hematologic tumors.15^,^16 Among them, Kim et al. showed that an anti-CD47 mAb (B6H12) significantly inhibited the growth of a myeloma xenograft model and led to tumor regression (CR = 15), suggesting a scavenging effect on myeloma-initiating cells.12 Magrolimab (hu5F9-G4), a humanized IgG4 mAby and the first anti-CD47 antibody,17 has demonstrated encouraging efficacy in patients with acute myeloid leukemia,18 DLBCL,19 and high-risk myelodysplastic syndromes.20 However, because CD47 is widely expressed on erythrocytes and platelets, systemic toxicity, including severe anemia and even death, is induced after targeted CD47 treatment, which severely affects the therapeutic utility of CD 47 mAb.17^,^21 Current mitigation strategies, including bispecific and safety-engineered antibodies, are constrained by their systemic administration,17^,^22^,^23^,^24^,^25 thereby driving the search for localized treatment strategies.

Oncolytic virus (OVs) therapy is a promising approach to cancer immunotherapy.26 Recently, the use of OVs in hematological malignancies and solid tumors has gained attention, with the advantage of selectively infecting and directly killing tumor cells and inducing immunogenic cell death to activate the body’s innate and adaptive immune responses. This immune-activating effect helps to transform immunologically “cold” tumors into “hot” tumors and enhance the immune response in the tumor microenvironment (TME).27^,^28^,^29 In addition, OVs have been genetically engineered to load therapeutic transgenes, and these transgenes can be expressed and secreted in infected cancer cells, thereby significantly enhancing their antitumor activity.30^,^31^,^32^,^33 One of the most widely studied platforms is the oncolytic vaccinia virus (OVV), which can be used as a delivery vector encoding several therapeutic genes.34^,^35^,^36 In previous research, we developed an OVV engineered to express a CD47 nanobody. This virus demonstrated significant synergistic antitumor activity by enhancing adaptive immune responses in coordination with innate immune mechanisms, as evidenced in preclinical models of solid tumors and B-cell lymphoma.37^,^38

In this work, we explored if arming OVV with αCD47nb could enhance the killing effect against MM cells. We assessed the effect of OVV-αCD47nb in preclinical tumor models and elucidated the underlying mechanism using RNA-sequencing (RNA-seq) and flow cytometry to evaluate functional and genetic changes of intratumoral immune cell populations of BALB/c mouse model following the treatment of OVV-αCD47nb. Furthermore, we demonstrated that OVV-αCD47nb could overcome chemotherapy resistance in MM, and investigated the therapeutic efficacy of combining bortezomib with OVV-αCD47nb in MM treatment.

Results

OVV-αCD47nb construction and characterization

We constructed a recombinant vaccinia virus (OVV-αCD47nb) by inserting an anti-mouse CD47 nanobody (αCD47nb) gene fragment into the thymidine kinase (TK)-destroyed vaccinia virus WR strain via homologous recombination. A control virus expressing GFP alone (OVV-GFP) was similarly constructed (Figure 1A). To assess potential impacts of αCD47nb gene insertion on viral cytotoxicity and infectivity, NS-1 and SP2/0 tumor cells were infected with GFP-expressing OVV or OVV-αCD47nb for 72 h. Both viruses exhibited comparable dose-dependent cytotoxicity across multiple MOIs (Figure 1B). Flow cytometry analysis further revealed equivalent GFP^+^ cell proportions over time, indicating unchanged infection efficiency between OVV and OVV-αCD47nb (Figure 1C). Immunoblot analysis of lysates and supernatants from OVV-αCD47nb-infected Vero cells confirmed intracellular and extracellular αCD47nb expression (Figure 1D), demonstrating functional nanobody production and secretion. These findings demonstrate that αCD47nb insertion preserves core viral functionality.Figure 1. Construction and characterization of OVV-αCD47nb in vitro(A) Schematic representation of OVV-αCD47nb. The anti-CD47 VHH domain was inserted into the thymidine kinase (TK) gene of vaccinia virus under the control of the Pse/L promoter.(B) Oncolytic activity of OVV and OVV-αCD47nb against murine tumor cells. NS-1 and SP2/0 cells were infected with serial dilutions of OVV or OVV-αCD47nb. Cell viability was assessed using the CCK-8 assay at 72 h post-infection.(C) Infectivity analysis of OVV-αCD47nb. NS-1 and SP2/0 cells infected with OVV or OVV-αCD47nb at an MOI of 20 were harvested at indicated time points and analyzed by flow cytometry.(D) Expression and secretion of anti-CD47 nanobody protein were detected by western blot. Vero cells were infected with OVV or OVV-αCD47nb at MOI 20. At 48 h post-infection, cell lysates and supernatants were collected and detected by anti-Flag antibody.(E) Binding of αCD47nb. NS-1 and SP2/0 cells were incubated with OVV and OVV-αCD47nb supernatants, and αCD47nb binding was detected by flow cytometry.(F) Representative flow cytometry plots of αCD47nb binding.(G) Phagocytosis of NS-1 cells by BMDMs. Phagocytosis was quantified as the percentage of CMFDA^+^ BMDMs that had ingested Deep Red^+^ tumor cells.(H) Representative flow cytometry plots of NS-1 phagocytosis by BMDMs.(I) CD47 blocking by OVV-αCD47nb. Blockade of CD47 on NS-1 and SP2/0 cells was assessed via flow cytometry.(J) Representative flow cytometry plots of CD47 blocking.(K) Representative fluorescent images of NS-1 phagocytosis by BMDMs. Scale bars represent 125 μm. Arrows highlight BMDMs that have phagocytosed tumor cells. Scale bars represent 30 μm. Data represent the mean ± standard deviation (SD) of ≥three independent experiments. Statistical significance was determined by two-way ANOVA with Tukey’s multiple-comparisons post-test. ns, not significant; ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001. OVV, oncolytic vaccinia virus; αCD47nb, anti-CD47 nanobody; gpt, guanine-hypoxanthine phosphoribosyl transferase; EGFP, enhanced green fluorescent protein; P7.5, vaccinia virus P7.5 early/late promoter; Pse/L, synthesized vaccinia virus early/late promoter; MOI, multiplicity of infection; BMDMs, bone marrow-derived macrophages.

CD47, which is widely expressed in human solid tumors, acts as a “don’t eat me” signal that critically regulates phagocytosis resistance.10^,^11 The bioinformatics analysis suggested a significant upregulation of CD47 expression in MM cells13 (Figure S1A), which was subsequently confirmed by flow cytometry to be highly expressed on the surface of mouse NS-1 and SP2/0 cells (Figures S1B and S1C). Supernatant from OVV-αCD47nb-infected cells effectively bound surface CD47 on both cell lines, as detected by PE-conjugated anti-FLAG antibody (Figures 1E and 1F). Subsequent CD47 blockade assessment revealed that OVV-αCD47nb significantly inhibited CD47 antibody binding compared to controls (Figures 1I and 1J).

To evaluate αCD47nb’s regulatory effect on macrophage phagocytosis, in vitro phagocytosis assays were performed. CMFDA-labeled bone marrow-derived macrophages (BMDMs) and deep red-labeled tumor cells (NS-1 or SP2/0) were co-cultured (effector:target ratio 1:2) for 5 h in the presence of supernatant from OVV or OVV-αCD47nb-infected cells. Flow cytometry analysis revealed that supernatant from OVV-αCD47nb-infected cells significantly enhanced phagocytosis of NS-1 cells by BMDMs compared to controls (Figures 1G and 1H). A comparable phagocytosis enhancement was observed with SP2/0 cells (Figures S1D and S1E). These findings were corroborated by fluorescence microscopy (Figures 1K and S1F). These data demonstrate that OVV-αCD47nb produces and secretes a functional αCD47nb capable of binding and blocking tumor cell CD47, thereby potently augmenting macrophage-mediated phagocytosis in vitro.

OVV-αCD47nb significantly enhances anti-tumor activity and activates systemic immune responses

Building on evidence that OVV-αCD47nb effectively enhances macrophage phagocytosis in vitro, we further evaluated its antitumor efficacy in vivo. To this end, we established subcutaneous NS-1 tumor models in mice (Figure 2A). When tumors reached an average volume of 50–100 mm^3^, tumor-bearing mice were randomized into three groups receiving: phosphate-buffered saline (PBS), OVV, or OVV-αCD47nb (2 × 10^7^ PFU/mouse). Results demonstrated that OVV-αCD47nb treatment significantly suppressed tumor growth (Figure 2B), prolonged survival (Figure 2C), and achieved a 16.7% complete response rate (1/6; Figure 2D) compared to controls. No significant body weight changes (Figure 2E) or hematologic toxicity (red blood cells, hemoglobin, and platelets) (Figures S2A–S2F) were observed during treatment, indicating favorable tolerability.Figure 2. Antitumor activity and systemic immune responses of OVV-αCD47nb in vivo(A) Experimental design. Balb/c mice (n = 5–6 per group) were subcutaneously injected with 5×10^6^ NS-1 cells in the right groin. When tumors reached 50–100 mm^3^, mice received intratumoral injections of PBS, OVV (2 × 10^7^ PFU), or OVV-αCD47nb (2 × 10^7^ PFU) every other day for three doses.(B) Tumor volume progression schematic. Mice were euthanized when tumor volume exceeded 2,000 mm^3^. (C) Survival analysis. Kaplan-Meier curves of mice treated with PBS, OVV, or OVV-αCD47nb.(D) Tumor growth curves for each mouse.(E) Body weight monitoring.(F) Representative flow cytometry plots of splenic immune profiling (NS-1 model). Flow cytometry analysis of CD8^+^ T cells and activation markers in splenocytes from treated tumor-bearing mice.(G) Representative flow cytometry plots of splenic CD8^+^ T cells and activation markers (NS-1 model).(H) Representative flow cytometry plots of splenic CD8^+^ T cells and activation markers (SP2/0 model). Data represent the mean ± standard deviation (SD) of ≥three independent experiments. Statistical significance was determined by two-way ANOVA with Tukey’s multiple-comparisons post-test. ns, not significant; ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001. S.C., subcutaneous; I.T., intratumoral; CR, complete response; TNF-α, tumor necrosis factor-alpha.

Quantitative PCR analysis revealed that OVV-αCD47nb replicated preferentially in tumors, with minimal viral levels in normal tissues (Figure S2G). Expression of the secreted αCD47nb was likewise concentrated within tumor sites (Figure S2H), indicating localized activity and a favorable safety profile. Spleens collected from the aforementioned tumor models were processed into single-cell suspensions. Subsequent flow cytometry analysis revealed that OVV-αCD47nb significantly increased splenic CD8^+^ T cell infiltration, elevated proportions of granzyme B^+^/TNF-α^+^ effector T cells, and upregulated the early activation marker CD69 (Figures 2F and 2G). These changes indicate enhanced early T cell activation and cytotoxic effector function. A comparable increase in CD8^+^ T cell infiltration, along with elevated expression of the effector molecules granzyme B, TNF-α, and perforin, was observed in SP2/0 myeloma models (Figure 2H). Collectively, these data demonstrate that OVV-αCD47nb not only suppresses local tumor growth but also systemically recruits and enhances the cytotoxic function of CD8^+^ T cells, conferring durable anti-myeloma immunity.

OVV-αCD47nb enhances immune infiltration in tumor microenvironment while reprogramming tumor-associated macrophages

To elucidate the mechanism by which OVV-αCD47nb enhances antitumor activity, we established NS-1 tumor models and administered intratumoral injections of PBS, OVV, or OVV-αCD47nb every 48 h for three consecutive treatments. Tumors were harvested within 24 h after the final injection for single-cell suspension preparation and flow cytometric analysis of immune cell populations. Compared to PBS or OVV controls, OVV-αCD47nb significantly increased CD8^+^ T cell infiltration in the TME (Figures 3A and 3B). Further analysis revealed enhanced effector function in CD8^+^ T cells, characterized by elevated expression of the early activation marker CD69, the degranulation marker CD107a, and the cytotoxic molecules granzyme B and TNF-α (Figures 3C–3F). This was accompanied by reduced expression of the exhaustion-associated molecules LAG-3 and CTLA-4 (Figures 3G and 3H). Together, these changes in the expression profiles of effector and inhibitory receptors are consistent with an overall improved functional state of tumor-infiltrating CD8^+^ T cells. Enhanced CD8^+^ T cell effector responses and reduced immunosuppressive lymphocyte infiltration were similarly observed in SP2/0 myeloma models (Figure S3).Figure 3OVV-αCD47nb enhances immune infiltration and reprograms tumor-associated macrophages(A) BALB/c mice bearing subcutaneous NS-1 tumors received intratumoral injections of PBS, OVV, or OVV-αCD47nb (2 × 10^7^ PFU/mouse). Tumors were harvested within 24 h after the final injection for single-cell suspension preparation and flow cytometric analysis of CD8^+^ T cell proportions. Parallel experiments assessed TNF production, CD107a expression, and LAG-3 surface levels on tumor-infiltrating CD8^+^ T cells.(B) Representative diagram of flow cytometric analysis of CD8^+^ T cells.(C) Representative diagram of flow cytometric analysis of the expression of CD69 on CD8^+^ T cells.(D) Representative diagram of flow cytometric analysis of the expression of CD107a on CD8^+^ T cells.(E) Representative diagram of flow cytometric analysis of secreting granzyme B in CD8^+^ T in tumors.(F) Representative diagram of flow cytometric analysis of producing TNF-α in CD8^+^ T in tumors.(G) Representative diagram of flow cytometric analysis of CD8^+^ T cell surface depletion indices LAG-3 in tumors.(H) Representative diagram of flow cytometric analysis of CD8^+^ T cell surface depletion indices CTLA-4 in tumors.(I) Flow cytometric analysis of the proportion of F4/80^+^ CD11b^+^ macrophages.(J) Flow cytometric analysis of the proportion of CD86^+^ M1 macrophages.(K) Flow cytometric analysis of the proportion of CD206^+^ M2 macrophages.(L) Detection of CD8^+^ T and CD68^+^ macrophages infiltration in NS-1 tumors by immunohistochemistry. Scale bars represent 100 μm. Data represent the mean ± standard deviation (SD) of ≥three independent experiments. Statistical significance was determined by one-way ANOVA with Tukey’s multiple-comparisons post-test. ns, not significant; ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001. LAG-3, lymphocyte-activation gene 3; CTLA-4, cytotoxic T-lymphocyte-associated protein 4; TAMs, tumor-associated macrophages.

Immunohistochemical (IHC) staining confirmed significantly increased infiltration of CD8^+^ T cells and CD68^+^ macrophages in OVV-αCD47nb-treated tumors (Figure 3L). This enhanced immune cell infiltration may be attributed to OVV-αCD47nb’s dual mechanisms: viral oncolysis releases tumor antigens and DAMPs to recruit macrophages, while CD47 blockade promotes phagocytosis and creates a pro-inflammatory milieu that facilitates T cell recruitment and effector differentiation. Collectively, these results demonstrate that OVV-αCD47nb effectively recruits and potentiates the anti-tumor activity of CD8^+^ T cells while reprogramming the immunosuppressive TME.

Given the critical role of the CD47/SIRPα axis in macrophage regulation, we assessed the impact of OVV-αCD47nb on tumor-associated macrophages (TAMs). Flow cytometry revealed that OVV-αCD47nb treatment significantly increased total macrophage infiltration within tumors (Figure 3I). Notably, it markedly upregulated expression of the costimulatory molecule CD86 (associated with M1 polarization) on macrophages (Figure 3J), while downregulating the protumorigenic M2 marker CD206 (Figure 3K). These flow cytometry data provide direct evidence that OVV-αCD47nb reprograms TAMs in the TME from a pro-tumorigenic “M2-like” phenotype toward an antitumor “M1-like” phenotype.

Mechanism of multiple myeloma microenvironment remodeling by OVV-αCD47nb

To gain further insight into the broader transcriptional changes underlying this immune remodeling, we performed transcriptomic analysis on CD45^+^ cells isolated from tumors of NS-1 myeloma-bearing mice treated with PBS, OVV, or OVV-αCD47nb (Figure 4A). KEGG enrichment analysis revealed significant activation of immune response pathways (cytokine-cytokine receptor interaction, TNF/NF-κB signaling) and TME remodeling pathways (ECM-receptor interaction, cell adhesion molecules [CAMs]) in the OVV-αCD47nb group compared to PBS and OVV controls (Figure 4B). Crucially, enrichment of phagosome and NOD-like receptor pathways confirmed enhanced tumor cell phagocytosis and oncolytic virus-induced immune activation (Figure 4B).Figure 4. Mechanism of OVV-αCD47nb in remodeling the multiple myeloma microenvironment(A) Experimental workflow. BALB/c mice bearing subcutaneous NS-1 tumors were treated with PBS, OVV, or OVV-αCD47nb (2 × 10^7^ PFU/mouse). Two days post-treatment, tumors were excised and processed into single-cell suspensions. CD45^+^ cells were isolated via fluorescence-activated cell sorting (FACS), followed by total RNA extraction and purification for sequencing analysis.(B) KEGG pathway enrichment analysis (bubble plot) of differentially expressed genes.(C) Heatmap depicting M1-associated gene expression in macrophages following treatment with OVV-αCD47nb versus OVV.(D) Heatmap depicting M2-associated gene expression in macrophages following treatment with OVV-αCD47nb versus OVV. Data represent the mean ± standard deviation (SD) of ≥three independent experiments. S.C., subcutaneous; I.T., intratumoral.

Strikingly, differential expression analysis revealed a transcriptional profile consistent with the M1-like polarization observed by flow cytometry, characterized by upregulation of M1-associated genes (Fcgr3, Il6, Cd80, and Cd86) (Figure 4C) and downregulation of M2-associated genes (Arg1, Cd209a, and Ly6c1/2) (Figure 4D). Thus, the transcriptomic data corroborate the flow cytometry findings and together demonstrate that OVV-αCD47nb effectively reprograms the immunosuppressive TME.

OVV-αCD47nb combined with proteasome inhibitor therapy

Bortezomib, a first-line chemotherapeutic agent for multiple myeloma, exhibits limited clinical utility due to acquired resistance.39^,^40 Studies demonstrate that autophagy activation serves as a key mechanism mediating bortezomib resistance.41^,^42^,^43^,^44 Transcriptomic analysis (Figure S4A) revealed that OVV-αCD47nb treatment significantly suppresses autophagic activity compared to PBS controls, evidenced by downregulation of pro-autophagic genes (BECN1, ATG5, MAP1LC3, and HMGB1) and upregulation of autophagy suppressors (mTOR, TRIM24, and SQSTM1/p62). Western blot analysis confirmed these findings, demonstrating that OVV-αCD47nb monotherapy significantly suppressed autophagic activity, as indicated by reduced LC3-II levels and SQSTM1/p62 accumulation. Notably, the combination of OVV-αCD47nb with bortezomib resulted in the most potent inhibition of autophagy (Figure S4B). Further investigation is warranted to elucidate the relative contributions of this direct pathway versus immunomodulatory effects in mediating the overall synergy.

To evaluate the antitumor efficacy of OVV-αCD47nb combined with bortezomib, we established NS-1 tumor-bearing mice following the experimental design in Figure 5A. Mice with tumor volumes of 50–100 mm^3^ were randomized into four treatment groups: PBS, OVV-αCD47nb monotherapy (1 × 10^7^ PFU/mouse), bortezomib monotherapy (1 mg/kg), or combination therapy. Tumor growth monitoring demonstrated that combination treatment resulted in the slowest tumor volume increase relative to PBS or monotherapy groups (Figure 5B). Individual growth curves confirmed significant tumor progression delay with combination therapy (Figure 5C). Furthermore, combination-treated mice exhibited significantly prolonged survival (Figure 5D) with stable body weights across all groups, indicating favorable treatment tolerability (Figure 5E). Collectively, these results demonstrate synergistic antitumor effects between OVV-αCD47nb and bortezomib, significantly inhibiting multiple myeloma growth in vivo.Figure 5. Enhanced antitumor efficacy of OVV-αCD47nb combined with proteasome inhibition in NS-1 tumors(A) Treatment scheme of OVV-αCD47nb and proteasome inhibitor combination therapy in NS-1 subcutaneous tumor model. Balb/c mice (n = 4–5 per group) bearing subcutaneous NS-1 tumors (5×10^6^ cells) received PBS, OVV-αCD47nb monotherapy (1 × 10^7^ PFU/mouse), bortezomib monotherapy (1 mg/kg), or combination therapy for three times.(B) Schematic representation of tumor volume progression in mice. Once the tumor volume exceeded 2,000 mm3, the mouse was considered dead.(C) Tumor growth curves for each mouse.(D) Mouse survival curves.(E) Mouse body weight change curve. Data represent the mean ± standard deviation (SD) of ≥three independent experiments. Statistical significance was determined by two-way ANOVA with Tukey’s multiple-comparisons post-test. ns, not significant; ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001. S.C., subcutaneous; I.T., intratumoral; I.P., intraperitoneal.

Discussion

Both oncolytic virotherapy and anti-CD47 immunotherapy represent promising therapeutic modalities for advanced malignancies. However, despite extensive preclinical investigation of these approaches, clinical translation has revealed limited efficacy, with only a subset of patients deriving clinical benefit. This therapeutic resistance stems from multiple immune-evasion mechanisms inherent to the complex and heterogeneous TME.45 Recent clinical studies have demonstrated superior antitumor activity with combination regimens pairing oncolytic viruses and PD-1 checkpoint inhibitors compared to either monotherapy.46^,^47^,^48 Preclinical models have further elucidated a combinatorial strategy where OV platforms (including engineered herpesviruses and myxomaviruses) are genetically modified to express antibodies targeting CD47.49^,^50^,^51 This approach enables dual mechanistic action: (1) direct oncolysis through viral replication in malignant cells, and (2) localized secretion of anti-CD47 antibodies that enhance innate immune recognition while remodeling the immunosuppressive TME. This engineered strategy synergizes innate immune activation with adaptive antitumor immunity, offering a multifaceted therapeutic approach that addresses both viral oncolysis and TME-mediated immunosuppression. These findings underscore the potential of rationally designed combination therapies to overcome treatment resistance in advanced malignancies.

Building on our previous development of an OVV expressing an anti-CD47 nanobody (OVV-αCD47nb), which demonstrated efficacy in solid tumors and lymphoma,37^,^38 the present study extends this platform to multiple myeloma, a hematologic malignancy characterized by a distinct and therapy-resistant microenvironment. The use of a nanobody offers advantages over full-length antibodies, including improved tumor penetration and more efficient viral packaging, while preserving effective CD47 blockade. Although the integration of viral oncolysis with CD47-SIRPα inhibition remains the core mechanism, this work reveals a significant translationally relevant advance. OVV-αCD47nb effectively alleviates immunosuppressive features of the myeloma microenvironment and exhibits strong synergistic antitumor activity with bortezomib. This synergy is mediated, at least in part, by suppression of autophagy-related pathways, providing a combinatorial strategy to overcome acquired chemotherapy resistance in multiple myeloma.

CD47-targeting antibodies have shown clinical potential but are hampered by systemic toxicity due to ubiquitous CD47 expression on erythrocytes and platelets.17^,^21 While our qPCR-based biodistribution analysis confirms tumor-enriched viral replication and transgene expression, we did not perform pharmacokinetic profiling to quantify serum nanobody levels. Low-level systemic exposure remains possible, and any circulating nanobody may be rapidly sequestered by CD47-expressing cells, making it difficult to detect in serum. Nevertheless, our strategy is designed to restrict nanobody secretion to sites of active viral replication, thereby minimizing systemic exposure while enhancing local immune activation. Future translational studies should include serum pharmacokinetics and hematologic evaluations to fully define the safety profile of this approach.

Consistent with prior evidence that oncolytic viruses convert “cold” tumors into “hot” tumors, OVV-αCD47nb remodeled the MM microenvironment, promoting an enhanced effector profile in tumor-infiltrating CD8^+^ T cells, characterized by elevated expression of granzyme B, CD107a, and TNF-α, coupled with reduced expression of the inhibitory receptors LAG-3 and CTLA-4, changes consistent with an improved functional state. As vaccinia virus is inherently immunogenic, it is acknowledged that some of the observed T cell activation may reflect antiviral immunity. Future studies employing tumor-antigen-specific assays, such as tetramer staining or TCR repertoire analysis, will be valuable to delineate the relative contributions of tumor-specific versus virus-specific CD8^+^ T cell responses to the overall antitumor efficacy. Transcriptomic profiling further confirmed activation of pro-inflammatory signaling pathways and suppression of autophagy-associated programs, the latter being a well-recognized mediator of bortezomib resistance.41^,^42^,^43^,^44

The combination of OVV-αCD47nb with bortezomib produced superior tumor control and survival compared with either monotherapy. Mechanistically, the virus reduced autophagic gene expression and sensitized tumor cells to proteasome inhibition, highlighting a rational approach to overcome chemotherapy resistance in MM. These results emphasize the therapeutic value of integrating viro-immunotherapy with established cytotoxic regimens.

Our findings highlight several important considerations for future research. By providing spatially confined delivery of immune-modulatory nanobodies, OVV-αCD47nb offers a preclinical proof-of-concept to minimize systemic adverse effects while achieving durable immune activation. However, a key limitation of the current study is the use of a mouse-specific nanobody, which precludes direct testing in human MM cells or patient samples. Therefore, the development of a human-specific counterpart (OVV-αhCD47nb) and its evaluation in humanized models represent critical next steps toward any potential clinical translation. Moreover, while this study focused on synergy with bortezomib, the immune-activating properties of OVV-αCD47nb provide a rationale for future exploration of combinations with other immunotherapies, such as immunomodulatory drugs or mAbies. Finally, to address disseminated disease, future work should investigate intravenous delivery in models of systemic myeloma.

In conclusion, OVV-αCD47nb represents a multifunctional platform capable of simultaneously inducing direct tumor lysis, reprogramming the TME, and overcoming drug resistance in preclinical models. This study provides mechanistic insights and a proof-of-concept in murine systems. Further validation in human-relevant models, along with exploration of additional drug combinations and delivery routes, is warranted to assess the potential of this strategy for MM and other CD47-expressing malignancies.

Limitations of the study

This study has several limitations. The nanobody used is mouse-specific, preventing direct evaluation in human systems, and its pharmacokinetics and potential systemic exposure were not quantified. Furthermore, the distinct contributions of antiviral versus antitumor T cell responses remain unclear. In addition, the findings are derived from a subcutaneous tumor model and a single drug combination, warranting further validation in more diverse models and therapeutic regimens. Finally, this study did not assess sex-based differences in antitumor responses due to the exclusive use of female mice.

Resource availability

Lead contact

Requests for further information and resources should be directed to and will be fulfilled by the lead contact, Shibing Wang (wangshibing@hmc.edu.cn).

Materials availability

Recombinant oncolytic vaccinia virus generated in this study are available from the lead contact upon request.

Data and code availability

The RNA sequencing data generated in this study have been deposited in the NCBI Sequence Read Archive (SRA) under accession code SRP629910. All other data reported in this article will be shared by the lead contact upon request. This article does not report original code. Any additional information required to reanalyze the data reported in this work study is available from the lead contact upon request.

Acknowledgments

We thank Figdraw (www.figdraw.com) for graphical abstract creation and the Experimental Animal Center of Zhejiang University of Technology for providing the animal facilities and technical support. This work was supported by the Key R&D Program of 10.13039/501100003786Hangzhou Science and Technology Bureau (2025SZD1B13), 10.13039/501100001809National Natural Science Foundation of China (82470182), Foundation of Science Technology Department of Zhejiang Province (LKLY25H160006), Construction Fund of Key Medical Disciplines of Hangzhou-Laboratory Diagnostics (2025HZZD01), Natural Science Foundation of Hangzhou (2024SZRYBC080002), Zhejiang Provincial Traditional Chinese Medicine Science and Technology Plan Project (2023ZL733), and Shaoxing Health Science and Technology Project (2022KY080).

Author contributions

Conceptualization, S.W., X.T., Y.X., and L.P.; methodology, L.P., X.Z., W.S., S.W., and X.T.; investigation, L.P., X.Z., J.Z., Q.Z., and H.W.; visualization, L.P. and X.Z.; funding acquisition, S.W., X.T., and Y.X.; project administration, S.W. and X.T.; supervision, S.W. and X.T.; writing – original draft, L.P. and X.Z.; writing – review and editing, S.W., X.T., and L.P.

Declaration of interests

The authors declare no competing interests.

STAR★Methods

Key resources table

REAGENT or RESOURCESOURCEIDENTIFIERAntibodiesAnti-FLAG antibodySigma-AldrichCat#F1804HRP-conjugated goat anti-mouse IgG (H + L)BeyotimeCat#A0216PE anti-DYKDDDDK Tag antibodyBioLegendCat#637309PE anti-mouse CD47 antibodyBioLegendCat#127507Live/Dead Fixable Violet Dead Cell StainInvitrogenCat# L34955Brilliant Violet 510 anti-mouse CD45 antibodyBioLegendCat# 157219PE anti-mouse CD45 antibodyBioLegendCat# 103106APC anti-mouse CD3 antibodyBioLegendCat# 100236FITC anti-mouse CD4 antibodyBioLegendCat# 100406FITC anti-mouse CD8 antibodyBioLegendCat# 100706APC anti-mouse CD69 antibodyBioLegendCat# 104514APC/Cy7 anti-mouse CD107a antibodyBioLegendCat# 121616APC anti-mouse CD11b antibodyBioLegendCat# 101212PE/Cy7 anti-mouse F4/80 antibodyBioLegendCat# 157307PerCP anti-mouse CD86 antibodyBioLegendCat# 105026FITC anti-mouse CD206 antibodyBioLegendCat# 141704Brilliant Violet 650 anti-mouse LAG-3 antibodyBioLegendCat# 125227Brilliant Violet 711 anti-mouse Tim-3 antibodyBioLegendCat# 119727PerCP/Cy5.5 anti-mouse CD152 antibodyBioLegendCat# 106315PerCP/Cy5.5 anti-mouse TNF-α antibodyBioLegendCat# 506322Alexa Fluor 700 anti-mouse Granzyme B antibodyBioLegendCat# 372222PE anti-mouse Perforin antibodyBioLegendCat# 154306Anti-rabbit CD8 antibodyAbcamCat# ab217344Anti-rabbit CD68 antibodyAbcamCat# ab283654HRP-conjugated goat anti-rabbit IgG secondary antibodyServicebioCat# G1213Anti-SQSTM1/p62 antibodyAbcamCat# ab211324LC3B Rabbit mAbAbcamCat# A19665Chemicals, peptides, and recombinant proteinsLipofectamine transfection reagentThermoFisherCat# A12621XanthineSangonCat# A601197Mycophenolic acidSangonCat# A600640HypoxanthineSangonCat# A500336BCA assay kitThermoCat# 23227Enhanced chemiluminescence kitFDbioCat# FD8020CCK-8 solutionBeyotimeCat# C0037Recombinant mouse M-CSFAmizona ScientificCat# AM10003-010CellTracker™ Deep Red DyeThermoFisherCat# A66433CellTracker™ Green CMFDA DyeThermoFisherCat# A66434BortezomibMedChemExpressCat# HY-10227Collagenase IISolarbioCat# C8150BD Cytofix/Cytoperm KitBD BiosciencesCat# 5547143,3′-diaminobenzidine (DAB)ServicebioCat# G1212-2HematoxylinServicebioCat# G1004TRIzolAmbionCat# 15596026Viral DNA/RNA KitCowin BiotechCat# CW0548SFastPure Cell/Tissue Total RNA Isolation KitVazyme BiotechCat# RC101–01HiScript II Reverse TranscriptaseVazyme BiotechCat# R201–01Deposited dataRNA sequencingNCBI databaseSRP629910Experimental models: Cell linesP3/NSI/1-Ag4-1 (NS-1)ATCCCat# TIB-18Sp2/0-Ag14 (SP2/0)ATCCCat# CRL-1581HEK293ATCCCat# CRL-1573HeLa-S3ATCCCat# CCL-2.2VeroATCCCat# CCL-81Experimental models: Organisms/strainsMouse: BALB/cthe Model Animal Research Center of Nanjing UniversityN/AWestern reserve strain of VV (WR-VV)ATCCCat# VR-1354Recombinant DNApVV-αCD47nbThis paperN/AαCD47 nanobody gene fragmentGenScriptN/ASoftware and algorithmsNovoCyte QuanteonAgilentN/AInvitrogen EVOS M7000GermanyN/AMicroscopeNikon Eclipse 100N/AOmicStudio platformN/AN/AGraphPad PrismN/AN/AAdobe IllustratorAdobeN/AFigdrawN/AN/A

Experimental model and study participant details

Mice

Six-to eight-week-old female BALB/c mice were used for the experiments. All experimental procedures were conducted in accordance with the Guide for the Care and Use of Laboratory Animals in the Zhejiang University of Technology, and conformed to the National Institutes of Health Guide for Care and Use of Laboratory Animals (Publication No. 85-23, revised 1996).

Cell lines

The study employed the following cell lines: P3/NSI/1-Ag4-1 (NS-1; TIB-18), Sp2/0-Ag14 (SP2/0; CRL-1581), HEK293 (CRL-1573), HeLa-S3 (CCL-2.2) and Vero (CCL-81). All cell lines were obtained from ATCC, authenticated by short tandem repeat (STR) analysis, and confirmed to be mycoplasma-free throughout the study. SP2/0 cells were maintained in RPMI-1640 medium (Cat#11875093, Gibco) supplemented with 10% fetal bovine serum (FBS; Cat#16000044, Gibco). NS-1, HEK293, and Vero cells were cultured in DMEM (Cat#11965092, ThermoFisher) containing 10% FBS. FBS. HeLa-S3 cells were grown in serum-free suspension medium (H740KJ, Basalmedia). using spinner flasks. All cells were incubated at 37°C with 5% CO_2_.

Method details

Construction of recombinant oncolytic vaccinia viruses

The αCD47 nanobody gene fragment was synthesized by GenScript (Nanjing, China) and cloned into the pVV-control shuttle plasmid to generate pVV-αCD47nb. This recombinant plasmid features: (1) a synthetic early/late promoter (pSE/L) controlling αCD47nb expression, and (2) a VV p7.5K early/late promoter regulating enhanced green fluorescent protein (EGFP) and guanine-hypoxanthine phosphoribosyltransferase (GPT) expression. OVV-αCD47nb was constructed by homologous recombination using the shuttle plasmid pVV-αCD47nb and a western reserve strain of VV (WR-VV; VR-1354; ATCC). HEK293 cells were transfected with pVV-αCD47nb using the lipofectamine transfection reagent (Cat#A12621, ThermoFisher) after being infected with WR-VV for 2 h at a MOI of 1. After 48 h, the EGFP-positive plaques were chosen and sown in plates with Hela-S3 cells. The growth of WR-VV was inhibited by a conditional DMEM medium containing 250 g/mL xanthine (Cat#A601197, Sangon) 25 g/mL mycophenolic acid (Cat#A600640, Sangon) and 15 g/mL hypoxanthine (Cat#A500336, Sangon). It was confirmed by PCR and DNA sequencing that the recombinant virus was free of WR-VV contamination following numerous picking and seeding cycles. After that, the isolated virus was gradually grown in 6-well plates, cell culture dishes, and cell culture spinner flasks using Hela-S3 cells. The viral titer was established using the TCID50 method. The calculation formula is virus titer = 0.7 × 10×10^(1+S) (D−0.5)^, where S is log_10_ (dilution) and D is the sum of the EGFP positive ratios in each dilution.

Western blot

Vero cells were infected with OVV-αCD47nb or OVV at an MOI of 20. After 48 h, cell lysates and supernatants were collected. Protein concentrations were quantified using a BCA assay kit (Cat#23227, Thermo). Samples were separated on 10–15% gradient SDS-PAGE gels and transferred to PVDF membranes. Membranes were blocked with 5% non-fat milk, then incubated overnight at 4°C with mouse anti-FLAG antibody (Cat#F1804, Sigma-Aldrich). After washing, membranes were probed with HRP-conjugated goat anti-mouse IgG (H + L) (Cat#A0216, Beyotime) for 1 h at room temperature. Protein bands were visualized using an enhanced chemiluminescence kit (Cat# FD8020, FDbio).

For autophagy analysis, NS-1 cells were pretreated with rapamycin (100 nM) for 24 h, followed by respective treatments with PBS, OVV, OVV-αCD47nb, bortezomib (10 nM), or OVV-αCD47nb combined with bortezomib for 48 h. Cells were lysed and proteins were separated as described above. Membranes were incubated with primary antibodies against LC3B (Cat# A19665, Abcam) and SQSTM1/p62 (Cat# ab211324, Abcam), followed by appropriate HRP-conjugated secondary antibodies.

CCK8 assay

NS-1 and SP2/0 cells were seeded in 96-well plates at 8 × 10^3^ cells/well. Cells were infected in triplicate with OVV or OVV-αCD47nb at MOIs of 0.1, 1, 10, and 100. After 72 h incubation, 20 μL CCK-8 solution (Cat# C0037, Beyotime) was added to each well. Plates were incubated for 4 h at 37°C, and optical density was measured at 450 nm using a microplate reader. Cell viability was calculated as: (A treatment − A blank)/(A control − A blank) × 100% is the cell viability (%).

Viral infectivity assay

NS-1 and SP2/0 cells were infected in parallel with OVV or OVV-αCD47nb (MOI = 20) in 24-well plates. At 24, 48, 72and 96 h post-infection, cells were harvested and centrifuged at 300×g for 3 min. Infection rates were quantified by flow cytometry (NovoCyte Quanteon, Agilent) based on GFP fluorescence in the FITC channel.

Blocking and binding assays

NS-1 and SP2/0 cells were incubated for 24 h with supernatants from cells infected with either OVV or OVV-αCD47nb. αCD47nb binding and CD47 blocking were assessed using PE anti-DYKDDDDK Tag antibody (Cat#637309, BioLegend) and PE anti-mouse CD47 antibody (Cat#127507, BioLegend) followed by flow cytometry (NovoCyte Quanteon, Agilent).

Phagocytosis assay

Bone marrow-derived macrophages (BMDMs) were differentiated from tibial and femoral bone marrow cells of BALB/c mice. Cells were cultured for 6–7 days in DMEM supplemented with 10% FBS and 20 ng/mL recombinant mouse M-CSF (Cat#AM10003-010, Amizona Scientific), with BMDM identity confirmed by CD11b (Cat#101211, BioLegend) and F4/80 (Cat#157307, BioLegend) double-positive staining. For phagocytosis assessment, tumor cells were labeled with CellTracker Deep Red Dye (Cat#A66433, ThermoFisher) for 30 min at 37°C, while BMDMs were stained with CellTracker Green CMFDA Dye (Cat#A66434, ThermoFisher). BMDMs and tumor cells were co-cultured at a 1:2 ratio for 5 h at 37°C with 5% CO_2_ in the presence of supernatants from OVV or OVV-αCD47nb-infected cells. Phagocytosis was quantified via fluorescence microscopy (Invitrogen EVOS M7000, Germany) and flow cytometry (NovoCyte Quanteon, Agilent) by measuring the percentage of CMFDA^+^ macrophages containing Deep Red^+^ tumor cells.

Animal models

All animal procedures and experiments were performed following the guidelines that had been approved by the Animal Care and Use Committee of Zhejiang University of Technology (ZH20250423018). BALB/c mice were purchased from the Model Animal Research Center of Nanjing University (Nanjing, China). To establish a subcutaneous tumor model, NS-1 cells (5 × 10^6^cells in 100 μL) were injected subcutaneously into 6-week-old BALB/c mice. When tumors reached 50–100 mm^3^, mice were randomized for intratumoral treatment with OVV-αCD47nb, OVV, or PBS. Tumor diameters were measured every 2 days, and volumes were calculated as volume = length×width^2^×0.5. Mice were euthanized when tumor volumes reached 2000 mm^3^. For combination therapy with OVV-αCD47nb and bortezomib (Cat# HY-10227, MCE), the NS-1 subcutaneous model was identically established. Mice received intraperitoneal bortezomib (1 mg/kg) thrice consecutively at 2-day intervals. Intratumoral OVV-αCD47nb (1 × 10^7^ PFU/mouse) was administered thrice at 2-day intervals, starting on day 2 of bortezomib treatment.

Flow cytometry

Fresh tumor tissue was minced and subsequently digested with Collagenase II (Cat# C8150, Solarbio) to obtain a single-cell suspension, which was filtered through a 100 μm nylon mesh. For splenocyte preparation, spleens harvested from euthanized mice were mechanically dissociated into single-cell suspensions by grinding through a 100 μm nylon mesh using a syringe plunger. Cells (2 × 10^6^) were stained in 100 μL PBS. The single-cell suspension was stained with the following antibodies: Live/Dead Fixable Violet Dead Cell Stain (Cat# L34955, Invitrogen), Brilliant Violet 510 anti-CD45 (Cat# 157219, BioLegend), APC anti-CD3 (Cat# 100236, BioLegend), FITC anti-CD4 (Cat# 100406, BioLegend), FITC anti-CD8 (Cat# 100706, BioLegend), APC anti-CD69 (Cat# 104514, BioLegend), APC/Cy7 anti-CD107a (Cat# 121616, BioLegend), APC anti-CD11b (Cat# 101212, BioLegend), PE/Cy7 anti-F4/80 (Cat# 157307, BioLegend), PerCP anti-CD86 (Cat# 105026, BioLegend), FITC anti-CD206 (Cat# 141704, BioLegend), Brilliant Violet 650 anti-LAG-3 (Cat# 125227, BioLegend), Brilliant Violet 711 anti-Tim-3 (Cat# 119727, BioLegend), PerCP/Cy5.5 anti-CD152 (Cat# 106315, BioLegend), PerCP/Cy5.5 anti-TNF-α (Cat# 506322, BioLegend), Alexa Fluor 700 anti-Granzyme B (Cat# 372222, BioLegend) and PE anti-Perforin (Cat# 154306, BioLegend). Following surface staining, cells were incubated for 30 min in the dark. For intracellular staining, samples were fixed and permeabilized using the BD Cytofix/Cytoperm Kit (Cat# 554714, BD Biosciences) and incubated with intracellular antibodies for 30 min. Stained cells were analyzed using a NovoCyte Quanteon flow cytometer (Agilent).

Immunohistochemistry

After establishing the NS-1 tumor model and administering three intratumoral injections, tumor tissues were harvested and fixed in 4% paraformaldehyde. Paraffin-embedded sections (3–5 μm thick) were dewaxed and rehydrated. Endogenous peroxidase activity was quenched by incubating sections in 3% hydrogen peroxide. Sections were then sequentially incubated with primary antibodies: rabbit anti-CD8 (ab217344, Abcam) and rabbit anti-CD68 (ab283654, Abcam), followed by an HRP-conjugated goat anti-rabbit IgG secondary antibody (G1213, Servicebio). Detection was performed using 3,3′-diaminobenzidine (DAB; G1212-2, Servicebio), and sections were counterstained with hematoxylin (G1004, Servicebio). Images were acquired and analyzed using a Nikon Eclipse 100 microscope.

Quantitative polymerase chain reaction (qPCR)

Frozen mouse tissues were pulverized in liquid nitrogen using a mortar and pestle. To assess viral biodistribution, viral DNA was extracted from 25 mg of tissue homogenate using the Viral DNA/RNA Kit (Cat# CW0548S, Cowin Biotech). To evaluate αCD47nb expression, total RNA was isolated with the FastPure Cell/Tissue Total RNA Isolation Kit (Cat# RC101–01, Vazyme Biotech). cDNA was synthesized from 1 μg of RNA using HiScript II Reverse Transcriptase (Cat# R201–01, Vazyme Biotech). Quantitative PCR was performed using 100 ng of viral genomic DNA or 40 ng of cDNA as template on a 7500 Fast Real-Time PCR System. The cycling protocol consisted of an initial denaturation at 95°C for 10 min, followed by 40 cycles of 95°C for 15 s, 60°C for 1 min, and 72°C for 7 min. Primers targeting the viral A56 gene and αCD47nb were as follows: A56-F: 5′-CTGGATCTACACATTCACCGGA-3′; A56-R: 5′-CGGAGTCTCGTCTGTTGTGG-3′; αCD47nb-F: 5′-GGTATTCTTGGTGGCTCTTT-3′; αCD47nb-R: 5′-GTACCATCCCATGTCGTTG-3′.

RNA sequencing (RNA-seq)

Fresh murine tumor tissue was dissociated into a single-cell suspension. Cells were stained with PE anti-mouse CD45 antibody (Cat# 103106, BioLegend) and CD45^+^ cells were isolated using fluorescence-activated cell sorting (FACS). Total RNA was isolated from the sorted cells and mouse tumor tissues using TRIzol reagent (Cat# 15596026, Ambion) and stored at −80°C. Transcriptome analysis was then performed on these RNA samples using the Illumina NovaSeq X Plus sequencing platform (Hangzhou Lianchuan Biological Information Co., Ltd). Bioinformatics analyses were carried out on the OmicStudio platform (https://www.omicstudio.cn/home). These data are available at the NCBI Sequence Read Archive (SRA) under accession code SRP629910.

Quantification and statistical analysis

All statistical analyses were performed using GraphPad Prism software. Differences between groups were assessed by analysis of variance (ANOVA). Survival curves were generated using the Kaplan–Meier method, with between-group statistical significance determined by the log rank test. A p-value <0.05 was considered statistically significant. Data represent the mean ± standard deviation (SD) of ≥three independent experiments. ns, not significant; ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001.