Harnessing SARS-CoV-2-specific CD8⁺ T cells to kill target tumor cells for cancer immunotherapy

Dear Editor,

During the coronavirus disease 2019 (COVID-19) pandemic era, many individuals were naturally infected with or vaccinated against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), resulting in the presence of cytotoxic CD8⁺ T cells (CTLs) specific to SARS-CoV-2 antigens, including the spike protein [1, 2]. Although these SARS-CoV-2-specific CTLs cannot eliminate cancer cells lacking the virus antigens on their surface, they can be repurposed to target and eliminate cancer cells by mimicking SARS-CoV-2-infected cells [3, 4]. However, the repurposing of SARS-CoV-2-specific CTLs for cancer immunotherapy remains unexplored. Here, we engineered CD8⁺ T cell epitope-delivering antibodies (TEDbodies) [5] to deliver spike antigen-derived CTL epitope peptides into the tumor cell cytosol. This process led to degradation of the antigenic peptide mediated by the cytosolic ubiquitin proteasome system, generating mature epitopes that bind to the major histocompatibility complex class I (MHC-I), with the epitope peptide-MHC-I complex (pMHCI) presented on the tumor cell surface. Consequently, tumor cells were recognized as SARS-CoV-2-infected cells and killed by the spike-specific CTLs (Figure 1A).

As TEDbody payloads, we selected 2 immunodominant CTL epitope peptides, HLA-A*24:02-restricted ¹²⁰⁸QYIKWPWYI¹²¹⁶ (S_1208-1216) and HLA-A*02:01-restricted ²⁶⁹YLQPRTFLL²⁷⁷ (S_269-277), from the SARS-CoV-2 spike protein [6]. These epitopes can be presented on the target tumor cell surface as S_1208–1216-loaded HLA-A*24:02 (A*24/S₁₂₀₈ pMHCI) and S_269–277-loaded HLA-A*02:01 (A*02/S₂₆₉ pMHCI). From 30 COVID-19-vaccinated donors, we isolated A*24/S₁₂₀₈- and A*02/S₂₆₉-CTLs (Supplementary Table S1). Among these individuals, 28 had detectable A*24/S₁₂₀₈- or A*02/S₂₆₉-CTLs, representing 0.22%-3.20% of all CD8⁺ T cells in their blood (Figure 1B, Supplementary Table S1). The spike-specific CTLs, namely A*24/S₁₂₀₈- and A*02/S₂₆₉-CTLs, were efficiently expanded ex vivo, through cognate epitope peptide stimulation, into a dominant effector memory subset with potent cytotoxicity (Figure 1C, Supplementary Table S2, Supplementary Figures S1-S2).

For cytosolic CTL epitope delivery, TEDbodies were designed by genetically fusing S_1208–1216- and S_269–277-encompassing peptides with natural N/C-terminal flanking residues to the heavy chain C-terminus of the human IgG1/κ antibody, inCT [5, 7], which can specifically penetrate the cytosol of integrin αv-expressing target tumor cells (Figure 1D-E, Supplementary Figure S3A). However, their ability to induce target tumor cell lysis, mediated by A*24/S₁₂₀₈- and A*02/S₂₆₉-CTLs, was limited (Figure 1F-G, Supplementary Figure S3B-C). Therefore, we optimized the N/C-flanking sequences surrounding the mature CTL epitope using in silico proteasomal cleavage prediction tools [8, 9] (Supplementary Figure S4). This sequential N/C-flanking engineering produced optimized inCT†A*24:S#6 and inCT†A*02:S#2 TEDbodies (Figure 1E), which efficiently induced the lysis of HLA-matched target tumor cells mediated by A*24/S₁₂₀₈- and A*02/S₂₆₉-CTLs, respectively (Figure 1F-G). These findings highlight the significant impact of N/C-flanking sequences on antigen processing and pMHCI presentation [10].

Upon tumor cell treatment with inCT†A*24:S#6, A*24/S₁₂₀₈-CTLs specifically lysed HLA-A*24:02⁺ SNU-216 and SW480 cells, while sparing HLA-A*24:02⁻ MDA-MB-231 and LoVo cells (Figure 1H). Similarly, inCT†A*02:S#2 induced A*02/S₂₆₉-CTL-mediated cytolysis, selectively killing HLA-A*02:01⁺ MDA-MB-231 and SW480 cells, while sparing HLA-A*02:01⁻ SNU-216 and LoVo cells (Figure 1I). These findings demonstrated the HLA-type specificity of TEDbodies. The extent of tumor cell lysis was proportional to the TEDbody concentration, effector-to-target ratio (Figure 1H-I), and incubation time (Supplementary Figure S5). The cytosolically inaccessible inCT(AAA)-based fusion antibodies, inCT(AAA)†A*24:S#6 and inCT(AAA)†A*02:S#2, which could not deliver the epitope payload into the tumor cell cytosol [5], caused negligible lysis across all tested cells (Figure 1H-I). These results emphasize the crucial role of the TEDbody-mediated cytosolic delivery of engineered S_1208–1216- and S_269–277-encompassing peptides for the surface presentation of A*24/S₁₂₀₈ and A*02/S₂₆₉ pMHCI antigens, enabling their recognition by cognate spike-specific CTLs.

The TEDbody demonstrated a conventional IgG1 antibody serum half-life and favorable intratumoral distribution in nonobese diabetic/severe combined immunodeficiency/interleukin-2γ-null (NSG) mouse models (Supplementary Figure S6). The in vivo antitumor efficacy of TEDbodies combined with the adoptive transfer of ex vivo-expanded spike-specific CTLs was evaluated using immunodeficient NSG mice carrying pre-established SW480 and SNU-216 cell-derived subcutaneous tumor xenografts (Supplementary Figure S7). Compared to that in the inCT control, inCT†A*24:S#6 and inCT†A*02:S#2 significantly delayed HLA-A*24:02⁺/HLA-A*02:01⁺ SW480 tumor growth, showing tumor growth inhibition rates of approximately 70% and 68%, respectively (Figure 1J, Supplementary Figure S8). Importantly, inCT†A*24:S#6, but not inCT†A*02:S#2, suppressed HLA-A*24:02⁺/HLA-A*02:01⁻ SNU-216 tumor growth (Figure 1K), confirming the HLA-restricted in vivo antitumor activity. In contrast, the cytosolically inaccessible inCT(AAA)†A*24:S#6 and inCT(AAA)†A*02:S#2 controls exhibited no antitumor activity. The injection of S_1208-1216 and S_269-277 CTL epitopes in synthetic peptide form did not result in antitumor activity (Figure 1J-K), despite the substantial target tumor cell lysis in vitro, mediated by spike-specific CTLs through direct extracellular surface loading (Figure 1F-I). This suggests inefficient in vivo pMHCI formation through the extracellular surface loading of epitope peptides after systemic injection. Treatment with inCT†A*24:S#6 and inCT†A*02:S#2 led to a ∼60-fold increase in the total numbers of A*24/S₁₂₀₈- and A*02/S₂₆₉-CTLs, respectively, within the tumor and a ∼3-fold increase in the percentages of cells expressing granzyme B, a cytotoxic effector of activated CTLs, compared to those in the control groups (Figure 1L-M). These findings indicate that TEDbody-mediated pMHCI presentation on target tumors likely induces sustained antigenic spike-specific CTL stimulation, resulting in enhanced tumor infiltration and cytotoxicity within the tumor microenvironment. The materials and methods used in this study were mentioned in Supplementary Materials.

In conclusion, we developed TEDbodies capable of delivering spike protein-derived CTL epitope peptides into the target tumor cell cytosol, repurposing spike-specific CTLs against tumors. We observed the functional activity of A*24/S₁₂₀₈- and A*02/S₂₆₉-CTLs in COVID-19-vaccinated individuals, demonstrating their ability to eliminate target tumor cells presenting corresponding A*24/S₁₂₀₈ and A*02/S₂₆₉ pMHCI antigens. Our results highlight the potential of TEDbody technology to deliver MHC-I-restricted SARS-CoV-2 epitope peptides into the target tumor cell cytosol, harnessing pre-existing SARS-CoV-2-specific CTLs from natural infection or vaccinations to combat cancer.

DECLARATIONS

AUTHOR CONTRIBUTIONS

Yong-Sung Kim conceived and designed the experiments. Keunok Jung, Deok-Han Ko, and Jeong-Yun Jang performed the experiments. Jung Yeon Heo and Young Rong Kim supervised the collection of blood samples from vaccinated donors. All authors analyzed and interpreted the data. Keunok Jung, Deok-Han Ko, and Yong-Sung Kim wrote the manuscript. All authors have read and approved the final version of the manuscript.

COMPETING INTERESTS

The authors declare that they have no competing interests.

FUNDING INFORMATION

This work was supported by the Samsung Future Technology Center (SRFCMA1802-09) and the Korea Health Technology R&D Project (HR22C173402) through the Korea Health Industry Development Institute (KHIDI) funded by the Ministry of Health and Welfare.

ETHICS APPROVAL AND CONSENT TO PARTICIPATE

PBMCs from healthy donors were acquired using protocols approved by the Institutional Review Board of Ajou University (approval ID: AJOUIRB-SMP-2021-238). We obtained written informed consent from individual healthy donors who donated blood samples used in this study. All animal experiments were approved by the Animal and Ethics Review Committee of Woojung Bio Inc. (approval ID: IACUC2003-004) and performed in accordance with the guidelines established by the Institutional Animal Care and Use Committee.

CONSENT FOR PUBLICATION

Not applicable.

AVAILABILITY OF DATA AND MATERIALS

The datasets used in the current study are available from the corresponding author upon reasonable request.