Tumor-intrinsic TTLL12 drives resistance to cancer immunotherapy via modulating myeloid-derived suppressor cells

Abstract

Background: The majority of patients treated with immunotherapy as a standalone therapy experience little to no benefits. Tubulin tyrosine ligase 12 (TTLL12), as a member of the tubulin tyrosine ligase protein family, is associated with the prognosis of patients with cancer and implicated in regulating innate immunity. However, the role of TTLL12 in modulating antitumor immunity remains unclear.

Methods: We analyzed the expression of TTLL12 in several cancer types and evaluated the putative correlation between TTLL12 expression and the immune infiltration in our own proteomic profile of human colorectal cancer. The gain-of-function or loss-of-function approaches were then implemented in vitro and in vivo, followed by flow cytometry to quantify immune cell population frequency. In vitro assays were used to confirm the influence of TTLL12 on the migration and proliferation of the immune subset. Mass spectrometry and chromatin-immunoprecipitation were used to further explore how TTLL12 influenced the immune subset. Finally, whether TTLL12 could enhance the antitumor efficacy of anti-programmed cell death protein 1 (PD-1) therapy was assessed in an immunocompetent mouse model.

Results: We demonstrated that TTLL12 was upregulated in several cancer types and was linked with poor prognosis in patients with colorectal cancer. TTLL12 was found to be negatively correlated with the immune effector signature on the protein level in our proteomic datasets. TTLL12 ectopic expression in tumor cells did not influence cell proliferation but promoted tumor progression in syngeneic mouse models by modulating myeloid-derived suppressor cells (MDSCs), resulting in a suppressive immune response. TTLL12 was further proved to enrich MDSCs by promoting MDSC migration and proliferation in vitro. Mechanistically, tumor-derived TTLL12 induced the secretion of chemokine CCL9 via promoting its transcription, probably through binding to the promoter region of CCL9. Downregulating TTLL12 could significantly improve the efficacy of anti-PD-1 therapy in an immunocompetent murine model.

Conclusion: We identify TTLL12 as a key determinant of suppressive immune response, highlighting a previously unknown therapeutic target to increase the efficacy of antitumor immunotherapy.

Keywords: Cytokine; Immunotherapy; Myeloid-derived suppressor cell - MDSC.

Figure 1. A TTLL12 acts as an oncogene and may be a potential immunosuppressive target. (A) TTLL12 protein expression in CRC tissues and paired non-tumor tissues (n=116). Two-tailed t-tests, ****p<0.0001. (B–C) The protein levels of TTLL12 were detected by IHC analysis based on the CRC samples in our center (n=26). (B) TTLL12 was significantly elevated in the tumor tissues compared with the paired tumor-adjacent tissues. Two-tailed t-tests, ***p<0.001. (C) Representative images of IHC staining of TTLL12 expression in CRC tissues and paired tumor-adjacent tissues. Scale bars: 50 µm. (D–E) Kaplan-Meier survival plots for TTLL12 staining were constructed according to the IHC staining intensity. Patients with low expression of TTLL12 showed better distant metastasis-free survival time than the patients with high expression of TTLL12. Scale bars: 50 µm. (F) The workflow to identify the immunosuppressive role of TTLL12. (G–J) Scatter plot presenting the correlation of TTLL12 with the immune cell subsets, effector and memory CD8⁺ T cell, Type 1 T helper cell and natural killer cells infiltration. The expression levels were derived from 116 CRC samples in The Six Affiliated Hospital, Sun Yat-sen University and the ssGSEA algorithm was used to analyze. CRC, colorectal cancer; IHC, immunohistochemistry; ssGSEA, single sample gene set ecrichment analysis; TTLL12, tubulin tyrosine ligase 12.

Figure 2. TTLL12 promotes tumor progression in an immune-dependent way. (A–B) Western blot analysis of CT26 and B16 cells transfected with plasmid harboring mTTLL12, Psin-mTTLL12 or shRNA targeting mTTLL12, pLKO-mTTLL12 plasmids with TTLL12 antibody. (C–D) Cell proliferation of TTLL12-knockdown or TTLL12-overexpressing CT26 cells in vitro. n=3. Values are represented as mean±SD. Two-way ANOVA, ns: not significant. (E–F) Tumor volume (E) and endpoint tumor weight (F) of vector control and TTLL12-knockdown CT26 tumors. In each case, about 2×10⁵ tumor cells were injected subcutaneously and observed for tumor formation in nude mice. n=7 mice for each group. Values are represented as mean±SEM. P values were calculated by two-way ANOVA in (E) and by two-tailed t-tests in (F), ns: not significant. (G) Evaluation model of antitumor activity in CT26 subcutaneous tumors in Balb/c mice. (H–I) Tumor size and tumor weight of vector control and TTLL12-overexpressing CT26 cell line in Balb/c mice. n=4 mice for both groups. Values are represented as mean±SEM. P values calculated by two-way ANOVA, ***p<0.001, *p<0.05. (J–K) Cell proliferation of TTLL12-knockdown or TTLL12-overexpressing B16 cells in vitro. n=3. Values are represented as mean±SD. Two-way ANOVA, ns: not significant. (L) Evaluation model of antitumor activity in B16 subcutaneous tumors in C57BL/6J mice. (M–P) Tumor size (M) and tumor weight (O) of vector control and TTLL12-overexpressing B16 cell line in C57BL/6J mice. Tumor size (N) and tumor weight (P) of vector control and TTLL12-knockdown B16 cell line in C57BL/6J mice. n=4–6 mice for each group. Values are represented as mean±SEM. P values calculated by two-way ANOVA, ****p<0.0001, *p<0.05. ANOVA, analysis of variance; TTLL12, tubulin tyrosine ligase 12.

Figure 3. TTLL12 reshapes the distribution of immune cells in the TME. (A) The detailed gating strategies used for immune phenotyping. Live cells were selected by the live/dead dye. CD45⁺γδT⁺ cells were defined as γδT cells. CD45⁺NKp46⁺ cells were defined as NK cells. DC cells were defined as the CD45⁺CD11b⁺CD11c⁺ subset. Protumoral macrophages (M2) were defined as the F4/80⁺CD206⁺ subpopulation of the CD45⁺CD11b⁺ subset. MDSCs were defined as the Ly6G⁺Ly6C^low subpopulation of the CD45⁺CD11b⁺ subset. CD8⁺ and CD4⁺ T lymphocytes were from the CD45⁺CD8⁺ and CD45⁺CD4⁺ subpopulation, respectively. Tregs were subdivided from CD4⁺ T lymphocytes and were defined as the CD45⁺CD4⁺Foxp3⁺ population. IFN-γ⁺ and TNF-α⁺ were used to test the CD8⁺ T cells’ activity. (B–E) Proportions of DC cells (B), γδT cells (C), M2 macrophages (D) and Tregs (E) in the TME of BALB/c mice injected subcutaneously with vector control and TTLL12-overexpressing CT26 cell line. Values are represented as mean±SEM. P values were determined by two-tailed t-tests, ns: not significant. (F–G) Representative FACS plots (F) and proportions (G) of G-MDSCs in the TME of BALB/c mice injected subcutaneously with vector control and TTLL12-overexpressing CT26 cell line. Values are represented as mean±SEM. P values were determined by two-tailed t-tests, **p<0.01. (H–I) Proportions of CD8⁺ T cells (H) and NK cells (I) in the TME of BALB/c mice injected subcutaneously with vector control and TTLL12-overexpressing CT26 cell line. Values are represented as mean±SEM. P values were determined by two-tailed t-tests, **p<0.01. (J–K) Representative FACS plots (J) and quantification (D) of IFN-γ expression among CD8⁺ T cells in CT26 tumor. Two-tailed t-tests, *p<0.05. (L–M) Representative FACS plots (J) and quantification (D) of TNF-α expression among CD8⁺ T cells in CT26 tumor. Two-tailed t-tests, *p<0.05. DC, dendritic cell; FACS, fluorescence-activated cell sorter; G-MDSC, granulocytic myeloid-derived suppressor cell; IFN, interferon; NK, natural killer; TME, tumor microenvironment; Treg, regulatory T cells; TTLL12, tubulin tyrosine ligase 12.

Figure 4. TTLL12 facilitates tumor progression in an MDSC-dependent manner. (A) The protein levels of TTLL12 and CD33 were detected by IHC analysis based on CRC samples in our center (n=151). Representative images of IHC staining from two tissue samples were shown. Red arrows: CD33⁺MDSCs. Scale bars: 25 µm. (B) The correlation between the expression of TTLL12 and numbers of CD33⁺MDSCs in CRC tissues. Two-tailed Spearman correlation is reported. **p<0.01. (C–G) Effect of MDSCs depletion on CT26 tumor growth. (C) Experimental protocol of an MDSCs-depletion model by intraperitoneally injecting anti-Ly6G antibodies (200 µg per mouse) and its isotype antibodies (200 µg per mouse) every 3 days into the immunocompetent BALB/c mice bearing CT26 shNC or CT26 shTTLL12 tumors. (D) Flow cytometry analysis for the percentage of MDSCs after anti-Ly6G antibody treatment. (E) Growth curve of the CT26 tumors. Values are represented as mean±SEM. P values calculated by two-way analysis of variance, ns: not significant, *p<0.05, **p<0.01, ***p<0.001. (F–G) Flow cytometry analysis for the percentage of MDSCs and CD8⁺ T cells in tumors and spleens. Values are represented as mean±SEM. P values were determined by two-tailed t-tests, *p<0.05, ***p<0.001, ****p<0.0001. CRC, colorectal cancer; IHC, immunohistochemistry; MDSC, myeloid-derived suppressor cells; TTLL12, tubulin tyrosine ligase 12.

Figure 5. TTLL12 induces immunosuppressive activity, migration and proliferation of MDSCs. (A) Workflow of T-cell suppression assay. (B–C) T-cell suppression assay showed that MDSCs isolated from CT26-shTTLL12 allografts have attenuated ability to suppress T-cell proliferation. Values are represented as mean±SEM. P values were determined by two-tailed t-tests, ****p<0.0001. (D) The specific experimental methods to confirm the migration ability, apoptosis level and differentiation capacity of MDSCs in vitro. (E) The migration of GR1⁺MDSCs (left) and Ly6G⁺MDSCs (right) in vitro induced by the culture supernatants from both CT26 shNC and CT26 shTTLL12 cells. Values are represented as mean±SEM. P values were determined by two-tailed t-tests, *p<0.05, **p<0.01. (F–G) Flow cytometric analysis of mouse BM cells apoptosis in vitro induced by the culture supernatants from both CT26 shNC and CT26 shTTLL12 cells. Values are represented as mean±SEM. P values were determined by two-tailed t-tests, *p<0.05. (F) Representative FACS plots. (H–I) Flow cytometric analysis of the differentiation capacity of MDSCs in vitro induced by the culture supernatants from both CT26 shNC and CT26 shTTLL12 cells. Values are represented as mean±SEM. P values were determined by two-tailed t-tests, **p<0.01. (H) Representative FACS plots. (J) The migration of GR1⁺MDSCs and Ly6G⁺MDSCs in vitro induced by the culture supernatants from both CT26-Psin and CT26-TTLL12 cells. Values are represented as mean±SEM. P values were determined by two-tailed t-tests, **p<0.01. (K) Flow cytometric analysis of mouse BM cells apoptosis in vitro induced by the culture supernatants from both CT26-Psin and CT26-TTLL12 cells. Values are represented as mean±SEM. P values were determined by two-tailed t-tests, *p<0.05. (L) Flow cytometric analysis of the differentiation capacity of MDSCs in vitro induced by the culture supernatants from both CT26-Psin and CT26-TTLL12 cells. Values are represented as mean±SEM. P values were determined by two-tailed t-tests, *p<0.05. 7-AAD, 7-aminoactinomycin D; BM, bone marrow; CFSE, carboxyfluorescein diacetate succinimidyl ester; FACS, fluorescence-activated cell sorter; GM-CSF, granulocyte-macrophage colony stimulating factor; MDSC, myeloid-derived suppressor cells; TTLL12, tubulin tyrosine ligase 12.

Figure 6. TTLL12 promotes chemokine CCL9 secretion in colorectal cancer. (A) The workflow to identify the cytokines bridging the connection between tumor TTLL12 and MDSCs in the tumor microenvironment. (B) The heatmap of 33 candidate proteins with p<0.01 and secretory ability predicted by the Human Protein Atlas database. (C) ELISA analysis in the supernatant of CT26 shNC and CT26 shTTLL12 cells. Values are represented as mean±SEM. P values were determined by two-tailed t-tests, ***p<0.001. (D) qRT-PCR analysis of CT26 shNC and CT26 shTTLL12 cells. Values are represented as mean±SEM. P values were determined by two-tailed t-tests, ***p<0.001. (E) Tumor image from the immunocompetent Balb/c mice subcutaneously implanted with CT26-control cells, TTLL12-overexpressing CT26 cells as well as TTLL12-overexpressing CT26 cells with CCL9 silence. (F–G) Tumor volume (F) and tumor weight (G) in Balb/c mice. n=5 mice for each group. Values are represented as mean±SEM. P values calculated by two-way ANOVA, *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001. (H) Quantification of CCL9 concentration in tumors. Values are represented as mean±SEM. P values calculated by two-way ANOVA, ****p<0.0001. (I) Flow cytometry analysis to test the MDSCs infiltration in tumors. Values are represented as mean±SEM. P values calculated by two-way ANOVA, ***p<0.001, ****p<0.0001. ANOVA, analysis of variance; MDSC, myeloid-derived suppressor cells; qRT-PCR, quantitative reverse transcription-polymerase chain reaction; TTLL12, tubulin tyrosine ligase 12.

Figure 7. TTLL12 induces CCL9 expression, probably by binding to its promoter region. (A) The intracellular location of TTLL12 was detected by IHC analysis based on colorectal cancer samples in our center. Representative images of IHC staining from three tissue samples were shown. Red arrows: cell nucleus. Scale bars: 50 µm (left panel), 25 µm (right panel). (B–C) Western blot analysis of the cytoplasm and nucleus of CT26 cells transfected with plasmid harboring TTLL12 overexpression or knockdown with TTLL12 antibody, α-tubulin (a cytosolic marker) antibody and Lamin-B (a marker for nuclear compartments) antibody. (D) The promoter region of CCL9 covered by the primer. (E–F) ChIP-PCR assay in DLD1/TTLL12 and CT26/TTLL12 cells. Values are represented as mean±SEM. P values were determined by two-tailed t-tests, ****p<0.0001. ChIP, chromatin iImmunoprecipitation; IHC, immunohistochemistry; TTLL12, tubulin tyrosine ligase 12.

Figure 8. Targeting TTLL12 improves the efficacy of anti-PD-1 therapy. (A) Experimental protocol of an anti-PD-1 therapy mice model by intraperitoneally injecting anti-PD-1 antibodies (200 µg per mouse) and its isotype antibodies (200 µg per mouse) every 3 days into the immunocompetent BALB/c mice bearing CT26 shNC or CT26 shTTLL12 tumors. (B) Growth curve of the CT26 tumors. n=6–7 mice for each group. Values are represented as mean±SEM. P values calculated by two-way analysis of variance, ns: not significant, *p<0.05, **p<0.01, ***p<0.001. (C) Tumor weight of the CT26 tumors. Values are represented as mean±SEM. P values calculated by two-tailed t-tests, *p<0.05. (D–E) Flow cytometry analysis to test the MDSCs and CD8⁺ T cells infiltration of the CT26 shNC or CT26 shTTLL12 tumors in the anti-PD-1 therapy model. Values are represented as mean±SEM. P values calculated by two-tailed t-tests, *p<0.05, **p<0.01. MDSC, myeloid-derived suppressor cell; PD-1, programmed cell death protein 1; TTLL12, tubulin tyrosine ligase 12.

Figure 9. Working model of TTLL12-mediated antitumor immune response. Tumor-intrinsic TTLL12 depends on MDSCs to promote tumor progression and knockdown of TTLL12 can enhance the antitumor efficacy of anti-programmed cell death protein 1 therapy in an immunocompetent mouse model. Mechanistically, tumor-derived TTLL12 promotes chemokine CCL9 secretion to modulate MDSCs by inducing the transcriptional expression of CCL9, probably by binding to its promoter region. MDSC, myeloid-derived suppressor cell; NK, natural killer; TTLL12, tubulin tyrosine ligase 12.