NEK8 kinase-mediated lactate increase impairs antitumor immunity decreasing radiotherapy sensitivity in colorectal cancer

Mingzhou Li; Yunfei Ni; Jieqiong Wu; Xin Zou; Yining Chen; Junfeng Qiu; Yifang Li; Huayu Cai; Li Wang; Feifei Wang; Hongxia Zhang; Fangyi Han; Jinghao Huang; Zilong Chen; Bingyu Xu; Li Liang

doi:10.1038/s41467-026-70657-z

. 2026 Mar 28;17:4565. doi: 10.1038/s41467-026-70657-z

NEK8 kinase-mediated lactate increase impairs antitumor immunity decreasing radiotherapy sensitivity in colorectal cancer

Mingzhou Li ^1,^2,^3,^4,^5,^#, Yunfei Ni ^1,^2,^3,^#, Jieqiong Wu ^1,^2,^3,^#, Xin Zou ^1,^2,^3,^4,^#, Yining Chen ^1,^2,³, Junfeng Qiu ^1,^2,³, Yifang Li ^1,^2,³, Huayu Cai ^1,^2,³, Li Wang ^1,^2,³, Feifei Wang ^1,^2,³, Hongxia Zhang ^1,^2,³, Fangyi Han ^1,^2,³, Jinghao Huang ^1,^2,³, Zilong Chen ^1,^2,³, Bingyu Xu ^1,^2,³, Li Liang ^1,^2,^3,^4,^✉

PMCID: PMC13195097 PMID: 41904143

Abstract

Radiotherapy effectively treats colorectal cancer (CRC), but local recurrence remains common and abscopal effects—regression of tumors distant from irradiated sites—are rarely observed even with immune checkpoint inhibitors. Here we show that the protein kinase NEK8, highly expressed in CRC, promotes radioresistance by suppressing anti-tumor immunity. In radiation-resistant tumors, NEK8 phosphorylates lactate dehydrogenase A (LDHA), driving lactate overproduction. This metabolite promotes histone modifications that silence antigen presentation machinery, while extracellular lactate directly impairs CD8⁺ T cell function, collectively excluding CD8⁺ T cell from the tumor microenvironment. Pharmacological inhibition of NEK8 using CX6258 restores CD8⁺ T cell infiltration and enhances both local and systemic tumor control following radiotherapy. These findings establish NEK8 as a promising therapeutic target for overcoming radioresistance and inducing abscopal responses in CRC.

Subject terms: Radiotherapy, Cancer microenvironment, Colorectal cancer

Radiotherapy response in colorectal cancer (CRC) is often limited and recurrence is common. Here, the authors show that the kinase NEK8 induces radioresistance by lactate-mediated immunosuppression and that inhibiting NEK8 restores antitumor immunity and improves radiotherapy sensitivity in preclinical CRC.

Introduction

Colorectal cancer (CRC) is a highly lethal malignancy, ranking among the top three in both incidence and mortality rates globally. Approximately 22% of colorectal cancer patients are already dealing with advanced distant metastases at the time of initial diagnosis. For these metastatic cases, the primary clinical treatment modalities are mainly radiotherapy and chemotherapy. However, despite ongoing advancements in treatment, many patients still face treatment failure due to local radioresistance.The cytotoxic effect of ionizing radiation (IR) primarily relies on inflicting DNA damage in tumor cells¹. As a result, most radiosensitizing strategies focus on either inhibiting DNA repair or enhancing DNA damage². The immune contexture is vital in radiocurability³. Radiation boosts antitumor immunity via several mechanisms. For example, cellular stress and death caused by radiation release signals like DAMPs (damage-associated molecular patterns), DNA, chemokines, and cytokines, which link innate and adaptive immunity^4–6. Thus, tumor-targeted RT can be seen as an “in situ tumor vaccination.” Here, activated antigen-presenting cells (APCs) process engulfed tumor-associated antigens (TAAs) to cross-prime CD8⁺ T cells, which can drive systemic cancer rejection, known as the abscopal effect^7–9. However, about 30% of CRC patients will develop radiotherapy resistance¹⁰, and the incidence of RT-induced abscopal responses remains extremely low ( < 5%)^11–13. Our limited understanding of how CRC evades the immune system after radiation therapy continues to hinder efforts to enhance radiotherapy efficacy and radiation-induced abscopal immunity.

Understanding the immune mechanisms underlying intrinsic or acquired resistance is crucial to optimize existing combination therapies. In particular, combining radiotherapy with immunotherapies, especially those that systemically activate T cells, is gaining increasing attention^14,15. Previous studies have shown that radiotherapy can have both immunostimulatory effects, such as in situ vaccination and T-cell recruitment, and immunosuppressive effects, like hypoxia and the expansion of regulatory T (Treg) cells^6,16–18. Tumor-derived metabolites play a key role in regulating immune cells within tumor microenvironment (TME). Lactate, the most elevated metabolite in tumors, promotes cancer progression and therapeutic resistance¹⁹. It serves as a carbon source for tumor-infiltrating Treg cells, MDSCs (Myeloid-derived suppressor cells), and M2-like macrophages, fueling their tumor-promoting activities^20–22. High lactate levels in the TME also significantly impair CD8⁺ T cell function. Lactate boosts PD-1 expression on Tregs but suppresses it on effector T cells²³. Moreover, lactate binds directly to GLUT10 (Glucose Transporter Type 10), reducing its glucose transport activity and thus inhibiting CD8⁺ T cell effector functions²⁴. During radiotherapy, pancreatic cancer cells ramp up glycolysis and secrete more lactate. This lactate activates MDSCs, fostering an immunosuppressive microenvironment that promotes pancreatic cancer progression and recurrence²⁵. However, the impact and regulatory mechanisms of lactate on radioresistance and abscopal effect in CRC remain unclear.

In this study, using single-cell RNA sequencing (scRNA-seq), we comprehensively characterized the immune microenvironment changes and mechanisms in radioresistant CRC, with a focus on reinvigorating CD8⁺ T cells to enhance therapeutic efficacy. By conducting RNA-seq on resistant cell lines and analyzing public databases, we identified Never-in-MitosisA-related kinase 8 (NEK8) as a radiotherapy-resistant gene associated with CD8⁺ T cell anti-tumor immunity. Research indicates that NEK8 is closely linked to tumor development. In breast cancer, NEK8 is upregulated and promotes cell proliferation, migration, invasion, and stemness via the β-catenin pathway^26,27. It also drives gastric cancer cell proliferation²⁸. Moreover, NEK8 plays a key role in tumor metabolic reprogramming. For instance, NEK8 interacts with aspartate synthetase (ASNS) to regulate its ubiquitination, influencing tumor growth in gastric cancer²⁹. In CRC, NEK8 enhances c-MYC protein stability by phosphorylating serine 405, thereby promoting CRC progression³⁰. Here, we demonstrate that overexpression of NEK8 in CRC promotes lactate production by activating LDHA. This dual mechanism not only enhances lactate secretion but also increases histone lactylation levels in the promoter regions of MHC-I genes, thereby suppressing their expression. As a result, CD8⁺ T cell infiltration and antitumor activity are inhibited, leading to radioresistance and suppression of the abscopal effect. Targeting NEK8 can enhance the efficacy of radiotherapy in tumors and induce the abscopal effect, thereby inhibiting the growth of distant metastatic tumors.

Results

CD8⁺ T cells as the primary mediators of radiotherapy resistance in colorectal cancer

To investigate the potential mechanisms involved in radiotherapy resistance in CRC, we used a syngeneic mouse tumor model to generate an in vivo resistance model (Fig. 1A). As shown in Fig. S1A–B, after performing cycles of radiotherapy along with serial passage, resistant MC38 (IR-MC38) tumors were nonresponsive to radiotherapy, as shown by the similar tumor growth to the untreated group (Supplemental Fig. 1A–B). Radiation clonogenic assays and CCK8 assays further verified IR-MC38 is a radiotherapy-resistant cell line (Supplemental Fig. 1C–D). To elucidate the tumor immune microenvironment (TIME) differences between the IR-sensitive cell line MC38 and IR-resistant cell line IR-MC38, we characterized CD45⁺ immune cells isolated from MC38 and IR-MC38 tumors by high-throughput single cell RNA sequencing (scRNA-seq). We identified four major cell lineages including myeloid cells, granulocytes, T cells or natural killer (NK) cells and B cells, based on gene expression signatures (Fig. 1B). We then characterized the changes of these cell subtypes in IR-MC38 tumors compared with MC38 tumors. The percentage of T cells or natural killer (NK) cells significantly decreased, while myeloid cells and granulocytes increased (Fig. 1C). ScRNA-seq resolved myeloid cells into nine transcriptionally defined subsets (Supplemental Fig. 1E): C01 (Ccr2⁺ Trem1⁺ M/MdM), C02 (Gpnmb⁺ Mph), C03 (Ms4a7⁺ Mrc1^hi Mph), C04 (Cd3e⁺ Myeloid Cells), C05 (Mki67⁺ Myeloid Cells), C06 (APCs), C07 (Ifitm6⁺ Mono), C08 (Tnfrsf4⁺ Ikzf2⁺ Myeloid Cells), C09 (S100a9⁺ M/MdM), and a residual C10 (Others). Cluster percentages diverged sharply between conditions (Supplemental Fig. 1F): C02 (Gpnmb⁺ Mph), C03 (Ms4a7⁺ Mrc1^hi Mph), and C04 (Cd3e⁺ Myeloid Cells) showed the greatest shifts in IR-MC38 versus MC38 tumors, reflecting radiation-driven remodeling of the myeloid niche.Focusing on the T cells or natural killer (NK) cells compartment, we identified the percentage of CD8⁺ T cells dramatically decreased in IR-MC38 tumors compared with MC38 tumors, while regulatory T cells (Tregs) increased (Fig. 1D–F). Thus, the tumor microenvironments of MC38 tumors and IR-MC38 tumors exhibited significant differences, particularly in the infiltration of immune cells such as CD8⁺ T cells, Tregs, myeloid cells, and granulocytes. Among these, the infiltration of CD8⁺ T cells showed the most pronounced difference.

To investigate the anti-tumor CD8⁺ T cell response in IR-MC38 tumors and MC38 tumors, we clustered Tumor-infiltrating CD8⁺ T cells into four clusters based on well-established marker genes: effector-like (Prf1, Gzmd, Gzmf), memory-like (Xcl1, Ccr7), naive-like (Lef1, Tcf7), and proliferating-like (Mki67, Top2a) (C01–C04) (Fig. 1G, Supplemental Fig. 1G). We found that the majority of CD8⁺ T cells isolated from MC38 tumors clustered in the C01 effector-like cluster and C02 memory-like cluster, while CD8⁺ T cells from IR-MC38 tumors showed increased density in the C03 naive-like cluster and C04 proliferating-like cluster (Fig. 1H). Notably, we observed that CD8⁺ T cells from IR-MC38 tumors displayed decreased expression of cytotoxic molecules and effector molecules compared with MC38 tumors (Supplemental Fig. 1H). Consistent with scRNA-seq, flow cytometry revealed a sharp drop in CD8⁺ T cells and a concomitant surge in Tregs in IR-MC38 tumors (Fig. 1I, Supplemental Fig. 1I). The same analysis showed depletion of effector-like C01 (GZMB⁺ Perforin⁺) and memory-like C02 (CCR7⁺) subsets, offset by expansion of naive-like C03 (TCF7⁺) and proliferating-like C04 (MKi67⁺) subsets (Fig. 1J, Supplemental Fig. 1J).

We further investigated the differentially expressed genes in C01 CD8⁺ T cells from MC38 and IR-MC38 tumors (Supplemental Fig. 1K, Supplementary Data 1) and conducted Gene Ontology (GO) enrichment analysis on the upregulated genes in IR-MC38 C01 CD8⁺ T cells. Our analysis revealed that these genes were significantly enriched in the negative regulation of cytokine production pathway (Supplemental Fig. 1L). These suggested that cytokine secretion by CD8⁺ T cells in IR-MC38 tumors was substantially inhibited, thereby diminishing their anti-tumor activity.

NEK8 is a radiotherapy resistance gene negatively correlated with CD8⁺ T cell anti-tumor immunity

To identify potential oncogenic targets that influenced CD8⁺ T cell infiltration and function during radiotherapy, we first performed RNA sequencing (RNA-seq) between the MC38 cells and IR-MC38 cells without IR. DEG (differentially expressed genes) analysis revealed a series of genes with dramatic differences in expression (Fig. 2A, Supplemental Data 2). Among these DEGs, Nek8, Casp1, Atp1a3, Mgp, and Nes were the top five upregulated genes upon radiotherapy resistance IR-MC38 (Fig. 2A–B). We classified TCGA-COADREAD patients into two groups based on Tcell antitumor immunity using a scoring system³¹. DEGs between high (top 10%, n = 64) and low (bottom 10%, n = 64) CD8⁺ T-score CRC tumors in TCGA (Fig. 2C), and those differentially expressed in IR-resistant vs. IR-sensitive samples in GSE133057 and GSE119409, were identified. NEK8 and CELSR3 were the common genes in all three groups (Fig. 2D). CELSR3 was down-regulated in IR-resistant samples. NEK8 was the sole gene convergently identified by both screens and up-regulated in IR-resistant samples, so we focused our study on NEK8. In the TCGA-COADREAD cohort, NEK8 expression correlated with the radiosensitivity index (RSI)³²—a genomic proxy inversely linked to tumor radiosensitivity—being markedly higher in RSI-high than in RSI-low CRC (Fig. 2E). Furthermore, NEK8 expression exhibited an inverse correlation with the transcriptional levels of genes linked to CD8⁺ T-cell infiltration and antitumor immune responses (Supplemental Fig. 2A). Analysis using the TIMER algorithm revealed that colorectal cancer tumors with elevated NEK8 expression displayed reduced infiltration of CD8⁺ T cells (Fig. 2F). These results suggested that NEK8 might play a critical role in attenuating CD8⁺ T cell-mediated antitumor immunity.

Fig. 2 — Heatmaps (A) and volcano plots (B) revealed significant differentially expressed genes identified through RNA-seq screening between the IR-resistant MC38 cell line and the parental MC38 cell line. Genes meet the specified criteria (p < 0.05 and |log₂fold change | > 1.5). independent samples = 3, Wald’s test. C Heatmap illustrating the clustering of CRC tumors from the TCGA dataset, based on high (top 10%, patients = 64) or low (bottom 10%, patients = 64) CD8⁺ T-cell-mediated antitumor immunity scores. D Venn diagram showing the overlap of differentially expressed genes between tumors with high and low CD8⁺ T-cell scores in TCGA-CRC, and those differentially expressed in IR-resistant versus IR-sensitive samples from GSE133057 and GSE119409 datasets, Wald’s test. E RSI was calculated based on data from the TCGA database, and the expression of NEK8 in the low RSI group and high RSI group was analyzed. means ± SD, two- tailed t test. F TIMER algorithm analysis of tumor-infiltrating immune cells, including CD8⁺ T cells and dendritic cells, in TCGA-CRC stratified by high versus low NEK8 expression. Patients = 620, means ± SD, the box plots centre = median, minima= non-outlier minimum, maxima= non-outlier maximum, bounds of box = upper and lower quartiles, two- tailed t test. G Radiation clonogenic assays were performed on MC38 cells overexpressing *Nek8* and MC38 cells with *Nek8* knockdown, following exposure to increasing doses of ionizing radiation (IR: 0, 2, 4, 6, and 8 Gy). Survival fractions were calculated using the multi-target single-hit model. independent experiments = 3, two- tailed t test. H The proliferation of MC38 cells overexpressing *Nek8* and MC38 cells with *Nek8* knockdown was assessed by CCK-8 assay after 4 Gy irradiation. independent experiments = 3, means ± SD, two-way ANOVA. I MC38-shNC or MC38-shNek8 cells (5 × 10⁵) were injected subcutaneously into mice. When tumor volume reached 100 mm³, tumor-bearing mice were treated with local tumor irradiation (15 Gy, single dose). Tumor growth was monitored. mice = 5, means ± SEM, two-way ANOVA. J CT26-shNC or CT26-shNek8 cells (5 × 10⁵) were injected subcutaneously into mice. When tumor volume reached 100 mm³, tumor-bearing mice were treated with local tumor irradiation (15 Gy, single dose). Tumor growth was monitored. mice = 5, means ± SEM, two-way ANOVA. K IR-MC38-shNC or IR-MC38-shNek8 cells (5 × 10⁵) were injected subcutaneously into mice. When tumor volume reached 100 mm³, tumor-bearing mice were treated with local tumor irradiation (15 Gy, single dose). Tumor growth was monitored. mice = 5, means ± SEM,two-way ANOVA. Source data are provided as a Source data file.

We next examined NEK8’s role in radioresistance. Western blotting confirmed markedly higher NEK8 expression in IR-MC38 than in parental MC38 cells, with the most pronounced elevation observed 24 h after 4 Gy irradiation (Supplemental Fig. 2B). NEK8 has been previously described as a ciliary kinase, we discovered that NEK8 can colocalize with cilia (axoneme marker: acetylated tubulin, a basal body marker: Gamma-tubulin), but a significant amount of NEK8 remains uncolocalized in the cytoplasm, suggesting that NEK8 may function through other pathways in addition to its role via the cilia signaling (Supplemental Fig. 2C).Nek8 was stably knocked down in MC38, IR-MC38, CT26, the melanoma B16 cell lines and the breast cancer 4T1 cell lines by two short hairpin RNAs (shRNA), and overexpressed in MC38 cell lines (Supplemental Fig. 2D). Radiation clonogenic assays and CCK8 assays showed that NEK8 overexpression induced radioresistance in MC38 cells, whereas NEK8 depletion increased radiosensitivity in both MC38 and IR-MC38 cells (Fig. 2G–H, Supplemental Fig. 2E–F). In MC38 model, primary tumor growth was significantly suppressed in Nek8 knockdown tumors compared to the control group (Fig. 2I). Moreover, local irradiation treatment could further inhibit the tumor growth of Nek8 knockdown tumors compared with the control group (Fig. 2I), and this similar phenotype was also observed in the CT26 model (Fig. 2J), while overexpression of Nek8 had opposite effects (Supplemental Fig. 2G). We also implanted Nek8-deficient (shNek8) or Nek8-sufficient (shNC) IR-MC38 cells on the flanks of mice and then monitored tumor growth after local radiation treatment and observed a similar phenotype. In addition, in the IR-MC38 model, local irradiation therapy led to a more pronounced inhibition of tumor growth in Nek8 knockdown tumors compared with the control group (Fig. 2K). Moreover, to verify whether NEK8 has a similar effect in other cancer types, we used melanoma B16 cells and obtained consistent results (Supplemental Fig. 2H). Taken together, these data indicated that deleting NEK8 in cancer cells enhanced the efficacy of radiotherapy.

NEK8 inhibition leads to abscopal effects of radiotherapy in CRC

Radiotherapy can induce antitumor immune responses and has the potential to mediate abscopal effects³³. We investigated whether NEK8 deficiency enhances the abscopal effects of radiotherapy.

We implanted WT MC38 tumors into left flank of C57BL/6 J mice and Nek8-deficient or Nek8-sufficient MC38 tumors into right flank. Right tumors were irradiated at 15 Gy. Abscopal tumor growth was significantly delayed in mice bearing primary Nek8-deficient MC38 tumors post-RT (Fig. 3A Supplementary Fig. 3A). Similar results were obtained in murine CT26 and IR-MC38 colorectal tumors (Fig. 3B, Supplemental Fig. 3B–D).

Fig. 3 — WT MC38 (A) or CT26 (B) tumors were implanted subcutaneously on the left flank of C57BL/6 J mice, while *Nek8*-deficient or *Nek8*-sufficient MC38 (A) or CT26 (B) tumors were simultaneously transplanted on the right flank. When tumor volume reached 100 mm³, right flank tumors were treated with 15 Gy irradiation. Tumor growth was monitored. mice = 5, means ± SEM, two-way ANOVA. C Diagram illustrating the construction of orthotopic and subcutaneous multiple tumor models. Created in BioRender. Mingzhou, L. (2026) https://BioRender.com/wsrgmbf. MC38-luciferase cells (1 × 10⁶) were implanted orthotopically into the cecum, and *Nek8*-deficient or *Nek8*-sufficient MC38 cells (5 × 10⁵) were transplanted into the subcutaneous tissue. The subcutaneous tumors were then treated with 15 Gy radiotherapy 7 days after subcutaneous injection. D Subcutaneous tumor growth curve following radiation treatment (upper) and corresponding gross diagram (lower). mice = 5, means ± SEM,two-way ANOVA. E Representative bioluminescent images of orthotopic tumor (left) and quantitative analysis (right) of mean intensity. mice = 5, means ± SD, two-tailed t test. F Gross diagram of the orthotopic tumor at the non-irradiated abscopal site. G Hematoxylin and eosin (H&E) staining of orthotopic tumors at the non-irradiated abscopal site. Scale bar, 100 μm, independent samples = 5. H Diagram illustrating the construction of CRC liver metastasis and subcutaneous multiple tumor models. Created in BioRender. Mingzhou, L. (2026) https://BioRender.com/wsrgmbf. MC38-luciferase cells (2 × 10⁵) were injected into the spleen to establish a CRC liver metastasis model, and *Nek8*-deficient or *Nek8*-sufficient MC38 cells (5 × 10⁵) were transplanted into subcutaneous tissue. The subcutaneous tumors were then treated with 15 Gy radiotherapy 9 days after subcutaneous injection. I Subcutaneous tumor growth curve following radiation treatment (upper) and corresponding gross diagram (lower). mice = 6, means ± SEM,two-way ANOVA. J Representative bioluminescent images of CRC liver metastasis tumor (left) and quantitative analysis (right) of mean intensity. mice = 6, means ± SD, two-tailed t test. K Gross diagram of the CRC liver metastasis tumor (left, red arrows indicate metastatic lesions), corresponding H&E staining at the non-irradiated abscopal site and number of liver metastatic lesions(right). Scale bar, 100 μm, mice = 6, means ± SD, two-tailed t test. Source data are provided as a Source data file.

Colon orthotopic tumors and orthotopic liver metastasis tumors are more resistant to immunotherapy than subcutaneous colorectal cancer models due to their distinct tumor microenvironments^34,35. To address these challenges, we implanted MC38-luciferase cells orthotopically into the cecum or liver by injecting into the spleen and simultaneously transplanted Nek8-deficient or Nek8-sufficient MC38 cells into the subcutaneous tissue. The subcutaneous tumors were then treated with radiotherapy (15 Gy; Fig. 3C, Fig. 3H). We observed that radiotherapy significantly increased the complete response (CR) rates in both irradiated (subcutaneous) and abscopal (cecal or liver metastasis) tumors in the Nek8-deficient group compared to the control group (Fig. 3D-G, Fig. 3I-K). Similarly, we implanted IR-MC38-luciferase cells orthotopically into the cecum and simultaneously transplanted Nek8-deficient or Nek8-sufficient IR-MC38 cells into the subcutaneous tissue. The subcutaneous tumors were then treated with radiotherapy (15 Gy; Supplemental Fig. 3E). We also observed that radiotherapy significantly increased the complete response (CR) rates in both irradiated (subcutaneous) and abscopal (cecal) tumors in the Nek8-deficient group compared to the control group (Supplemental Fig. 3F-I). These results suggested that tumor cell-intrinsic NEK8 deficiency in irradiated primary tumors restricted the growth of unirradiated secondary tumors by enhancing antitumor adaptive immune responses.

NEK8 impedes CD8⁺ T cell migration and suppressive function after radiotherapy

To determine whether tumor cell-intrinsic NEK8 deficiency is required for the reconstitution of immune cell populations in tumors after radiation, we used flow cytometry to analyze the proportions of various immune cell types within the TME. In MC38 and CT26 tumors, Nek8 knockdown alone increased DCs, M-MDSCs, and CD8⁺ T cells while decreasing macrophages and CD4⁺ T cells. B cells rose only in MC38, neutrophils fell only in CT26. Under RT, Nek8 knockdown increased DCs and CD8⁺ T cells, increased granulocytes only in MC38, increased NK cells and decreased CD4⁺ T cells only in CT26 (Fig. 4A, Supplemental Fig. 4A–C). Inspiring, radiation combined with Nek8 knockdown boosted intratumoral CD8⁺ T-cell infiltration in both Nek8-deficient MC38 and CT26 tumors, an effect recapitulated in IR-MC38 tumors. (Fig. 4A, Supplemental Fig. 4C). Immunohistochemistry (IHC) results were consistent with the flow cytometry data (Supplemental Fig. 4D). To further investigate this, we harvested Nek8-sufficient or Nek8-deficient MC38 tumor fragments containing pre-existing CD8⁺ T cells from CD45.1 wild-type (WT) mice and transplanted them into CD45.2 mice, which were then treated with irradiation. Four days after IR, there was a significant increase in tumor-infiltrating CD45.2⁺ CD8⁺ T cells in Nek8-deficient tumors compared to controls. Conversely, a similar result was obtained (Fig. 4B). We further explored whether newly infiltrating T cells are necessary for the observed effects by administering the sphingosine-1-phosphate receptor agonist fingolimod (FTY720), which blocks T cells from exiting the thymus and secondary lymphoid organs and prevents their migration to sites of inflammation³⁶. In the absence of new lymphocyte infiltration, the radiosensitizing effect and antitumor effect of Nek8 knockdown were abolished (Fig. 4C). Additionally, the proportions of Perforin⁺ and GZMB⁺ cells among tumor-infiltrating CD8⁺ T cells were significantly higher in irradiated Nek8-deficient MC38, CT26 and IR-MC38 tumors compared to controls (Fig. 4D, Supplemental Fig. 4E). These additional effector markers (including IFN-γ, IL-2 and TNF-α) were increased in irradiated Nek8-deficient MC38 tumors, while the exhaustion markers (PD-1, TIM-3 and CTLA4) were decreased in irradiated Nek8-deficient MC38 tumors compared to controls (Supplemental Fig. 4F). This indicated that IR-induced activation of the CD8⁺ T cell antitumor immune response was enhanced in Nek8-deficient tumors. Furthermore, the depletion of CD8⁺ T cells abolished the growth delay of Nek8-deficient tumors after radiation treatment (Fig. 4E, Supplemental Fig. 4G). To investigate the impact of tumor-derived NEK8 on T cells during T cell receptor (TCR)-mediated tumor antigen recognition, we co-cultured MC38-OVA-shNC and MC38-OVA-shNek8 cells (with or without 4 Gy irradiation) with OT-1 CD8⁺ T cells for 24 hours. Notably, OT-1 T cells co-cultured with Nek8-deficient tumor cells exhibited robust production of Perforin and GZMB (Fig. 4F-G), along with enhanced cytotoxic T lymphocyte (CTL) activity against target tumor cells. This effect was further amplified under radiotherapy conditions (Fig. 4G).

Fig. 4 — A Flow cytometry was used to assess the infiltration of CD8⁺T cells in *Nek8*-deficient and *Nek8*-sufficient MC38 subcutaneous tumors. independent samples = 3, means ± SD, one-way ANOVA. B *Nek8*-sufficient or *Nek8*-deficient MC38 tumor fragments from CD45.1 WT mice were transplanted into CD45.2 mice. Three days later, the tumors were treated with local irradiation (15 Gy, single dose). Four days after irradiation, the number of tumor-infiltrating CD45.2⁺ CD8⁺ T cells was determined by flow cytometry. Conversely, *Nek8*-sufficient or *Nek8*-deficient MC38 tumor fragments from CD45.2 WT mice were transplanted into CD45.1 mice, and the same procedure was performed. mice = 5, means ± SD, one-way ANOVA.Schematic illustration were created in BioRender. Mingzhou, L. (2026) https://BioRender.com/wsrgmbf. C Tumor growth curves of *Nek8*-sufficient and *Nek8*-deficient MC38 tumors with FTY720 administration starting from the indicated treatment. mice = 5, two-way ANOVA. D FACS analysis of granzyme B⁺ and perforin⁺ of CD8⁺ T cells in *Nek8*-deficient and *Nek8*-sufficient MC38 or CT26 subcutaneous tumors. independent samples = 3, means ± SD, one-way ANOVA. E Tumor growth curves of *Nek8*-sufficient and *Nek8*-deficient MC38 tumors following administration of anti-CD8 antibodies, starting from the indicated treatment. mice = 6, means ± SEM, two-way ANOVA. F Schematic illustration for analysis of OT-1 CTL cytotoxicity.Created in BioRender. Mingzhou, L. (2026) https://BioRender.com/wsrgmbf. G FACS analysis of granzyme B⁺ and perforin⁺ in anti-CD3/CD28-activated CD8⁺ T cells co-cultured with MC38-OVA-shNC and MC38-OVA-shNek8 cells (with or without 4 Gy irradiation). independent samples = 3, means ± SD, one-way ANOVA. H FACS analysis of tumor-infiltrating CD8⁺ T cells in the non-irradiated abscopal site of *Nek8*-deficient and *Nek8*-sufficient MC38 tumors or IR-MC38 tumors. independent samples of MC38 = 3 and independent samples of IR-MC38 = 4, means ± SD, two-tailed t test. I FACS analysis of granzyme B⁺ and perforin⁺ of CD8⁺T cells in the non-irradiated abscopal site of *Nek8*-deficient and *Nek8*-sufficient MC38 tumors or IR-MC38 tumors. independent samples = 3, means ± SD, two-tailed t test. Source data are provided as a Source data file.

To investigate the abscopal tumor microenvironment, we used flow cytometry to analyze immune cell proportions. Compared to controls, macrophages significantly decreased and CD8⁺ T cells significantly increased in the abscopal tumor of irradiated Nek8-deficient MC38 and IR-MC38 tumors. (Fig. 4H, Supplemental Fig. 4H). Additionally, the proportions of Perforin⁺ and GZMB⁺ cells among tumor-infiltrating CD8⁺ T cells were higher in the abscopal tumor of irradiated Nek8-deficient MC38 and IR-MC38 tumors (Fig. 4I). Notably, blocking CD8⁺ T cells with a neutralizing antibody effectively restored abscopal effects mediated by Nek8 deficiency, while blocking macrophages with anti-CSF1 did not enhance abscopal effects in Nek8-sufficient tumors (Supplemental Fig. 4I). The above results indicated that CD8⁺ T cells were the critical cell population mediating the abscopal effects in the context of NEK8 deficiency.

NEK8 interacts with LDHA to induce its Ser196 phosphorylation and promotes lactate production

MS, conducted post-Co-IP, was performed on MC38 and CT26 cells to identify NEK8-interacting proteins. Ingenuity Pathway Analysis of NEK8-interacting proteins revealed that lactate biosynthetic process and lactate metabolic process were among the top enriched metabolic pathways (Fig. 5A, Supplemental Data 3). We identified lactate metabolism-related proteins (LDHA, LDHC, PARK7) among NEK8-binding proteins. LDHA, which reduces pyruvate to lactate, was our primary focus. Using the HDOCK website, we predicted NEK8-LDHA binding in both humans and mice, confirming direct interaction (Supplemental Fig. 5A). This interaction was further confirmed using Co-IP, which showed that LDHA was a potent NEK8-interacting protein (Fig. 5B). We collected MC38 cells with stable Nek8 overexpression or knockdown and their supernatants, then measured lactate levels. Nek8 knockdown significantly decreased lactate in cells and supernatant, while Nek8 overexpression increased it, with the most pronounced elevation observed 24 h after 4 Gy irradiation (Fig. 5C, Supplemental Fig. 5B). Additionally, Nek8 knockdown inhibited LDH enzymatic activity, whereas Nek8 overexpression enhanced it in MC38, with the most pronounced elevation observed 24 h after 4 Gy irradiation (Fig. 5D, Supplemental Fig. 5C). The above results indicated that NEK8 can significantly enhance the enzymatic activity of LDHA, thereby promoting lactate production.

Fig. 5 — A The results from the MS analysis were analyzed for enrichment and the top enriched metabolic pathways. B MC38 cells were immunoprecipitated with anti-NEK8 or anti-LDHA and analyzed, independent experiments = 3. C Lactate levels in *Nek8* knockdown MC38 cells with or without 4 Gy irradiation, as well as in their culture supernatants, were measured using a lactate assay kit. independent samples = 3, means ± SD, one-way ANOVA. D LDH enzymatic activity was measured in *Nek8*-knockdown MC38 cells, with or without 4 Gy irradiation. independent samples = 3, means ± SD, one-way ANOVA. E Phosphorylation of LDHA Serine-196 was detected in the *Nek8* stable strain with or without 4 Gy radiation. Western blot image (upper) and quantitative data (lower), independent experiments = 3, means ± SD,two- tailed t test. F Tumor growth curves of *Nek8*-overexpressing MC38 tumors following administration of 2-DG or oxamate, starting from the indicated treatment. mice = 5, means ± SEM,two-way ANOVA. G A diagram illustrating the construction of bilateral subcutaneous MC38 tumor models treated with L-sodium lactate (upper) and macroscopic images of the tumors (lower). mice = 5. Created in BioRender. Mingzhou, L. (2026) https://BioRender.com/wsrgmbf. H Tumor growth curves of bilateral subcutaneous MC38 tumors model following administration of L-sodium lactate, starting from the indicated treatment. mice = 5, means ± SEM,two-way ANOVA. I Lactate levels were measured in bilateral subcutaneous MC38 tumors, including both the irradiated side and the abscopal side. independent samples = 3, means ± SD, one-way ANOVA. J FACS analysis of granzyme B⁺ and perforin⁺ in anti-CD3/CD28-activated CD8⁺ T cells co-cultured with MC38-OVA-shNC and MC38-OVA-shNek8 cells, following 4 Gy irradiation with or without lactate. independent samples = 3, means ± SD, one-way ANOVA. K FACS analysis of granzyme B⁺ and perforin⁺ in anti-CD3/CD28-activated CD8⁺ T cells co-cultured with MC38-OVA-vector and MC38-OVA-*Nek8* cells, after 4 Gy irradiation with or without oxamate. independent samples = 3, means ± SD, one-way ANOVA. Source data are provided as a Source data file.

Given that NEK8 is a protein kinase involved in substrate phosphorylation, we hypothesized that it might act as a LDHA kinase. The phosphorylation LDHA by the NEK8 kinase was characterized using [γ−32P]-ATP, with a kinase-inactive NEK8 (NEK8 Lys33Met) as a negative control (Supplemental Fig. 5D), and the level of phosphorylation continuously increased over time (Supplemental Fig. 5E). The Michaelis constant (Km) for the LDHA–NEK8 interaction was determined by autoradiography to be 0.35 μM, with a Vmax of 9.2 × 10⁵ signal per 5 min, indicating that protein LDHA is a highly efficient substrate for NEK8 (Supplemental Fig. 5F-G).An in vitro kinase assay and MS analysis showed that NEK8 phosphorylates LDHA at S196, T213, and S310 (Supplemental Fig. 5H). To investigate this further, we analyzed the conservation of three LDHA phosphorylation sites (T213, S310, and S196) across multiple species. Only the S196 site was highly conserved (Supplemental Fig. 5I), suggesting it may be the primary site through which NEK8 regulates LDHA activity. Additionally, A previous study reported that ULK1 can interact with LDHA and phosphorylate Ser196 under nutrient scarcity, enhancing LDHA activity and lactate production³⁷. This supports our conclusion that LDHA-S196 is a critical site for regulating LDHA activity and lactate production. We developed a polyclonal antibody specific to LDHA phosphorylated at Serine-196 and validated it in LDHA-KO cell (followed by expression of either LDHA^WT, LDHA^S196A, or LDHA^S196D) (Supplemental Fig. 5J). Then, we used the antibody to detect the Nek8 stable strain, and the results showed that Nek8 knockdown significantly reduced LDHA Serine-196 phosphorylation, while Nek8 overexpression significantly increased it. This effect was even more pronounced under radiotherapy conditions (Fig. 5E). Therefore, these results suggested that NEK8 as an upstream regulator mediated LDHA S196 phosphorylation.

To verify whether LDHA is a crucial downstream target molecule of NEK8, we constructed cell lines with Ldha knockdown in Nek8-overexpressing cells and cell lines with Ldha overexpression in Nek8-knockdown cells, confirmed by Western blot (Supplemental Fig. 5K). We then subcutaneously implanted these cell lines into mice to evaluate their responses to radiotherapy. Results showed that Ldha knockdown significantly reversed the radiotherapy resistance mediated by Nek8 overexpression (Supplemental Fig. 5L), while Ldha overexpression significantly rescued the radiosensitization caused by Nek8 knockdown (Supplemental Fig. 5M). Furthermore, we knocked out endogenous Ldha in the Nek8 overexpressing MC38 cells, and overexpressed Ldha^WT, Ldha^S196A, or Ldha^S196D, respectively, followed by validation of successful construction through Western blot analysis (Supplemental Fig. 5N). We then subcutaneously implanted these cell lines into mice to evaluate their responses to radiotherapy. Results showed that LDHA KO or LDHA^S196A significantly reversed the radiotherapy resistance mediated by NEK8 overexpression (Supplemental Fig. 5O). These findings further confirmed that LDHA was a critical downstream target of NEK8.

Lactate inhibits radiosensitization and abscopal effects mediated by NEK8 knockdown

To elucidate the impact of lactate metabolism on the efficacy of radiotherapy and the abscopal effects in CRC, we treated MC38-Vector and MC38-Nek8 subcutaneous tumor models with 2-deoxy-D-glucose (2-DG, a glycolysis inhibitor) and oxamate (a lactate dehydrogenase inhibitor). The results showed that 2-DG and oxamate significantly reversed NEK8-mediated radioresistance and their combination with radiotherapy markedly suppressed tumor growth (Fig. 5F). Moreover, we examined lactate levels in subcutaneous tumor tissues and found that Nek8 overexpression boosts lactate production, especially under radiotherapy. However, this effect was significantly dampened by 2-DG and oxamate (Supplemental Fig. 5P). Then, we treated bilateral subcutaneous tumor models of MC38-shNC and MC38-shNek8 with lactate and found that it significantly rescued the anti-tumor and abscopal effects induced by Nek8 knockdown combined with radiotherapy (Fig. 5G–H). We measured lactate in tumor tissues and serum. Lactate increased serum and bilateral tumor lactate levels in MC38-shNek8 mice (Fig. 5I, Supplemental Fig. 5Q). These findings indicated that lactate might play a crucial role in NEK8-mediated suppression of radiotherapy sensitivity and the abscopal effects in CRC.

To investigate the impact of lactate on CD8⁺ T cell infiltration and function during radiotherapy, we used IHC to examine CD8⁺ T cell infiltration in bilateral tumor tissues from a subcutaneous model treated with lactate. Nek8 knockdown with radiotherapy increased CD8⁺ T cell infiltration in both irradiated and non-irradiated tumors, but exogenous lactate reversed this effect (Supplemental Fig. 5R-S). We co-cultured OT1 CD8⁺ T cells with MC38-OVA-shNek8 or control cells after radiotherapy. One group received lactate, the other PBS. After 24 hours, flow cytometry showed Nek8 knockdown with RT increased Perforin and GZMB secretion, but lactate inhibited this effect (Fig. 5J). In a parallel experiment, OT1 CD8⁺ T cells were co-cultured with MC38-OVA-Nek8 or control cells after radiotherapy. One group received oxamate, the other PBS. Flow cytometry after 24 h showed Nek8 overexpression with RT inhibited Perforin and GZMB secretion, but oxamate reversed this effect (Fig. 5K). The above results demonstrated that lactate significantly inhibited the infiltration and anti-tumor activity of CD8⁺ T cells.

NEK8-mediated H3K18 lactylation inhibits MHC-I gene transcription in CRC

Our previous findings showed that NEK8 not only increased lactate levels in tumor cell culture supernatants but also significantly raised intracellular lactate levels in tumor cells (Supplemental Fig. 5B). Lactate accumulation within cells can drive histone lysine lactylation, and cancer-associated alterations in lactylation can promote tumor progression and immune evasion³⁸. We hypothesized that NEK8-mediated lactate accumulation promotes protein lactylation, influencing gene expression and immune evasion in CRC. Using pan-lactylation antibodies, we found that Nek8 knockdown reduced protein lactylation in MC38 and IR-MC38, especially in histone regions (around 15 kDa), while Nek8 overexpression increased it in MC38 (Fig. 6A, Supplemental Fig. 6A). This effect was more pronounced in cells treated with 4 Gy radiation. Lactylation levels at various histone H3 and H4 sites were also significantly reduced by Nek8 knockdown in MC38 and IR-MC38 (Fig. 6B, Supplemental Fig. 6B). Among them, H3K18la is a key histone lactylation modification involved in tumorigenesis, immune cell regulation, and metabolism-epigenetics crosstalk. Previous studies showed that H3K18la enhances immune evasion in non-small cell lung cancer by activating the POM121/MYC/PD-L1 pathway³⁹. Thus, we focused on investigating NEK8’s effects via H3K18la.

A The pan-lactyl lysine antibody was used to detect lactate protein levels in *Nek8* stable strains, with and without 4 Gy irradiation, independent experiments = 3. B Screened for the primary histone lactylation sites regulated by NEK8 in MC38 cells with *Nek8* knockdown, with and without 4 Gy irradiation, independent experiments = 3. C The heatmap illustrates the distribution of H3K18la peaks surrounding the translation start site (TSS) in MC38-shNC and MC38-shNek8 cells. independent samples = 2. D The genomic distribution of differentially enriched H3K18la peaks in MC38-shNC and MC38-shNek8 cells. E Volcano plots revealed significant differentially expressed genes identified via bulk RNA-seq between MC38-shNek8 + IR and MC38-shNC + IR tumors. Genes meet the specified criteria (p < 0.05 and |log₂fold change | > 2), Wald’s test. F A Venn diagram illustrates the overlap between genes with decreased H3K18la levels in their promoter regions following *Nek8* knockdown (p < 0.05 and log₂fold change <0) and genes that were upregulated in the shNek8 + IR versus shNC + IR comparison (p < 0.05 and log₂fold change > 0), Wald’s test. G Gene Ontology (GO) enrichment analysis of the overlapping genes shown in Fig. 6F was performed using the Metascape website. H Heatmaps of qRT-PCR data were generated to analyze the expression of 21 genes enriched in the R-MMU-983169: Class I MHC mediated antigen processing & presentation pathway in subcutaneous tumor tissues from MC38-shNC+IR and MC38-shNek8+IR groups. independent samples = 3, two-tailed t test. I Levels of the promoter region H3K18la of *Calr*, *Tapbp*, and *Psmb4* genes in Cut-tag sequencing data. J Cut&Tag qPCR was used to detect the levels of H3K18la in the promoter regions of *Calr*, *Tapbp*, and Psmb4 genes in *Nek8* knockdown MC38 cells. independent samples = 3, means ± SD, two-tailed t-test. K Flow cytometry was employed to detect MHC-I expression on the surface of *Nek8*-knockdown MC38 cells, with and without 4 Gy irradiation. independent samples = 4, means ± SD, one-way ANOVA. L Western blot analysis was used to detect the expression levels of H3K18la in *Ldha* stable strains, with and without 4 Gy irradiation, independent experiments = 3. M Flow cytometry was used to assess the expression levels of MHC-I molecules on the surface of *Ldha* stable strains following 4 Gy irradiation. independent samples = 4, means ± SD, one-way ANOVA. Source data are provided as a Source data file.

The H3K18la-regulated downstream targets were identified in MC38-shNek8 through Cut&Tag, which revealed H3K18la enrichment in gene promoter regions (Fig. 6C–D, Supplemental Data 4). We also performed bulk RNA-seq on subcutaneous MC38-shNC and MC38-shNek8 tumor tissues of mice after radiotherapy, identifying 230 upregulated genes and 384 downregulated genes (Fig. 6E, Supplemental Data 5). KEGG enrichment of the differentially expressed genes revealed significant over-representation of the cytokine–cytokine receptor interaction and antigen processing and presentation pathways (Supplemental Fig. 6C). Subsequently, we performed a joint analysis of the Cut&Tag sequencing and bulk RNA-seq results. We intersected the genes with decreased H3K18la levels in their promoter regions following Nek8 knockdown with the genes that were upregulated in the shNek8 + IR vs. shNC + IR comparison, identifying a total of 246 genes (Fig. 6F). These 246 genes were analyzed using the Metascape website, revealing significant enrichment in the R-MMU-983169: Class I MHC mediated antigen processing & presentation pathway (Fig. 6G).

We further used qRT-PCR to analyze the expression of 21 genes enriched in the R-MMU-983169: Class I MHC mediated antigen processing & presentation pathway in subcutaneous tumor tissues from MC38-shNC+IR and MC38-shNek8+IR groups. The results showed that 17 of the 21 genes were significantly upregulated in MC38-shNek8+IR tumor tissues, except Arhgap1, Mdd, Uba52, and Actr1b (Fig. 6H). We analyzed the Cut&Tag sequencing data and found that Nek8 knockdown significantly reduced H3K18la levels in the promoter regions of representative genes including Calr, Tapbp and Psmb4 (Fig. 6I), which was further confirmed by Cut&Tag qPCR experiments (Fig. 6J).

Additionally, we measured MHC-I expression on tumor cells after Nek8 knockdown via flow cytometry. Nek8 knockdown increased MHC-I expression, further enhanced by radiotherapy (Fig. 6K).

We obtained the same results in breast cancer 4T1 cells (Supplemental Fig. 6D). Pharmacologic inhibition of histone acetyltransferase p300 with C646 fully reversed the Nek8-driven elevation of H3K18la and restored MHC-I expression in Nek8-high MC38 cells, establishing a causal link between Nek8-induced histone lactylation and antigen-presentation capacity (Supplemental Fig. 6E). Moreover, we treated MC38-Vector and MC38-Nek8 subcutaneous tumor models with C646 and the results showed that C646 significantly reversed NEK8-mediated radioresistance and their combination with radiotherapy markedly suppressed tumor growth (Supplemental Fig. 6F). In Ldha stable cell lines, Ldha overexpression reversed Nek8 knockdown-induced H3K18la downregulation, while Ldha knockdown reversed Nek8-induced H3K18la upregulation and all other histone H3 and H4 sites can also rescued(Fig. 6L, Supplemental Fig. 6G-H). In the Ldha KO-Nek8 overexpressing MC38 cells (followed by expression of either Ldha^WT, Ldha^S196A, or Ldha^S196D), Ldha KO or Ldha^S196A significantly reversed the upregulation of H3K18la mediated by Nek8 overexpression, while Ldha^S196D only partially reversed it (Supplemental Fig. 6I). Similarly, under radiotherapy, Ldha reintroduction restored MHC-I upregulation caused by Nek8 knockdown, while Ldha knockdown restored MHC-I downregulation induced by Nek8 overexpression (Fig. 6M). These results suggested that NEK8 knockout combined with radiotherapy reduced H3K18la levels in the promoter region of MHC-I genes, thereby promoting MHC-I expression and enhancing antigen presentation.

Pharmacological inhibition of NEK8 enhances responses to radiotherapy and abscopal effects

Given the significant radiosensitization and abscopal effects mediated by NEK8 knockdown, targeting NEK8 could be a promising strategy for cancer treatment. To preclinically prove this concept, we aimed to identify NEK8 inhibitors. By virtual screening of chemical libraries consisting of 50,000 small-molecule drugs, the chemical compounds capable of binding to the ATP binding site (key amino acids include VAL10/VAL18/LYS33) of NEK8 were scored. The top 5 hits were selected (Supplemental Fig. 7A) and their ability to inhibit tumor growth in vitro was tested. Notably, CX6258, a potent and kinase-selective pan-Pim kinase inhibitor⁴⁰, also acts as an inhibitor of haploid germ cell-specific nuclear protein kinase (Haspin)⁴¹ and effectively kills tumors at the nM level (Supplemental Fig. 7B). CX6258 formed two hydrogen bonds with the NEK8 protein (Fig. 7A). We used surface plasmon resonance (SPR) to monitor the real-time interaction between CX6258 and NEK8. The results showed that the dissociation constant (KD) was 3.38 µM, the association rate constant (ka) was 4443.454/Ms, and the dissociation rate constant (kd) was 0.015/s (Fig. 7B). We then purified human NEK8 protein in vitro and conducted a kinase activity assay, revealing that NEK8 can consume ATP at a concentration of 200 nM (Fig. 7C). Further kinase inhibition assays showed that CX6258 inhibited NEK8 kinase activity with an IC50 of 3.011 µM (Fig. 7D-E). These findings demonstrated that CX6258 was a small-molecule inhibitor of NEK8.

Fig. 7 — A 2D diagram (upper) and 3D diagram (lower) illustrating the computer-simulated binding of CX6258 to NEK8. B Surface plasmon resonance (SPR) technology was used to detect the interaction between CX6258 and NEK8. C In vitro detection of NEK8 protein kinase activity. independent experiments = 3, means ± SD, one-way ANOVA. D In vitro kinase inhibition assay of CX6258 on the NEK8 protein by luciferase-coupled ATP assay. independent experiments = 3. E In vitro kinase inhibition assay of CX6258 on the NEK8 protein by autoradiography using [32 P]-γ-ATP. Coomassie staining of the proteins (upper) and autoradiographic signals (lower), independent experiments = 3. F CT26 cells were implanted into BALB/c mice, which were then treated with or without 15 Gy irradiation or CX6258.Tumor growth curves of CT26 tumors following treatment with IR or CX6258. mice = 6, means ± SEM,two-way ANOVA. G MC38 cells were implanted into C57BL/6 J mice, which were then treated with or without 15Gy irradiation or oral administration of CX6258 or anti-PD1. mice = 6, means ± SEM,two-way ANOVA. H–I Bilateral subcutaneous MC38 or IR-MC38 tumors were treated with or without 15 Gy irradiation or CX6258, when the right flank tumor reached 100 mm³. mice = 5, means ± SEM, two-way ANOVA. J Diagram illustrating the construction of orthotopic and subcutaneous multiple tumor models. Created in BioRender. Mingzhou, L. (2026) https://BioRender.com/wsrgmbf. K Subcutaneous tumor growth curve following radiation treatment or oral administration of CX6258 (upper) and corresponding gross diagram (lower). mice = 5, means ± SEM,two-way ANOVA. L Representative bioluminescent images of orthotopic tumor (upper) and quantitative analysis (lower) of mean intensity. mice = 5, means ± SD, one-way ANOVA. M Gross diagram of the orthotopic tumor (upper) and H&E staining of orthotopic tumors at the non-irradiated abscopal site (lower). Scale bar, 100 μm, independent samples = 5. FACS analysis of tumor-infiltrating CD8⁺ T cells (N) or granzyme B⁺ and perforin⁺ of CD8⁺ T cells (O) in MC38 tumors treated with 15 Gy radiotherapy or oral administration of CX6258. independent samples = 3, means ± SD, one-way ANOVA. Source data are provided as a Source data file.

To investigate whether CX6258 improves the response to IR at a similar level as NEK8 genetic deletion, we treated CT26 subcutaneous tumors and AOM/DSS-induced CRC models with CX6258. We found that CX6258 alone inhibited tumor growth, while its combination with radiotherapy significantly amplified this effect without affecting mouse body weight. (Fig. 7F, Supplemental Fig. 7D). Given CX6258’s multiple targets, we tested if its radiosensitizing effect on CRC is mediated by NEK8. Using a subcutaneous MC38-shNek8 tumor model treated with CX6258 and radiotherapy, we found that Nek8 knockdown abolished CX6258’s ability to inhibit tumor growth and enhance radiotherapy efficacy (Supplemental Fig. 7E-F). We also explored whether CX6258 could enhance the efficacy of IR and anti-PD1 treatment using the MC38 model. Compared to any single treatment, combining CX6258 with anti-PD1 significantly slowed MC38 tumor growth. The triple therapy of CX6258, IR, and anti-PD1 delivered the most potent antitumor effects (Fig. 7G).

To examine if CX6258 potentiates the abscopal effects of radiotherapy, we implanted C57BL/6 J mice with bilateral subcutaneous MC38 or IR-MC38 tumors and treated them with or without: (i) 15-Gy irradiation to the right flank tumors, and (ii) oral CX6258. Significant growth delay in abscopal tumors was observed in mice treated with both radiotherapy and CX6258 (Fig. 7H-I). To verify whether CX6258 has a similar effect in other cancer types, we used melanoma B16 cells and obtained consistent results (Supplemental Fig. 7G). Moreover, we implanted MC38-luciferase cells into the cecum and MC38 cells into the subcutaneous tissue. The subcutaneous tumors were then treated with radiotherapy or oral CX6258 (Fig. 7J). We found that combining radiotherapy and CX6258 significantly increased complete response (CR) rates in both irradiated (subcutaneous) and abscopal (cecal) tumors (Fig. 7K-M).

To interrogate the underlying immunological mechanisms, we profiled tumor-infiltrating immune cells in MC38 tumors following CX6258 treatment by flow cytometry. CX6258 alone significantly increased DC and CD8⁺ T cell infiltration, with no significant effects on other immune cells. However, combining CX6258 with radiotherapy further enhanced the infiltration of macrophages, DCs, granulocytes, and CD8⁺ T cells, while suppressing PMN-MDSC, M-MDSC, and neutrophils. Notably, this combination had no significant impact on B cells, CD4⁺ T cells, and NK cells (Fig. 7N, Supplemental Fig. 7H). The combination of CX6258 and radiotherapy significantly enhanced CD8⁺ T cell infiltration and their anti-tumor activity, consistent with our previous findings in irradiated NEK8-deficient tumors (Fig. 7N-O). MC38 tumors treated with irradiation and CX6258 exhibited a significant shift in immune activation profiles, showing markedly elevated effector cytokines (IFN-γ, IL-2, and TNF-α) alongside downregulated exhaustion markers (PD-1, TIM-3, and CTLA-4) compared to the irradiation control group (Supplemental Fig. 7I). We also examined the microenvironment of abscopal tumors. Compared to radiotherapy alone, the combination of CX6258 and radiotherapy significantly increased the infiltration of macrophages, dendritic cells, granulocytes, and CD8⁺ T cells, while reducing M-MDSC infiltration in abscopal tumors (Supplemental Fig. 7J). Moreover, the cytotoxic activity of CD8⁺ T cells was markedly enhanced (Supplemental Fig. 7K). The results demonstrated that the antitumor effect of combining NEK8 inhibition with IR depended on adaptive immunity and mirrored the effects seen in NEK8-deficient tumors.

Clinical relevance of NEK8 expression and LDHA Ser196 phosphorylation levels in CRC patients

According to the TCGA-COADREAD datasets, NEK8 expression was significantly upregulated in CRC (Fig. 8A). Examination of NEK8 in an in-house cohort of 54 CRC patients by in situ multi-color immunofluorescence revealed that NEK8 expression in tumor cells was inversely correlated with CD8⁺ T cell infiltration (Fig. 8B-C). Kaplan–Meier survival analysis of the TCGA-COADREAD datasets revealed that CRC patients with high NEK8 expression had significantly poorer overall survival compared to those with low NEK8 expression (Fig. 8D). IHC showed that the amounts of NEK8 expression, LDHA Ser196 phosphorylation levels and H3K18 lactylation were increased in the radiotherapy-resistant tumors (Fig. 8E–F). NEK8 expression significantly correlated with T stage, TNM stage, and TRG classification but showed no correlation with M stage, histologic grade, or tumor location (Supplemental Table 2). Further analyses demonstrated positive correlations between NEK8 expression, LDHA Ser196 phosphorylation and H3K18 lactylation levels in CRC tissues (Fig. 8G). These findings suggested NEK8 expression and LDHA Ser196 phosphorylation were correlated with clinical radiotherapy resistance to CRC (Fig. 8H).

Discussion

Radiotherapy (RT) is highly effective in controlling localized rectal cancer and has become a standard neoadjuvant treatment modality. However, distant relapse at non-irradiated sites remains a significant cause of disease progression, occurring in up to 30% of patients despite aggressive multimodal therapies⁴². Although the role of the immune system in mediating both local and abscopal antitumor effects of RT has been recognized for decades, the mechanisms underlying the efficiency of immune priming in this context remain poorly defined. Here, we report that the activation of CD8⁺ T cells is the critical bottleneck in converting radiotherapeutic cytotoxicity into effective antitumor immunity. Additionally, we demonstrate that the tumor kinase NEK8 inhibits antigen presentation and CD8⁺ T cell function by promoting lactate production, thereby suppressing radiotherapy efficacy and the abscopal effects.

Here, we found that IR-MC38 tumors had significantly fewer CD8⁺ T cells, correlating with reduced anti-tumor capacity. Through RNA-seq, we identified DEGs between radiotherapy-resistant IR-MC38 cells and parental MC38 cells. By further querying public databases, we identified tumor intrinsic serine/threonine kinase NEK8 as a key candidate associated with radiotherapy resistance and CD8⁺ T cell function. NEK8 is a ciliary kinase, part of the INVC complex, a well defined protein complex that defines a region at the proximal zone of the cilium and is involved in signaling during left-right determination and tissue development⁴³. In colorectal cancer tumors, NEK8 can colocalize with cilia, but a significant amount of NEK8 remains uncolocalized in the cytoplasm, suggesting that NEK8 may function through other pathways in addition to its role via the cilia signaling. NEK8, previously identified as an oncogene in various cancers (e.g., breast^26,27, gastric²⁸, and renal cancers⁴⁴), regulates cytoskeleton maintenance, cilia signaling, and DNA damage response and repair⁴⁵. In CRC, NEK8 enhances c-MYC stability by phosphorylating serine 405, promoting cancer progression³⁰. However, its precise role remains unclear. In the current study, we found that NEK8 was significantly overexpressed in radiotherapy-resistant patients and negatively correlated with CD8⁺ T cell infiltration. In vivo studies showed that NEK8 knockdown significantly enhanced radiotherapy efficacy and the abscopal effects in CRC, while also promoting CD8⁺ T cell infiltration and anti-tumor activity. Since MC38 and CT26 cells exhibit inherent sensitivity to radiotherapy, shNek8 tumors (shNek8+IR/shNek8) and shNC tumors (shNC+IR/shNC) both demonstrate significant effectiveness in response to radiotherapy. However, in the radiotherapy-resistant IR-MC38 tumors, we observed a significantly enhanced sensitivity of shNek8 tumors relative to shNC tumors. These findings suggest that NEK8 may serve as a biomarker for radiation resistance and the abscopal effects in CRC.

Mechanistically, NEK8, a protein kinase, functions primarily through substrate phosphorylation. Using Co-IP and MS, we identified that NEK8 could bind to lactate metabolism-related proteins LDHA, LDHC, and PARK7. LDHA is a key enzyme that converts pyruvate to lactate and is highly expressed in various digestive system tumors^46,47. We found that NEK8 could bind to LDHA, enhancing its enzymatic activity and thereby promoting lactate production. To identify the phosphorylation sites of LDHA mediated by NEK8, we conducted in vitro kinase assays and found that human NEK8 can phosphorylate human LDHA at three sites: T213, S310, and S196. Given that NEK8 also directly binds to LDHA in mice and modulates its activity, we further analyzed the sequence conservation of these sites across species. Only S196 was highly conserved, suggesting it may be the primary site through which NEK8 regulates LDHA activity. Our findings align with previous research showing that the kinase ULK1 can directly interact with LDHA and phosphorylate its Serine-196 site, enhancing LDHA activity and lactate production under nutrient-limited conditions³⁷. Additionally, LDHA’s glutamate 192, S196, and S319 maintain their spatial conformation through hydrogen bonding, facilitating the binding of LDHA histidine-193 (H193) to NADH³⁷. These observations support the critical role of the S196 site in regulating LDHA activity and lactate production. However, our results also suggest that NEK8 may regulate LDHA via multiple phosphorylation sites in humans and mice, with S196 being the most important.

Once dismissed as a waste product of glucose metabolism, lactate is now recognized as a critical regulator of multiple biological processes, including macrophage polarization, T helper cell differentiation, and tumor immune surveillance^48–50. During radiotherapy, pancreatic cancer cells ramp up glycolysis, secreting more lactic acid. This lactic acid activates MDSCs, creating an immunosuppressive environment that fuels cancer progression and recurrence²⁵. Lactic acid also aids DNA damage repair, driving resistance to radiotherapy and chemotherapy⁴⁶. Our study shows that lactic acid significantly hampers the ability of NEK8 knockdown plus radiotherapy to boost CD8⁺ T cell infiltration in both irradiated and non-irradiated tumor tissues. It also curtails the secretion of Perforin and GZMB driven by this combination therapy, thereby weakening CD8⁺ T cell anti-tumor activity. Moreover, lactic acid inhibits antigen presentation by upregulating H3K18la and downregulating MHC-I gene expression, further suppressing CD8⁺ T cell function. These results suggest that lactic acid is a key driver of NEK8-mediated radiotherapy resistance in CRC. Eliminating lactic acid enhances radiotherapy efficacy and amplifies the abscopal effects.

Targeting kinases has proven to be an effective therapeutic strategy⁵¹. As a protein kinase, NEK8 is a potential drug target. However, no specific small-molecule inhibitors for NEK8 have been reported. To identify NEK8-targeted drugs, we used computer-based virtual screening and identified CX6258 as a specific small-molecule inhibitor. Previous research has shown that CX6258 is a potent and selective pan-Pim kinase inhibitor, targeting Pim1, Pim2, and Pim3 with IC50 values of 5 nM, 25 nM, and 16 nM, respectively⁴⁰. Additionally, CX6258 has been identified as an inhibitor of Haspin kinase⁴¹. To determine whether CX6258’s radiosensitizing effect on CRC is mediated through NEK8, we established a subcutaneous MC38-shNek8 tumor model and treated it with either CX6258 alone, radiotherapy alone, or a combination of both. Results showed that in the absence of Nek8, CX6258’s inhibitory effect on tumor growth and its radiosensitizing effect were lost. This suggests that CX6258’s radiosensitizing effect on CRC is primarily driven by targeting NEK8. Combining CX6258 with radiotherapy could be a promising strategy to overcome CRC radioresistance and enhance the abscopal effects. Thus, NEK8 blockade could represent a paradigm shift in radiosensitization, enhancing the antitumor effects of radiotherapy in treated tumors and potentially modifying local radiation to suppress distant metastasis.

Methods

Mice

C57BL/6 and BALB/c mouse strains were procured from the Guangdong Medical Laboratory Animal Center, while CD45.1 mice were acquired from Cyagen Biosciences. OT-1 mice were obtained from Shanghai Model Organisms. All experimental animals, aged between 4 and 7 weeks, were maintained in specific pathogen-free (SPF) facilities at Southern Medical University. AOM/ DSS-induced colorectal cancer mice were male and other mice were female. Sex was not considered as a biological variable in the study design and analysis. The study exclusively used female mice to maintain consistency across experiments and reduce variability in immune microenvironment analyses. Male mice were used in the AOM/DSS model as they are more prone to successful modeling. The animal experiments were performed in compliance with the ethical guidelines and protocols approved by the Institutional Animal Care and Use Committee (IACUC) of Southern Medical University in accordance with the ARRIVE guidelines (Approval No. SMUL202404004).

Clinical specimens

Between January 2017 and January 2022, we collected 54 pathological specimens from colorectal cancer (CRC) patients at the Department of Pathology, Nanfang Hospital, Southern Medical University. Specimens were obtained from pre-treatment colonoscopy biopsies from patients undergoing neoadjuvant chemoradiotherapy, with no prior anti-tumor therapies. Therapeutic responses were evaluated using the tumor regression grading (TRG) system from the 8th edition of the American Joint Committee on Cancer (AJCC)⁵². Patients with TRG 0 or TRG 1 were classified as responders, whereas those with TRG 2 or TRG 3 were considered non-responders. All specimens were collected with written informed consent from the patients. The study was approved by the Ethics Committee of Nanfang Hospital, Southern Medical University (Guangzhou, China), in accordance with the Declaration of Helsinki (Approval No. NFEC-2024-507).

Establishment of radioresistance mouse models and animal studies

To establish an in vivo radiotherapy-resistant model, MC38 cells (5 × 10⁵) were inoculated into the right flank of mice. When tumors reached a volume of 100 mm³, they were irradiated with 8 Gy in three doses, with a one-day interval between each irradiation. After 7 days, mice were euthanized, tumors were dissociated, and tumor cells were sorted using anti-mouse-EPCAM (BioLegend, 118213, RRID:AB_1134105) and cultured. This treatment cycle was repeated four more times to obtain radiotherapy-resistant MC38 cells (IR-MC38 cells). The radiosensitivity of these cells was then evaluated using in vitro clonogenic survival assays, CCK8 experiments, and in vivo subcutaneous tumor experiments.

Tumor implantation and treatment: MC38, CT26, IR-MC38, or B16 cells (5 × 10⁵) were injected subcutaneously (unilateral or bilateral). For the tumor fragment model, MC38 tumors were excised, cut into fragments, and implanted subcutaneously into recipient mice. For the orthotopic model, 1 × 10⁶ MC38-luciferase or IR-MC38-luciferase cells in 25 µL of 100% high-concentration Matrigel (Corning) were implanted into the cecum following the described protocol⁵³. The mouse liver metastasis model was generated by injecting 2×10⁵ MC38-luciferase cells into the spleen. The AOM/DSS-induced colitis-associated CRC model was developed as previously described⁵⁴. When tumors reached ~100 mm³ (subcutaneous/fragment) or after bioluminescence confirmed engraftment (orthotopic/metastatic), mice were randomized to receive 15-Gy focal radiation or sham. In the FTY720 treatment experiments, FTY720 (TargetMol, 162359-55-9) was administered via gavage at 20 µg per dose daily for 7 days. For CD8⁺ T cell or macrophage depletion experiments, 200 µg of anti-CD8α (BioXCell, BE0061) or anti-CSF1 (BioXCell, BE0204) antibody was injected intraperitoneally, beginning one day prior to other treatments and continuing twice weekly. For the 2-Deoxy-D-glucose (2-DG), oxamate, and lactate treatment experiments, daily intraperitoneal injections were administered as follows: 2-DG (TargetMol, 154-17-6) at 10 mg/kg, oxamate (TargetMol,565-73-1) at 5 mg/kg, or sodium lactate (Sigma-Aldrich,71718-10 G) at 1 g/kg. In the C646 treatment experiments, C646 (TargetMol, 328968-36-1) was administered via intraperitoneal injections at 6 mg/kg per dose daily for 8 days. For the anti-PD1 treatment experiments, 200 µg of anti-PD1 antibody (BioXCell, BE0146) was injected intraperitoneally twice weekly for a total of four doses. For the NEK8 inhibitor treatment, 25 mg/kg of CX6258 (TargetMol, 1202916-90-2) was injected intravenously daily.

Disease monitoring and endpoints: Subcutaneous tumors were measured every 2 days with calipers (volume = length × width²/2). Orthotopic cecal and liver metastasis burden was quantified by bioluminescence imaging (Bruker In-Vivo Xtreme) 10 min after i.p. luciferin (200 µL, 15 mg mL⁻¹). Imaging was performed at baseline and 7 days post-treatment, or weekly until endpoint. For the AOM/DSS model, mice were weighed twice weekly;mice were euthanized and colons were collected for tumor counting after the treatment period.

Endpoint criteria: Animals were euthanized when any of the following was met: (i) subcutaneous tumor volume ≥ 2000 mm³ or longest diameter ≥ 1.5 cm; (ii) orthotopic/metastatic bioluminescence signal increased ≥ 10-fold over baseline or showed exponential growth despite therapy; (iii) loss of > 20 % body weight from baseline; (iv) signs of intestinal obstruction, ascites, or severe anemia (pale mucous membranes); (v) inability to access food/water, lethargy, or hunched posture persisting > 24 h. These humane endpoints were approved by the Institutional Animal Care and Use Committee and applied across all models.

Single cell RNA-seq (scRNA) analysis

To obtain a sufficient number of CD45⁺ leukocytes and biological replicates for single-cell RNA sequencing, we isolated CD45⁺ cells using flow cytometry. Cell viability was assessed by trypan blue staining and confirmed to be over 90%. Single-cell capture and library construction were performed by Novogene Co., Ltd. (Beijing, China). Sequencing was conducted using the Illumina 10x Genomics platform. Raw sequencing reads from each sample were processed using Cell Ranger to generate gene expression matrices.

These processed data were analyzed using R version 4.4.0. The analysis pipeline was implemented with the Seurat package (v5.0.0) and other complementary tools as described below.

To ensure the quality of the single-cell data and remove potential doublets, we employed the DoubletFinder package to identify and remove doublets based on gene expression profiles. The raw gene expression matrix was cleaned by removing mitochondrial, ribosomal, and hemoglobin genes.

After data integration and batch effect removal by Harmony algorithm, Uniform Manifold Approximation and Projection (UMAP) was used for nonlinear dimensionality reduction, with the first 25 principal components (PCs) as input. Differential gene expression analysis was performed to identify marker genes for each cluster using the FindAllMarkers function, with a log-fold change threshold of 0.25 and a minimum percentage of cells expressing the gene set to 25%.

Cells were annotated based on their expression profiles and known marker genes. This manual annotation process involved comparing the expression patterns of cells to known cell type-specific markers. The annotation was refined iteratively to ensure consistency and accuracy.

Cell culture

In this study, we used three cell lines: two CRC cell lines (MC38 and CT26), a melanoma cell line (B16) and a breast cancer cell line (4T1), all obtained from the American Type Culture Collection (ATCC). Cells including MC38, IR-MC38, CT26, 4T1 and B16 were maintained in DMEM medium (Gibco) supplemented with 10% fetal bovine serum (FBS) and cultured in a humidified incubator at 37 °C with 5% CO₂. Mycoplasma contamination was routinely monitored and found to be absent.

Plasmid and cell transfections

The Nek8 and Ldha genes were knocked down using shRNA delivered via a pLKO.1 plasmid and their overexpression was achieved using a pCDH plasmid. The Ldha genes were knocked out using sgRNA delivered via a lentiCRISPR v2 plasmid, and Ldha mutant plasmid including Ldha ^S196A, and Ldha ^S196D was achieved using a pCDH plasmid. The shRNA and sgRNA sequences are listed in Supplemental Table 1. Stable MC38 cell lines expressing OVA and luciferase were generated by lentiviral infection. Briefly, Lentivirus was produced by co-transfecting 293 T cells with the packaging vectors psPAX2 and pMD2.G using Lipofectamine 3000 (Invitrogen). The viral supernatant was collected after 48 hours, filtered through a 0.45-µm filter (Millipore), and concentrated. The expression of NEK8, LDHA, and OVA was confirmed by Western blot.

Flow cytometry

To investigate immune cell infiltration, we first isolated tumor-infiltrating lymphocytes (TILs). Tumor tissues were dissected, cut into small pieces, and digested in RPMI 1640 medium containing DNase I (100 U/mL, Sigma, D5025), collagenase VI (210 U/mL, Thermo, 17104019), and hyaluronidase (0.5 mg/mL, Sigma, H3506) for 30 minutes at 37 °C. The dissociated cells were passed through a 70 µm strainer, and the supernatant was collected and centrifuged at 1000 × g for 10 minutes. The cell pellet was then resuspended and subjected to density gradient centrifugation using 40% and 70% Percoll. The interphase containing the TILs was harvested and centrifuged at 1000 × g for 5 minutes. The isolated TILs were incubated with the indicated antibodies including CD4 (BioLegend, 100434, RRID: AB_893324), CD8a (BioLegend, 1623043, RRID: AB_2894434), Granzyme B (BioLegend, 515403, RRID: AB_2114575), Perforin (BioLegend, 154304, RRID: AB_2721463),

TNF-α(BioLegend, 506339, RRID: AB_2563127), IL-2 (BioLegend, 503825, RRID: AB_10895901),

IFN-γ(BioLegend, 505825, RRID: AB_1595591),CD152(BioLegend, 106316, RRID: AB_2564474), TIM-3(BioLegend, 134019, RRID: AB_2814028), PD-1(BioLegend, 109119, RRID: AB_2566640),

CD45 (BioLegend, 103116, RRID: AB_312981), CD3ε (BioLegend, 100357, RRID: AB_3662378), FOXP3 (BioLegend, 118904, RRID: AB_2936574), CD25 (BioLegend, 113709, RRID: AB_3674974), TCF7 (BD Biosciences, 566692, RRID: AB_2869822), CCR7 (BioLegend, 120124, RRID: AB_2616688), Ki67 (BioLegend, 151212, RRID: AB_2814055), CD11c (BioLegend, 117343, RRID: AB_10897814), CD11b (BioLegend, 101211, RRID: AB_312794), LY6C (BioLegend, 128005, RRID: AB_10639728), LY6G (BioLegend, 127607, RRID: AB_1186099), IA/IE (BioLegend, 107641, RRID: AB_2565975), F4/80 (BioLegend, 123113, RRID: AB_893478), CD19 (BioLegend, 115549, RRID: AB_3106188), CD45.1 (BioLegend, 110731, RRID: AB_10896425), CD45.2 (BioLegend, 109837, RRID: AB_2561393)then resuspended and analyzed by flow cytometry.

To assess the clearance efficiency of CD8⁺ T cells, cells were isolated by grinding spleen tissues through 70 µm filters. The cells were then washed twice with PBS, stained with 1:200 dilution of fluorescence-labeled antibodies including CD4 and CD8a for 30 minutes at 4 °C in the dark, and analyzed by flow cytometry.

To determine MHC class I expression, cells were stained with the respective primary antibodies followed by APC-conjugated anti-mouse MHC class I (H-2Kb) (BioLegend,116518, RRID:AB_10564404). All data were acquired using a Fortessa cytometer (BD) and analyzed with FlowJo software (Tree Star). The flow cytometry gating strategy is shown in Supplemental Fig. 8.

Western blot analysis

Western blotting was performed as previously described⁵⁵. Antibodies used were: NEK8 (Abcam, ab116721, RRID: AB_1090250, discontinued reagent); α-tubulin (Proteintech, 11224-1-AP, RRID: AB_2210206); LDHA (Proteintech, 19987-1-AP, RRID: AB_10646429); L-Lactyl Lysine(PTMBIO, PTM-1401RM, RRID: AB_2942013); H3K9la (PTMBIO, PTM-1419RM, RRID: AB_3076695); H3K14la (PTMBIO, PTM-1414RM, RRID: AB_3076697); H3K18la (PTMBIO, PTM-1427RM, RRID: AB_3076698); Histone H3 (PTMBIO, PTM-1001RM, RRID: AB_3676032); H4K5la (PTMBIO, PTM-1407RM, RRID: AB_3096309); H4K8la (PTMBIO, PTM-1415RM, RRID: AB_3101829); H4K12la (PTMBIO, PTM-1411RM, RRID: AB_2941896), H4K16la (PTMBIO, PTM-1417RM, RRID: AB_3101830); Histone H4 (PTMBIO, PTM-1015RM, RRID: AB_3101866); Anti-Rabbit IgG (Cell Signaling Technology, 7074), Anti-Rabbit LDHA (Ser196) (HUABIO, Homemade) and HLA-A (Boster, M00194-4).

RT–qPCR

Total RNA was isolated using a TRIzol reagent (Invitrogen, USA). cDNA was synthesized from RNA samples using the HiScript III RT SuperMix for qPCR ( + gDNA wiper; Vazyme, R323-01). Quantitative PCR (qPCR) was performed with ChamQ SYBR qPCR Master Mix (Vazyme, Q311-03) on an ABI PRISM 7500 Sequence Detection System (Applied Biosystems, USA). Relative gene expression levels were normalized to GAPDH and calculated using the 2^(-ΔΔCT) method. Primer sequences are listed in Supplemental Table 1.

Multiplex immunofluorescence (mfIHC)

Multiplexed immunofluorescence staining was performed using the Opal™ 4-color Manual IHC kit (PerkinElmer, NEL810001KT) according to the manufacturer’s instructions. Primary antibodies were applied sequentially: anti-CD8 (HUABIO, ET1606-31) and anti-NEK8 (Bioss, bs-7815R). This was followed by incubation with HRP-conjugated secondary antibodies and tyramide signal amplification (TSA). Slides were microwaved after each TSA cycle, and nuclei were stained with DAPI for 30 minutes. Sample scanning, spectral unmixing, and signal quantification were performed using an LSM 880 confocal microscope (Zeiss) and ZEN software (Zeiss). The average fluorescence intensity was calculated by dividing the total fluorescence by the area of interest.

Immunohistochemistry (IHC)

IHC staining was performed as previously described using specific antibodies⁵⁵. Primary antibodies used included CD8α Rabbit mAb (cell signaling technology, 98941), CD3ε Rabbit mAb (cell signaling technology, 78588), H3K18la (PTMBIO, PTM-1406RM, RRID: AB_2909438), Anti-Rabbit LDHA (Ser196) (HUABIO, Homemade) and NEK8 mAb (Bioss, 7815 R). Staining intensity was scored as follows: 0 (no staining), 1 (weak, light yellow), 2 (moderate, yellow-brown), and 3 (strong, brown). The proportion of positive tumor cells was scored as: 0 ( < 25%), 1 (26–50%), 2 (51–75%), and 3 ( > 75%). The staining index was calculated by multiplying the intensity score by the proportion score. Results were independently reviewed and scored by two observers. The density of CD3⁺ T cells and CD8⁺ T cells in tumor and stromal regions was assessed by counting the number of positive cells under a 40× objective lens.

Radiation clonogenic assay

The radiation clonogenic assay was performed as previously described¹⁸. Briefly, cells were seeded in six-well plates at densities of 4 × 10², 8 × 10², 1 × 10³, 5 × 10³, and 8 × 10³ cells per well and exposed to 0, 2, 4, 6, and 8 Gy of radiation, respectively. After 14 days of incubation at 37 °C, cells were washed with PBS, fixed with 4% paraformaldehyde for 30 minutes, and stained with crystal violet for 20 minutes. Colonies were then manually counted. The surviving fraction (SF) was calculated and plotted as a dose-response curve, which was fitted to the multi-target single-hit model using the formula: SF = 1- (1-e^(-D/D₀))^N.

Cell counting kit-8 (CCK-8) assay

Cell proliferation in the indicated groups was assessed using the CCK-8 assay (Glpbio, GK10001) according to the manufacturer’s instructions. After the cells adhered to the culture dish, they were either irradiated with 4 Gy using 6 MV X-rays from a Varian 2300 C/D linear accelerator or treated with a gradient concentration of drugs. Subsequently, 10 µL of CCK-8 solution and 90 µL of DMEM medium (total 100 µL) were added to each well, and the cells were cultured for an additional 2 hours. The absorbance (OD value) of each well was measured daily for 5 consecutive days.

Immunofluorescence

Paraformaldehyde (4%) was used to fix the cells for approximately 10 min, followed by incubation with 0.1% Triton X-100 for approximately 30 min. All samples were placed in a 1% BSA solution and closed on a horizontal shaker for 30 min. The primary antibody (axoneme marker: acetylated tubulin (sigma, T7451), a basal body marker: Gamma-tubulin (Boster, BM4273)) solution and incubated overnight at 4 °C. The corresponding fluorescent dye-conjugated secondary antibody was added and incubated in the dark for 1 h, and 10 μL DAPI was added for 15 min. Finally, images were collected using fluorescence microscopy or confocal microscopy.

T-cell co-culture assay

NEK8 knockdown or overexpressing MC38-OVA tumor cells were seeded in a 96-well plate with or without 4 Gy irradiation 24 hours prior and pre-incubated for 2 hours. OT-1 CD8⁺ T cells were isolated from OT-1 mouse spleens using the EasySep Mouse CD8⁺ T Cell Isolation Kit (Stemcell, 19853 A). Purified OT-1 CD8⁺ T cells were co-cultured with tumor cells at a 10:1 ratio for 24 hours in a complete medium containing IL-2. Four hours before collection, Brefeldin A (BioLegend, 420601) was added to inhibit cytokine secretion. T cells were then washed, resuspended in a staining buffer, and stained with anti-CD8a-PE (BioLegend, 162304) for 30 minutes on ice. After washing, intracellular staining was performed with anti-Perforin-APC (BioLegend, 154304) and anti-Granzyme B-FITC (BioLegend, 515403) as previously described. For other treatments, L-lactic acid sodium (20 mM) or oxamate (10 mM) was added to the co-culture of tumor cells and T cells.

Co-Immunoprecipitation (Co-IP) assays

Cell lysates were prepared from the indicated cells using lysis buffer (150 mM NaCl, 10 mM HEPES, pH 7.4, 1% NP-40). Lysates were then incubated with anti-NEK8 or anti-LDHA rabbit antibodies and protein G-conjugated agarose beads, or with Flag affinity agarose (Sigma–Aldrich) at 4 °C overnight. The beads containing the affinity-bound proteins were washed six times with IP wash buffer (150 mM NaCl, 10 mM HEPES, pH 7.4, 0.1% NP-40) and eluted with 1 M glycine (pH 3.0). The eluates were neutralized, mixed with sample buffer, denatured, and analyzed by Western blotting.

Mass spectrometry (MS) analysis

To identify NEK8-binding proteins, MC38 and CT26 cells were transfected with Flag-tagged NEK8. Lysates were immunoprecipitated using flag beads. The beads containing affinity-bound proteins were washed six times with wash buffer (150 mM NaCl, 10 mM HEPES, pH 7.4, 0.1% NP-40) and eluted with 1 M glycine (pH 3.0). The eluates were analyzed by mass spectrometry (MS), and the data were deposited in the NGDC database (accession No. OMIX009338). Supplemental Data 3 provided peptide counts and detailed information on NEK8-binding proteins identified by IP/MS, as well as a comprehensive list of enriched pathways for NEK8-interacting proteins.

Protein expression and purification

BL21 derivative competent cells (Rosetta) infected with human Bub1 baculovirus were harvested 48 hours post-infection and lysed by sonication. His6-Bub1740–1085 was purified and incubated with tobacco etch virus (TEV) protease and 1 mM ATP overnight at 4 °C. The phosphorylated Bub1740–1085 was further purified using Resource S and Superdex 200 columns (GE Healthcare). Bub1740–1085 mutants were expressed and purified using the same protocol. The purified Bub1740–1085 was then incubated with 10 mM ATP in storage buffer (20 mM Tris-HCl [pH 7.7], 150 mM NaCl, 10 mM MgCl₂, 10 mM DTT) for 30 minutes at room temperature and concentrated to 6 mg/ml for crystallization.

In vitro NEK8 kinase assay

The in vitro kinase assay was performed as previously described³⁷. Briefly, LDHA protein was purchased from Abcam (ab93699), and NEK8-His kinase and kinase inactive NEK8 (NEK8 Lys33Met) were expressed in BL21 (Rosetta) competent cells and purified using Ni²⁺-NTA resin (QIAGEN). Recombinant LDHA was incubated with NEK8-His in a kinase buffer (25 mM Tris-HCl [pH 7.5], 5 mM beta-glycerophosphate, 2 mM DTT, 0.1 mM Na₃VO₄, and 10 mM MgCl₂) containing 10 mM ATP in a 50 µL reaction mixture at 37 °C for 20 minutes. The reaction was terminated with a sample buffer and analyzed by MS. The data were deposited in the NGDC database (accession No. OMIX009336).

In addition, another in vitro kinase assay was performed by Nanjing RuiGan Biotechnology Co., Ltd (Nanjing, China) as previously described⁵⁶. Briefly, the 2 μg of NEK8 or kinase-inactive NEK8 (NEK8 Lys33Met), and 4 μg of LDHA proteins were incubated in the kinase reaction buffer (20 mM Tris-HCl, pH7.5, 20 mM MgCl₂, 5 mM EDTA, 1 mM DTT and 100 μM ATP) in the presence of 5 μCi [32 P]-γ-ATP for 2 hours at room temperature with or without kinase inhibitor. The reactions were stopped by adding SDS sample buffer, and protein phosphorylation was visualized by autoradiography (Typhoon 9410) in 10% SDS-PAGE. In vitro phosphorylation kinetics of the NEK8 by LDHA. Two μg of NEK8 were incubated with 0, 0.1, 0.2, 0.5, 1, 2.5, and 5 μM LDHA for 5 minutes. Phosphorylation was detected by autoradiography, and band intensities were quantified using ImageJ. Km and Vmax were determined by fitting the data to the Michaelis–Menten equation in GraphPad Prism 9.

Luciferase-coupled ATP assay

A luciferase-based luminescence assay was used to measure the ATPase activity of NEK8, following the manufacturer’s protocols (Beyotime, S0150M). Briefly, NEK8 protein was incubated with 10 mM ATP, in the presence or absence of CX6258, at 37 °C in kinase buffer (25 mM Tris-HCl [pH 7.5], 5 mM beta-glycerophosphate, 2 mM DTT, 0.1 mM Na₃VO₄, and 10 mM MgCl₂). The reaction was terminated by adding Kinase-Glo reagent (Beyotime, S0150M), and ATPase activity was assessed via luminescence measurements using a microplate reader.

Lactic dehydrogenase (LDH) activity assay

LDH activity was measured using an LDH assay kit (Abbkine KTB1110) according to the manufacturer’s instructions. Briefly, 50 µL of cell lysis buffer or tissue lysis buffer was added to each well of a 96-well plate. Then, 50 µL of LDH working reagent was added to each well and incubated for 30 minutes. All standards and samples were prepared in triplicate, and the assays were performed three times.

Lactate concentration assay

Following the manufacturer’s protocol, cell supernatant, cell lysis buffer, or tissue lysis buffer was processed using a CheKine™ Lactate Assay Kit (Abbkine, KTB1100). Lactate concentration was measured by absorbance at 450 nm.

Surface plasmon resonance (SPR) assay

The interaction between CX6258 and NEK8 was quantified using surface plasmon resonance (SPR) technology on a BIAcore T200 system (GE Healthcare). NEK8 proteins were covalently immobilized onto CM5 sensor chips, and the resulting binding kinetics were analyzed using the BIAcore T200 evaluation software, following the manufacturer’s recommended protocols.

Cut&Tag and Cut&Tag qPCR

The Cut&Tag assay was performed using the hyperactive in situ ChIP Library Prep Kit for Illumina (pG-Tn5) (Vazyme Biotech, TD904) according to the manufacturer’s instructions. Briefly, cells were collected after experimental treatment and bound to concanavalin A beads. The samples were then incubated with an anti-H3K18la primary antibody (PTM Bio, PAP™−599-16), followed by secondary antibodies. The pA-Tn5 transposase was added, and Tn5 transposase-mediated transposition was activated. DNA was extracted and amplified to construct the library, which was purified using VAHTS DNA Clean Beads (Vazyme Biotech, N411). The library was quantified using the VAHTS Library Quantification Kit for Illumina (Vazyme Biotech) and sequenced on the Illumina NovaSeq platform with 150-bp paired-end reads. Data analysis was performed by Epibiote (Guangzhou, China).

For qPCR, immunoprecipitated DNA fragments were purified and subjected to RT-qPCR using primers specific to the promoters of Tapbp, Psmb4, and Calr. IgG was used as a negative control. Primer sequences were listed in Supplemental Table 1.

Bioinformatics analysis

Gene expression data and corresponding clinical information were retrieved from the Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geo/) and The Cancer Genome Atlas (TCGA) database (https://www.cancer.gov/ccg/research/genome-sequencing/tcga). To evaluate CD8⁺ T-cell-mediated antitumor immunity in TCGA-COADREAD patients, we computed a score reflecting CD8⁺ T-cell infiltration and immune response for each sample. This score was generated through single-sample Gene Set Enrichment Analysis (ssGSEA) on the GenePattern platform (https://cloud.genepattern.org/), utilizing previously published gene sets associated with CD8⁺ T-cell activity³¹. Differential gene expression analysis was performed between colorectal cancer (CRC) tumors exhibiting high (top 10%, n = 64) and low (bottom 10%, n = 64) CD8⁺ T-cell scores.

Our study analyzed two GEO cohorts (GSE133057 and GSE119409) to identify differentially expressed genes (DEGs) in IR-resistant samples compared to IR-sensitive samples. Additionally, we used the Kaplan–Meier and log-rank tests, implemented with the R package survival to evaluate the association between NEK8 expression and overall survival in the TCGA-COADREAD cohort.

To investigate the correlation between NEK8 and radiosensitivity, the radiosensitivity index (RSI) was calculated using the reported formula³²: RSI = −0.0098009 × AR + 0.0128283 × JUN + 0.0254552 × STAT1 − 0.0017589 × PRKCB − 0.0038171 × RELA + 0.1070213 × ABL1 − 0.0002509 × SUMO − 0.0092431 × CDK1 + 0.0204469 × HDAC1 − 0.0441683 × IRF1. The RSI was calculated, and patients were divided into RSI-low and RSI-high groups according to the median RSI value in TCGA-COADREAD.

To investigate the correlation between NEK8 and immune infiltration in CRC, the TIMER algorithm was used to estimate immune cell proportions and generate immune scores for each patient in the TCGA-COADREAD cohort.

To determine whether NEK8 binds to LDHA, the structures of LDHA and NEK8 proteins were obtained from the UniProt database. Molecular docking of LDHA and NEK8 was performed using the HDOCK server (http://hdock.phys.hust.edu.cn/). The interactions within the LDHA-NEK8 protein complex were analyzed in detail using PDBePISA and visualized with PyMOL software.

RNA sequencing and dataset analysis

Tumor-bearing C57BL/6 J mice transplanted with shNC or shNEK8 tumors were subjected to 15 Gy radiation and then anesthetized. Tumors and cells were collected, and RNA was extracted using Trizol. RNA sequencing was performed as previously described¹⁸. Gene read counts were determined using Feature Counts, and differential gene expression analysis was conducted using the limma R package to compare MC38 and IR-MC38 groups, as well as shNC+IR and shNek8+IR tumors. Volcano plots of DEGs were generated using the ggplot2 package, and the expression of the top 30 DEGs was visualized in a heatmap created with the pheatmap package.

High-throughput virtual screening (HTVS)

To identify potential NEK8 inhibitors, this study performed a computer-based virtual screening targeting the ATP-binding site of human NEK8 protein (UniProt database), focusing on key amino acids VAL10, VAL18, and LYS33. The protein was prepared using the Protein Preparation Wizard module, which added hydrogen atoms and optimized energy (OPLS2005 force field, RMSD = 0.3 Å). A receptor grid centered on residues VAL10, VAL18, and LYS33 was generated with a box size of 20 Å × 20 Å × 20 Å. The 2D structures of 50,000 compounds from the MCE 50 K Diversity Library were processed using the LigPrep module in Schrödinger software for hydrogenation and energy optimization, then converted to 3D structures for virtual screening. The screening process involved three stages: (1) HTVS mode to screen the initial 50,000 compounds; (2) SP mode to screen the top 10% from HTVS; and (3) XP mode to screen the top 10% from SP, yielding the final ranking of small molecules. The top 5 molecules with the highest absolute docking scores were selected as candidates for further experimental validation (Supplemental Fig. 7A). A higher absolute docking score indicated stronger binding affinity. The 2D and 3D docking modes were visualized using PyMOL.

Quantification and statistical analysis

All results shown were representative of at least three independent experiments. Statistical analyses were performed using GraphPad Prism 9.0 and IBM SPSS Statistics 20. Comparisons between the two groups were made using two-tailed unpaired Student’s t-tests. One-way or two-way ANOVA followed by Tukey’s multiple comparisons test was used for multiple comparisons. Tumor growth curves were analyzed using repeated-measures two-way ANOVA (mixed model) with Tukey’s multiple comparisons test. Correlation coefficients were calculated using Spearman’s rank correlation test. Survival curves were plotted using the Kaplan–Meier method and compared with the log-rank test. Significance levels: *p < 0.05, **p < 0.01, ***p < 0.001, ns = not significant.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Supplementary information

Supplementary Information^{(4.1MB, pdf)}

41467_2026_70657_MOESM2_ESM.pdf^{(35KB, pdf)}

Description of Additional Supplementary Files

Supplementary Data 1^{(17.1KB, xlsx)}

Supplementary Data 2^{(29.9KB, xlsx)}

Supplementary Data 3^{(100.2KB, xlsx)}

Supplementary Data 4^{(105.9KB, xlsx)}

Supplementary Data 5^{(250.3KB, xlsx)}

Reporting summary^{(1.6MB, pdf)}

Transparent Peer Review file^{(2.5MB, pdf)}

Source data

Source Data^{(6.3MB, xlsx)}

Acknowledgements

This work was supported by grants from the Chongqing Technology Innovation and Application Development Special Major Project (CSTB2024TIAD-STX0003 to L.L.), the National Natural Science Foundation of China (Grant Nos. 82273358 to L.L.and 81872041 to L.L.), the Open Project of the Key Laboratory of Tumor Immunopathology, Ministry of Education of China, Chongqing, China (No. 2024jsz1006 to M.Z.L.), the President Foundation of Nanfang Hospital, Southern Medical University (Grant Nos. 2025B026 to M.Z.L. and 2025B034 to J.F.Q.) and the Scientific and Technological Innovation Cultivation of College Students in Guangdong Province (Climbing Program) (pdjh2026bk049 to J.Q.W).

Author contributions

M.Z.L. conceived and designed the study. M.Z.L.,Y.F.N., J.Q.W., Y.N.C, J.F.Q, Y.F.L., H.Y.C, L.W, F.F.W, H.X.Z, F.Y.H, J.H.H, Z.L.C and B.Y.X performed the experiments and analyzed the data. X.Z. conducted the bioinformatics analysis. M.Z.L. and Y.F.N. wrote the first draft of the manuscript. M.Z.L., Y.F.N., and J.Q.W. edited the manuscript,The corresponding author L.L. coordinated the overall organization, design, and writing of the article. All other authors contributed equally to the conception of the study, literature review, and editing of the manuscript and figures. All authors approved the final version of the manuscript.

Peer review

Peer review information

Nature Communications thanks Ilaria Elia, who co-reviewed with Carmen Escalona Noguero, Jian Jian Li, Maria Rodríguez Colman and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Data availability

The raw sequencing data reported in this study have been deposited in the Genome Sequence Archive (GSA) at the National Genomics Data Center (NGDC), China National Center for Bioinformation/Beijing Institute of Genomics, Chinese Academy of Sciences (accession No. PRJCA036988) and are publicly accessible at https://ngdc.cncb.ac.cn/gsa. The scRNA-seq data for IR-resistant IR-MC38 and IR-sensitive MC38 tumors, as well as the RNA-seq data for IR-resistant IR-MC38 and IR-sensitive MC38 cells, have been deposited in the NGDC database (accession nos. CRA023580 and CRA023695, respectively). Additionally, the data on NEK8-binding proteins identified by IP/MS in MC38 and CT26 cells, the in vitro kinase assay data screening NEK8 phosphorylation sites on LDHA, the CUT&Tag data for H3K8la in MC38-shNek8 and MC38-shNC cells, and the RNA-seq data for shNC_IR and shNek8_IR tumors are available in the NGDC database under accession nos. OMIX009338,OMIX009336(https://ngdc.cncb.ac.cn/omix/release/OMIX009336), CRA023632, and CRA023578, respectively. This paper does not report the original code. All the other data supporting the findings of this study are available within the article and its Supplementary Information files. Source data are provided as a Source Data file. Source data are provided with this paper.

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

These authors contributed equally: Mingzhou Li, Yunfei Ni, Jieqiong Wu, Xin Zou.

Supplementary information

The online version contains supplementary material available at 10.1038/s41467-026-70657-z.

References

1.Hopkins, J. L., Lan, L. & Zou, L. DNA repair defects in cancer and therapeutic opportunities. Genes Dev.36, 278–293 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Schulz, A., Meyer, F., Dubrovska, A. & Borgmann, K. Cancer stem cells and radioresistance: DNA repair and beyond. Cancers (Basel)11, 10.3390/cancers11060862 (2019). [DOI] [PMC free article] [PubMed]
3.Fridman, W. H., Zitvogel, L., Sautès–Fridman, C. & Kroemer, G. The immune contexture in cancer prognosis and treatment. Nat. Rev. Clin. Oncol.14, 717–734 (2017). [DOI] [PubMed] [Google Scholar]
4.Herrera, F. G., Bourhis, J. & Coukos, G. Radiotherapy combination opportunities leveraging immunity for the next oncology practice. CA Cancer J. Clin.67, 65–85 (2017). [DOI] [PubMed] [Google Scholar]
5.Krysko, D. V. et al. Immunogenic cell death and DAMPs in cancer therapy. Nat. Rev. Cancer12, 860–875 (2012). [DOI] [PubMed] [Google Scholar]
6.Deng, L. et al. STING-dependent cytosolic DNA sensing promotes radiation-induced type I interferon-dependent antitumor immunity in immunogenic tumors. Immunity41, 843–852 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Golden, E. B., Marciscano, A. E. & Formenti, S. C. Radiation therapy and the in situ vaccination approach. Int. J. Radiat. Oncol. Biol. Phys.108, 891–898 (2020). [DOI] [PubMed] [Google Scholar]
8.Wei, J. et al. Sequence of αPD-1 relative to local tumor irradiation determines the induction of abscopal antitumor immune responses. Sci. Immunol.6, 10.1126/sciimmunol.abg0117 (2021). [DOI] [PubMed]
9.Hsieh, R. C. et al. ATR-mediated CD47 and PD-L1 up-regulation restricts radiotherapy-induced immune priming and abscopal responses in colorectal cancer. Sci. Immunol.7, eabl9330 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Benson, A. B. et al. Rectal cancer, version 2.2022, NCCN clinical practice guidelines in oncology. J. Natl. Compr. Canc Netw.20, 1139–1167 (2022). [DOI] [PubMed] [Google Scholar]
11.Postow, M. A. et al. Immunologic correlates of the abscopal effect in a patient with melanoma. N. Engl. J. Med.366, 925–931 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Abuodeh, Y., Venkat, P. & Kim, S. Systematic review of case reports on the abscopal effect. Curr. Probl. Cancer40, 25–37 (2016). [DOI] [PubMed] [Google Scholar]
13.Formenti, S. C. et al. Radiotherapy induces responses of lung cancer to CTLA-4 blockade. Nat. Med24, 1845–1851 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Postow, M. A. et al. A prospective, phase 1 trial of nivolumab, ipilimumab, and radiotherapy in patients with advanced melanoma. Clin. Cancer Res26, 3193–3201 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Weichselbaum, R. R., Liang, H., Deng, L. & Fu, Y. X. Radiotherapy and immunotherapy: a beneficial liaison?. Nat. Rev. Clin. Oncol.14, 365–379 (2017). [DOI] [PubMed] [Google Scholar]
16.Vanpouille-Box, C. et al. DNA exonuclease Trex1 regulates radiotherapy-induced tumour immunogenicity. Nat. Commun.8, 15618 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Muroyama, Y. et al. Stereotactic radiotherapy increases functionally suppressive regulatory T cells in the tumor microenvironment. Cancer Immunol. Res5, 992–1004 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Li, M. et al. PREX2 contributes to radiation resistance by inhibiting radiotherapy-induced tumor immunogenicity via cGAS/STING/IFNs pathway in colorectal cancer. BMC Med22, 154 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Certo, M., Tsai, C. H., Pucino, V., Ho, P. C. & Mauro, C. Lactate modulation of immune responses in inflammatory versus tumour microenvironments. Nat. Rev. Immunol.21, 151–161 (2021). [DOI] [PubMed] [Google Scholar]
20.Watson, M. J. et al. Metabolic support of tumour-infiltrating regulatory T cells by lactic acid. Nature591, 645–651 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Noe, J. T. et al. Lactate supports a metabolic-epigenetic link in macrophage polarization. Sci. Adv.7, eabi8602 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
22.He, J. et al. The lactate receptor HCAR1 drives the recruitment of immunosuppressive PMN-MDSCs in colorectal cancer. Nat. Immunol. 10.1038/s41590-024-02068-5 (2025). [DOI] [PubMed]
23.Xu, W. et al. N6-methyladenosine modification of 3’tRF-AlaAGC impairs PD-1 blockade efficacy by promoting lactic acid accumulation in the tumor microenvironment of gastric carcinoma. Drug Resist Updat79, 101197 (2024). [DOI] [PubMed] [Google Scholar]
24.Liu, Y. et al. Activation and antitumor immunity of CD8+ T cells are supported by the glucose transporter GLUT10 and disrupted by lactic acid. Sci. Transl. Med16, eadk7399 (2024). [DOI] [PubMed] [Google Scholar]
25.Yang, X. et al. Lactate-modulated immunosuppression of myeloid-derived suppressor cells contributes to the radioresistance of pancreatic cancer. Cancer Immunol. Res8, 1440–1451 (2020). [DOI] [PubMed] [Google Scholar]
26.Kang, E., Kim, H. K., Lee, H. B. & Han, W. Never in mitosis gene A-related kinase-8 promotes proliferation, migration, invasion, and stemness of breast cancer cells via β-catenin signalling activation. Sci. Rep.13, 6829 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Bowers, A. J. & Boylan, J. F. Nek8, a NIMA family kinase member, is overexpressed in primary human breast tumors. Gene328, 135–142 (2004). [DOI] [PubMed] [Google Scholar]
28.Ding, X. F., Chen, J., Zhou, J., Chen, G. & Wu, Y. L. Never-in-mitosis A-related kinase 8, a novel target of von-Hippel-Lindau tumor suppressor protein, promotes gastric cancer cell proliferation. Oncol. Lett.16, 5900–5906 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Wang, M., Yu, K., Meng, F., Wang, H. & Li, Y. NEK8 promotes the progression of gastric cancer by reprogramming asparagine metabolism. Mol. Med31, 3 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Cao, B., Zhang, K., Pan, C., Dong, Y. & Lu, F. NEK8 regulates colorectal cancer progression via phosphorylating MYC. Cell Commun. Signal21, 209 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Cheng, W. C. et al. Uncoupling protein 2 reprograms the tumor microenvironment to support the anti-tumor immune cycle. Nat. Immunol.20, 206–217 (2019). [DOI] [PubMed] [Google Scholar]
32.Mohammadi, H. et al. Using the Radiosensitivity Index (RSI) to predict pelvic failure in endometrial cancer treated with adjuvant radiation therapy. Int J. Radiat. Oncol. Biol. Phys.106, 496–502 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Rodríguez-Ruiz, M. E., Vanpouille-Box, C., Melero, I., Formenti, S. C. & Demaria, S. Immunological mechanisms responsible for radiation-induced abscopal effect. Trends Immunol.39, 644–655 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Oliveira, R. C., Abrantes, A. M., Tralhão, J. G. & Botelho, M. F. The role of mouse models in colorectal cancer research-The need and the importance of the orthotopic models. Anim. Model Exp. Med3, 1–8 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Ho, W. W. et al. Dendritic cell paucity in mismatch repair-proficient colorectal cancer liver metastases limits immune checkpoint blockade efficacy. Proc. Natl. Acad. Sci. USA118, 10.1073/pnas.2105323118 (2021). [DOI] [PMC free article] [PubMed]
36.Halin, C. et al. The S1P-analog FTY720 differentially modulates T-cell homing via HEV: T-cell-expressed S1P1 amplifies integrin activation in peripheral lymph nodes but not in Peyer patches. Blood106, 1314–1322 (2005). [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Jia, M. et al. ULK1-mediated metabolic reprogramming regulates Vps34 lipid kinase activity by its lactylation. Sci. Adv.9, eadg4993 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Zhang, D. et al. Metabolic regulation of gene expression by histone lactylation. Nature574, 575–580 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Zhang, C. et al. H3K18 lactylation potentiates immune escape of non-small cell lung cancer. Cancer Res. 10.1158/0008-5472.Can-23-3513 (2024). [DOI] [PubMed]
40.Haddach, M. et al. Discovery of CX-6258. A potent, selective, and orally efficacious pan-pim kinases inhibitor. ACS Med. Chem. Lett.3, 135–139 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Melms, J. C. et al. Inhibition of haspin kinase promotes cell-intrinsic and extrinsic antitumor activity. Cancer Res80, 798–810 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Ogura, A. et al. Neoadjuvant (Chemo)radiotherapy with total mesorectal excision only is not sufficient to prevent lateral local recurrence in enlarged nodes: results of the multicenter lateral node study of patients with low cT3/4 rectal cancer. J. Clin. Oncol.37, 33–43 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Roig, J. NEK8, a NIMA-family protein kinase at the core of the ciliary INV complex. Cell Commun. Signal23, 170 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Ding, X. F., Zhou, J., Hu, Q. Y., Liu, S. C. & Chen, G. The tumor suppressor pVHL down-regulates never-in-mitosis A-related kinase 8 via hypoxia-inducible factors to maintain cilia in human renal cancer cells. J. Biol. Chem.290, 1389–1394 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
45.Panchal, N. K. & Evan Prince, S. The NEK family of serine/threonine kinases as a biomarker for cancer. Clin. Exp. Med23, 17–30 (2023). [DOI] [PubMed] [Google Scholar]
46.Chen, H. et al. NBS1 lactylation is required for efficient DNA repair and chemotherapy resistance. Nature10.1038/s41586-024-07620-9 (2024). [DOI] [PMC free article] [PubMed]
47.Sharma, D., Singh, M. & Rani, R. Role of LDH in tumor glycolysis: Regulation of LDHA by small molecules for cancer therapeutics. Semin Cancer Biol.87, 184–195 (2022). [DOI] [PubMed] [Google Scholar]
48.Peng, M. et al. Aerobic glycolysis promotes T helper 1 cell differentiation through an epigenetic mechanism. Science354, 481–484 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
49.Colegio, O. R. et al. Functional polarization of tumour-associated macrophages by tumour-derived lactic acid. Nature513, 559–563 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
50.Brand, A. et al. LDHA-associated lactic acid production blunts tumor immunosurveillance by T and NK cells. Cell Metab.24, 657–671 (2016). [DOI] [PubMed] [Google Scholar]
51.Attwood, M. M., Fabbro, D., Sokolov, A. V., Knapp, S. & Schiöth, H. B. Trends in kinase drug discovery: targets, indications and inhibitor design. Nat. Rev. Drug Discov.20, 839–861 (2021). [DOI] [PubMed] [Google Scholar]
52.Mace, A. G., Pai, R. K., Stocchi, L. & Kalady, M. F. American Joint Committee on Cancer and College of American Pathologists regression grade: a new prognostic factor in rectal cancer. Dis. Colon Rectum58, 32–44 (2015). [DOI] [PubMed] [Google Scholar]
53.Xu, Y., Zhang, L., Wang, Q. & Zheng, M. Comparison of different colorectal cancer with liver metastases models using six colorectal cancer cell lines. Pathol. Oncol. Res26, 2177–2183 (2020). [DOI] [PubMed] [Google Scholar]
54.Isermann, T. et al. Suppression of HSF1 activity by wildtype p53 creates a driving force for p53 loss-of-heterozygosity. Nat. Commun.12, 4019 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
55.Zeng, Z. et al. Cancer-derived exosomal miR-25-3p promotes pre-metastatic niche formation by inducing vascular permeability and angiogenesis. Nat. Commun.9, 5395 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
56.Czarnecki, P. G. et al. ANKS6 is the critical activator of NEK8 kinase in embryonic situs determination and organ patterning. Nat. Commun.6, 6023 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Information^{(4.1MB, pdf)}

41467_2026_70657_MOESM2_ESM.pdf^{(35KB, pdf)}

Description of Additional Supplementary Files

Supplementary Data 1^{(17.1KB, xlsx)}

Supplementary Data 2^{(29.9KB, xlsx)}

Supplementary Data 3^{(100.2KB, xlsx)}

Supplementary Data 4^{(105.9KB, xlsx)}

Supplementary Data 5^{(250.3KB, xlsx)}

Reporting summary^{(1.6MB, pdf)}

Transparent Peer Review file^{(2.5MB, pdf)}

Source Data^{(6.3MB, xlsx)}

Data Availability Statement

[CR1] 1.Hopkins, J. L., Lan, L. & Zou, L. DNA repair defects in cancer and therapeutic opportunities. Genes Dev.36, 278–293 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR2] 2.Schulz, A., Meyer, F., Dubrovska, A. & Borgmann, K. Cancer stem cells and radioresistance: DNA repair and beyond. Cancers (Basel)11, 10.3390/cancers11060862 (2019). [DOI] [PMC free article] [PubMed]

[CR3] 3.Fridman, W. H., Zitvogel, L., Sautès–Fridman, C. & Kroemer, G. The immune contexture in cancer prognosis and treatment. Nat. Rev. Clin. Oncol.14, 717–734 (2017). [DOI] [PubMed] [Google Scholar]

[CR4] 4.Herrera, F. G., Bourhis, J. & Coukos, G. Radiotherapy combination opportunities leveraging immunity for the next oncology practice. CA Cancer J. Clin.67, 65–85 (2017). [DOI] [PubMed] [Google Scholar]

[CR5] 5.Krysko, D. V. et al. Immunogenic cell death and DAMPs in cancer therapy. Nat. Rev. Cancer12, 860–875 (2012). [DOI] [PubMed] [Google Scholar]

[CR6] 6.Deng, L. et al. STING-dependent cytosolic DNA sensing promotes radiation-induced type I interferon-dependent antitumor immunity in immunogenic tumors. Immunity41, 843–852 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR7] 7.Golden, E. B., Marciscano, A. E. & Formenti, S. C. Radiation therapy and the in situ vaccination approach. Int. J. Radiat. Oncol. Biol. Phys.108, 891–898 (2020). [DOI] [PubMed] [Google Scholar]

[CR8] 8.Wei, J. et al. Sequence of αPD-1 relative to local tumor irradiation determines the induction of abscopal antitumor immune responses. Sci. Immunol.6, 10.1126/sciimmunol.abg0117 (2021). [DOI] [PubMed]

[CR9] 9.Hsieh, R. C. et al. ATR-mediated CD47 and PD-L1 up-regulation restricts radiotherapy-induced immune priming and abscopal responses in colorectal cancer. Sci. Immunol.7, eabl9330 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR10] 10.Benson, A. B. et al. Rectal cancer, version 2.2022, NCCN clinical practice guidelines in oncology. J. Natl. Compr. Canc Netw.20, 1139–1167 (2022). [DOI] [PubMed] [Google Scholar]

[CR11] 11.Postow, M. A. et al. Immunologic correlates of the abscopal effect in a patient with melanoma. N. Engl. J. Med.366, 925–931 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR12] 12.Abuodeh, Y., Venkat, P. & Kim, S. Systematic review of case reports on the abscopal effect. Curr. Probl. Cancer40, 25–37 (2016). [DOI] [PubMed] [Google Scholar]

[CR13] 13.Formenti, S. C. et al. Radiotherapy induces responses of lung cancer to CTLA-4 blockade. Nat. Med24, 1845–1851 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR14] 14.Postow, M. A. et al. A prospective, phase 1 trial of nivolumab, ipilimumab, and radiotherapy in patients with advanced melanoma. Clin. Cancer Res26, 3193–3201 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR15] 15.Weichselbaum, R. R., Liang, H., Deng, L. & Fu, Y. X. Radiotherapy and immunotherapy: a beneficial liaison?. Nat. Rev. Clin. Oncol.14, 365–379 (2017). [DOI] [PubMed] [Google Scholar]

[CR16] 16.Vanpouille-Box, C. et al. DNA exonuclease Trex1 regulates radiotherapy-induced tumour immunogenicity. Nat. Commun.8, 15618 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR17] 17.Muroyama, Y. et al. Stereotactic radiotherapy increases functionally suppressive regulatory T cells in the tumor microenvironment. Cancer Immunol. Res5, 992–1004 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR18] 18.Li, M. et al. PREX2 contributes to radiation resistance by inhibiting radiotherapy-induced tumor immunogenicity via cGAS/STING/IFNs pathway in colorectal cancer. BMC Med22, 154 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR19] 19.Certo, M., Tsai, C. H., Pucino, V., Ho, P. C. & Mauro, C. Lactate modulation of immune responses in inflammatory versus tumour microenvironments. Nat. Rev. Immunol.21, 151–161 (2021). [DOI] [PubMed] [Google Scholar]

[CR20] 20.Watson, M. J. et al. Metabolic support of tumour-infiltrating regulatory T cells by lactic acid. Nature591, 645–651 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR21] 21.Noe, J. T. et al. Lactate supports a metabolic-epigenetic link in macrophage polarization. Sci. Adv.7, eabi8602 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR22] 22.He, J. et al. The lactate receptor HCAR1 drives the recruitment of immunosuppressive PMN-MDSCs in colorectal cancer. Nat. Immunol. 10.1038/s41590-024-02068-5 (2025). [DOI] [PubMed]

[CR23] 23.Xu, W. et al. N6-methyladenosine modification of 3’tRF-AlaAGC impairs PD-1 blockade efficacy by promoting lactic acid accumulation in the tumor microenvironment of gastric carcinoma. Drug Resist Updat79, 101197 (2024). [DOI] [PubMed] [Google Scholar]

[CR24] 24.Liu, Y. et al. Activation and antitumor immunity of CD8+ T cells are supported by the glucose transporter GLUT10 and disrupted by lactic acid. Sci. Transl. Med16, eadk7399 (2024). [DOI] [PubMed] [Google Scholar]

[CR25] 25.Yang, X. et al. Lactate-modulated immunosuppression of myeloid-derived suppressor cells contributes to the radioresistance of pancreatic cancer. Cancer Immunol. Res8, 1440–1451 (2020). [DOI] [PubMed] [Google Scholar]

[CR26] 26.Kang, E., Kim, H. K., Lee, H. B. & Han, W. Never in mitosis gene A-related kinase-8 promotes proliferation, migration, invasion, and stemness of breast cancer cells via β-catenin signalling activation. Sci. Rep.13, 6829 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR27] 27.Bowers, A. J. & Boylan, J. F. Nek8, a NIMA family kinase member, is overexpressed in primary human breast tumors. Gene328, 135–142 (2004). [DOI] [PubMed] [Google Scholar]

[CR28] 28.Ding, X. F., Chen, J., Zhou, J., Chen, G. & Wu, Y. L. Never-in-mitosis A-related kinase 8, a novel target of von-Hippel-Lindau tumor suppressor protein, promotes gastric cancer cell proliferation. Oncol. Lett.16, 5900–5906 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR29] 29.Wang, M., Yu, K., Meng, F., Wang, H. & Li, Y. NEK8 promotes the progression of gastric cancer by reprogramming asparagine metabolism. Mol. Med31, 3 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR30] 30.Cao, B., Zhang, K., Pan, C., Dong, Y. & Lu, F. NEK8 regulates colorectal cancer progression via phosphorylating MYC. Cell Commun. Signal21, 209 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR31] 31.Cheng, W. C. et al. Uncoupling protein 2 reprograms the tumor microenvironment to support the anti-tumor immune cycle. Nat. Immunol.20, 206–217 (2019). [DOI] [PubMed] [Google Scholar]

[CR32] 32.Mohammadi, H. et al. Using the Radiosensitivity Index (RSI) to predict pelvic failure in endometrial cancer treated with adjuvant radiation therapy. Int J. Radiat. Oncol. Biol. Phys.106, 496–502 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR33] 33.Rodríguez-Ruiz, M. E., Vanpouille-Box, C., Melero, I., Formenti, S. C. & Demaria, S. Immunological mechanisms responsible for radiation-induced abscopal effect. Trends Immunol.39, 644–655 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR34] 34.Oliveira, R. C., Abrantes, A. M., Tralhão, J. G. & Botelho, M. F. The role of mouse models in colorectal cancer research-The need and the importance of the orthotopic models. Anim. Model Exp. Med3, 1–8 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR35] 35.Ho, W. W. et al. Dendritic cell paucity in mismatch repair-proficient colorectal cancer liver metastases limits immune checkpoint blockade efficacy. Proc. Natl. Acad. Sci. USA118, 10.1073/pnas.2105323118 (2021). [DOI] [PMC free article] [PubMed]

[CR36] 36.Halin, C. et al. The S1P-analog FTY720 differentially modulates T-cell homing via HEV: T-cell-expressed S1P1 amplifies integrin activation in peripheral lymph nodes but not in Peyer patches. Blood106, 1314–1322 (2005). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR37] 37.Jia, M. et al. ULK1-mediated metabolic reprogramming regulates Vps34 lipid kinase activity by its lactylation. Sci. Adv.9, eadg4993 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR38] 38.Zhang, D. et al. Metabolic regulation of gene expression by histone lactylation. Nature574, 575–580 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR39] 39.Zhang, C. et al. H3K18 lactylation potentiates immune escape of non-small cell lung cancer. Cancer Res. 10.1158/0008-5472.Can-23-3513 (2024). [DOI] [PubMed]

[CR40] 40.Haddach, M. et al. Discovery of CX-6258. A potent, selective, and orally efficacious pan-pim kinases inhibitor. ACS Med. Chem. Lett.3, 135–139 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR41] 41.Melms, J. C. et al. Inhibition of haspin kinase promotes cell-intrinsic and extrinsic antitumor activity. Cancer Res80, 798–810 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR42] 42.Ogura, A. et al. Neoadjuvant (Chemo)radiotherapy with total mesorectal excision only is not sufficient to prevent lateral local recurrence in enlarged nodes: results of the multicenter lateral node study of patients with low cT3/4 rectal cancer. J. Clin. Oncol.37, 33–43 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR43] 43.Roig, J. NEK8, a NIMA-family protein kinase at the core of the ciliary INV complex. Cell Commun. Signal23, 170 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR44] 44.Ding, X. F., Zhou, J., Hu, Q. Y., Liu, S. C. & Chen, G. The tumor suppressor pVHL down-regulates never-in-mitosis A-related kinase 8 via hypoxia-inducible factors to maintain cilia in human renal cancer cells. J. Biol. Chem.290, 1389–1394 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR45] 45.Panchal, N. K. & Evan Prince, S. The NEK family of serine/threonine kinases as a biomarker for cancer. Clin. Exp. Med23, 17–30 (2023). [DOI] [PubMed] [Google Scholar]

[CR46] 46.Chen, H. et al. NBS1 lactylation is required for efficient DNA repair and chemotherapy resistance. Nature10.1038/s41586-024-07620-9 (2024). [DOI] [PMC free article] [PubMed]

[CR47] 47.Sharma, D., Singh, M. & Rani, R. Role of LDH in tumor glycolysis: Regulation of LDHA by small molecules for cancer therapeutics. Semin Cancer Biol.87, 184–195 (2022). [DOI] [PubMed] [Google Scholar]

[CR48] 48.Peng, M. et al. Aerobic glycolysis promotes T helper 1 cell differentiation through an epigenetic mechanism. Science354, 481–484 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR49] 49.Colegio, O. R. et al. Functional polarization of tumour-associated macrophages by tumour-derived lactic acid. Nature513, 559–563 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR50] 50.Brand, A. et al. LDHA-associated lactic acid production blunts tumor immunosurveillance by T and NK cells. Cell Metab.24, 657–671 (2016). [DOI] [PubMed] [Google Scholar]

[CR51] 51.Attwood, M. M., Fabbro, D., Sokolov, A. V., Knapp, S. & Schiöth, H. B. Trends in kinase drug discovery: targets, indications and inhibitor design. Nat. Rev. Drug Discov.20, 839–861 (2021). [DOI] [PubMed] [Google Scholar]

[CR52] 52.Mace, A. G., Pai, R. K., Stocchi, L. & Kalady, M. F. American Joint Committee on Cancer and College of American Pathologists regression grade: a new prognostic factor in rectal cancer. Dis. Colon Rectum58, 32–44 (2015). [DOI] [PubMed] [Google Scholar]

[CR53] 53.Xu, Y., Zhang, L., Wang, Q. & Zheng, M. Comparison of different colorectal cancer with liver metastases models using six colorectal cancer cell lines. Pathol. Oncol. Res26, 2177–2183 (2020). [DOI] [PubMed] [Google Scholar]

[CR54] 54.Isermann, T. et al. Suppression of HSF1 activity by wildtype p53 creates a driving force for p53 loss-of-heterozygosity. Nat. Commun.12, 4019 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR55] 55.Zeng, Z. et al. Cancer-derived exosomal miR-25-3p promotes pre-metastatic niche formation by inducing vascular permeability and angiogenesis. Nat. Commun.9, 5395 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR56] 56.Czarnecki, P. G. et al. ANKS6 is the critical activator of NEK8 kinase in embryonic situs determination and organ patterning. Nat. Commun.6, 6023 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

NEK8 kinase-mediated lactate increase impairs antitumor immunity decreasing radiotherapy sensitivity in colorectal cancer

Mingzhou Li

Yunfei Ni

Jieqiong Wu

Xin Zou

Yining Chen

Junfeng Qiu

Yifang Li

Huayu Cai

Li Wang

Feifei Wang

Hongxia Zhang

Fangyi Han

Jinghao Huang

Zilong Chen

Bingyu Xu

Li Liang

Abstract

Introduction

Results

CD8+ T cells as the primary mediators of radiotherapy resistance in colorectal cancer

Fig. 1. CD8+ T cells as the primary mediators of radiotherapy resistance in colorectal cancer.

NEK8 is a radiotherapy resistance gene negatively correlated with CD8+ T cell anti-tumor immunity

Fig. 2. NEK8 is a radiotherapy resistance gene negatively correlated with CD8+ T cell anti-tumor immunity.

NEK8 inhibition leads to abscopal effects of radiotherapy in CRC

Fig. 3. RT and NEK8 inhibition combination therapy leads to abscopal effects in CRC.

NEK8 impedes CD8+ T cell migration and suppressive function after radiotherapy

Fig. 4. NEK8 controls CD8+ T cell migration and suppressive function in the context of IR.

NEK8 interacts with LDHA to induce its Ser196 phosphorylation and promotes lactate production

Fig. 5. NEK8 interacts with LDHA and induces its Ser196 phosphorylation to promote lactate production.

Lactate inhibits radiosensitization and abscopal effects mediated by NEK8 knockdown

NEK8-mediated H3K18 lactylation inhibits MHC-I gene transcription in CRC

Fig. 6. NEK8-mediated H3K18 lactylation inhibits MHC-I genes transcription in CRC.

Pharmacological inhibition of NEK8 enhances responses to radiotherapy and abscopal effects

Fig. 7. Pharmacological inhibition of NEK8 enhances responses to radiotherapy and abscopal effects.

Clinical relevance of NEK8 expression and LDHA Ser196 phosphorylation levels in CRC patients

Fig. 8. Clinical relevance of NEK8 expression and LDHA Ser196 phosphorylation levels in CRC patients.

Discussion

Methods

Mice

Clinical specimens

Establishment of radioresistance mouse models and animal studies

Single cell RNA-seq (scRNA) analysis

Cell culture

Plasmid and cell transfections

Flow cytometry

Western blot analysis

RT–qPCR

Multiplex immunofluorescence (mfIHC)

Immunohistochemistry (IHC)

Radiation clonogenic assay

Cell counting kit-8 (CCK-8) assay

Immunofluorescence

T-cell co-culture assay

Co-Immunoprecipitation (Co-IP) assays

Mass spectrometry (MS) analysis

Protein expression and purification

In vitro NEK8 kinase assay

Luciferase-coupled ATP assay

Lactic dehydrogenase (LDH) activity assay

Lactate concentration assay

Surface plasmon resonance (SPR) assay

Cut&Tag and Cut&Tag qPCR

Bioinformatics analysis

RNA sequencing and dataset analysis

High-throughput virtual screening (HTVS)

Quantification and statistical analysis

Reporting summary

Supplementary information

Source data

Acknowledgements

Author contributions

Peer review

Peer review information

Data availability

Competing interests

Footnotes

Supplementary information

References

CD8⁺ T cells as the primary mediators of radiotherapy resistance in colorectal cancer

Fig. 1. CD8⁺ T cells as the primary mediators of radiotherapy resistance in colorectal cancer.

NEK8 is a radiotherapy resistance gene negatively correlated with CD8⁺ T cell anti-tumor immunity

Fig. 2. NEK8 is a radiotherapy resistance gene negatively correlated with CD8⁺ T cell anti-tumor immunity.

NEK8 impedes CD8⁺ T cell migration and suppressive function after radiotherapy

Fig. 4. NEK8 controls CD8⁺ T cell migration and suppressive function in the context of IR.