Histone lactylation enhances GCLC expression and thus promotes chemoresistance of colorectal cancer stem cells through inhibiting ferroptosis

Abstract

Colorectal cancer stem cells (CCSCs) play a critical role in mediating chemoresistance. Lactylation is a post-translational modification induced by lactate that regulates gene expression. However, whether lactylation affects the chemoresistance of CCSCs remains unknown. Here, we demonstrate that histone lactylation enhances CCSC chemoresistance both in vitro and in vivo. Furthermore, our findings showed that p300 catalyzes the lactylation of histone H4 at K12, whereas HDAC1 facilitates its delactylation in CCSCs. Notably, lactylation at H4K12 (H4K12la) upregulates GCLC expression and inhibits ferroptosis in CCSCs, and the inhibition of p300 or LDHA decreases H4K12la levels, thereby increasing the chemosensitivity of CCSCs. Additionally, the GCLC inhibitor BSO promotes ferroptosis and sensitizes CCSCs to oxaliplatin. Taken together, these findings suggest that histone lactylation upregulates GCLC to inhibit ferroptosis signaling, thus enhancing CCSC chemoresistance. These findings provide new insights into the relationship between cellular metabolism and chemoresistance and suggest potential therapeutic strategies targeting p300, LDHA, and GCLC. We showed that histones H4K12 lactylation promoted chemoresistance in CSCs. p300 catalyzes the lactylation of histone H4 at K12, HDAC1 inhibits the histone lactylation at the same site. H4K12la in CSCs regulates the expression of the ferroptosis-related gene GCLC, thereby inhibiting ferroptosis and leading to chemoresistance. Targeting the p300, LDHA, or GCLC may be overcome tumor chemoresistance.

We showed that histones H4K12 lactylation promoted chemoresistance in CSCs. p300 catalyzes the lactylation of histone H4 at K12, HDAC1 inhibits the histone lactylation at the same site. H4K12la in CSCs regulates the expression of the ferroptosis-related gene GCLC, thereby inhibiting ferroptosis and leading to chemoresistance. Targeting the p300, LDHA, or GCLC may be overcome tumor chemoresistance.

Fig. 1. High levels of lactylation at histone are associated with an unfavorable prognosis in patients with colorectal cancer and increased histone lactylation in CCSCs.

A Immunohistochemical (IHC) staining was used to test lactylation levels in both normal and primary CRC tumor tissues. Scale bar: 50 μm. B Statistical results of IHC scores for lactylation levels in normal and primary CRC tissues. C Kaplan-Meier curves of overall survival indicate a significant difference between patients with colorectal cancer expressing low and high lactylation levels. High: IHC score >7, n = 39; Low: IHC score ≤7, n = 49; p = 0.008. D Lactylation levels in primary colorectal cancer tissues at different T stages. 11 cases were in the T_1-2 stage, 49 were in the T₃ stage, and 28 were in the T₄ stage. E Lactylation levels in primary colorectal cancer tissues at different N stages: 60 cases in the N₀ stage and 28 cases in the N_1-2 stage. F Lactation levels in primary colorectal cancer tissues at different TNM stages. 10 patients had I stage, 48 had II stage, and 30 had stages III and IV. G western blotting was used to evaluate the expression of Pan Kla in tumor and normal specimens; T, tumor; N, normal. H The UMAP plot of clustered single-cell RNA-seq datasets demonstrates the existence of distinct CRC subpopulations. The cells were clustered into cancer stem cells and other cancer cells, with each cell cluster labeled and colored according to its subcell type. Density distribution of the lactylation score in tumor subpopulations. The lactylation score signatures were calculated from the average relative expression level of a key lactylation-related gene. I Western blot analysis was used to evaluate the expression of CD133, Pan Kla, Nanog, and OCT4 in CRC sphere and adherent cells. J Western blot analysis was performed to evaluate the expression of H4K5la, H4K8la, H4K12la, and H4K16la in CRC spheres and adherent cells. Three biological replicates are shown. The data represent the mean ± SD. In (B) and (E), comparisons were made using unpaired Student’s t-test, whereas in (D) and (F), comparisons were made using One-way ANOVA with Tukey’s test. * p < 0.05, **p < 0.01.

Fig. 2. Lactate regulates histone lactylation and stemness of cancer stem cells.

A Various metabolic modulators regulate glycolysis and lactate production. B Western blot analysis was performed to test the expression of CD133, Pan Kla, Nanog, and H4K12la in LoVo CSCs cultured in varying lactate concentrations for 24 h. C Western blot analysis was performed to evaluate the expression of CD133, Pan Kla, Nanog, and H4K12la in LoVo CSCs cultured in varying NALA concentrations for 24 h. D Western blot analysis was used to evaluate the expression of CD133, Nanog, Pan Kla, and H4K12la in LoVo CSCs cultured in varying glucose concentrations for 24 h; Right: Intracellular lactate levels were measured. E LoVo CSCs were exposed to varying concentrations of 2-DG for 24 h. Left: Western blot analysis was used to assess the expression of CD133, Nanog, Pan Kla, and H4K12la; Right: Intracellular lactate levels were measured. F LoVo CSCs were cultured under hypoxic conditions for different durations. Left: Western blot analysis was used to evaluate the expression of CD133, Nanog, Pan Kla, and H4K12la; Right: Intracellular lactate levels were measured. G Western blot analysis was performed to test the expression of LDHA and LDHB in CRC sphere cells and adherent cells. H Two short hairpin RNAs of LDHA were used to knock down the expression of LDHA in LoVo CSCs. Left: Western blot analysis was used to evaluate the expression of LDHA, CD133, Nanog, Pan Kla, and H4K12la; Right: Intracellular lactate levels were measured. I Two short hairpin RNAs of LDHA were used to knock down the expression of LDHA in SW620 CSCs. Left: Western blot analysis was used to assess the expression of LDHA, CD133, Nanog, Pan Kla, and H4K12la; Right: Intracellular lactate levels were measured. J SW620 CSCs were exposed to varying concentrations of oxamate for 24 h. Left: Western blot analysis was used to evaluate the expression of CD133, Nanog, Pan Kla, and H4K12la; Right: Intracellular lactate levels were measured. K LoVo CSCs were exposed to varying concentrations of oxamate for 24 h. Left: Western blot analysis was used to evaluate the expression of CD133, Nanog, Pan Kla, and H4K12la; Right: Intracellular lactate levels. Three biological replicates were shown. Comparisons were conducted using one-way ANOVA with Tukey’s test. *p < 0.05, **p < 0.01, ***p < 0.001.

Fig. 3. H4K12 is lactylated by p300 and de-lactylated by HDAC1.

A Western blot analysis was used to evaluate the expression of p300, CBP, Pan Kla, and H4K12la in LoVo CSCs transfected with the Vector, HA-CBP, Flag-p300. B, C Western blot analysis was performed to assess the expression of CD133, Nanog, Pan Kla, H4K12la, and Pan Ac in LoVo and SW-620 CSCs cultured for 24 h with different concentrations of p300 inhibitor. D Western blot analysis was performed to evaluate the expression of p300, CD133, Nanog, Pan Kla, H4K12la, and Pan Ac in siP300 LoVo CSCs. E Immunofluorescence (IF) assay was performed to assess the expression of H4K12la and Pan Ac in response to p300 Scale bar: 200 μm. F Western blot analysis was used to evaluate the expression of the indicated molecules in shHDAC1 LoVo CSCs. G IF assay was performed to evaluate the expression of H4K12la and Pan Ac upon HDAC1 knockdown. Scale bar: 100 μm. H Western blot analysis was performed to evaluate the expression of the indicated molecules in LoVo CSCs overexpressing HDAC1. I IF assay of HCT-116 and LoVo sphere cells overexpressed HDAC1-GFP to assess H4K12la levels. The arrowhead indicates the cells overexpressing HDAC1 and under expressing H4K12la. Scale bar: 100 μm.

Fig. 4. Inhibition of histone lactylation sensitizes CCSCs to oxaliplatin.

A Cell survival was evaluated in LoVo and SW-620 CSCs treated with NALA plus the indicated doses of oxaliplatin for 24 h, vehicle treatment as negative control. B Cell survival was assessed in LoVo and SW-620 CSCs treated with lactate plus oxaliplatin for 24 h, with vehicle treatment as negative control. C Cell survival was tested in HCT-116 and LoVo CSCs treated with 2-DG plus oxaliplatin for 24 h, with vehicle treatment as negative control. D Cell survival was assessed in LoVo and SW-620 CSCs treated with oxamate plus oxaliplatin for 24 h. E–G Nude mice were injected subcutaneously with LoVo CSCs and then received the injection oxaliplatin (Oxa) at 10 mg/kg twice a week, and NALA at 120 mg/kg once a day, with vehicle injection as negative control. Tumor volumes were measured every 3 days (n = 5), growth curves were plotted (E), harvested tumors were photographed (F), and tumor weights (G) were measured (n = 5). H, I Cell survival was tested in SW-620 and LoVo CSCs treated with the indicated treatments and oxaliplatin for 24 h. J–L Nude mice were injected subcutaneously with LoVo CSCs and then treated with oxaliplatin (Oxa) at 10 mg/kg twice a week, along with either LDHi at 1000 mg/kg or p300i at 15 mg/kg every two days, respectively. Tumor volumes were measured every 3 days (n = 5) (J), harvested tumors were photographed (K), and weights were measured (n = 5) (L). Three biological replicates were shown. The data shown represent the mean ± SD. Statistical analyses were performed using the Student’s t-test. *p < 0.05; **p < 0.01; ***p < 0.001; ns stands for no significant differences.

Fig. 5. Lactylation of histones inhibits ferroptosis in CCSCs.

A Volcano plots displaying differential RNA expression in LoVo CSCs treated with oxaliplatin or oxaliplatin plus lactate. Genes that were identified as differentially expressed by DESeq2 (Wald test) are shown in red and blue, with 278 upregulated and 100 downregulated genes. Oxa: oxaliplatin; Lac: lactate. B Gene Set Enrichment analysis (GSEA) of bulk RNA-seq was performed using the dataset. Representative GSEA plots indicate the enrichment of ferroptosis-related genes in the WikiPathways (WP) cancer gene sets (NES = 1.545, p < 0.0). NES, normalized enrichment score; p-value, normalized p-value. C, D Relative lipid ROS and malondialdehyde (MDA) levels were assayed in the LoVo CSCs treated with RSL3 (5 μM) or erastin (5 μM) or oxaliplatin(5 μM) for 12 h (n = 3). E, F Relative levels of lipid ROS and MDA were measured in LoVo CSCs treated with oxaliplatin(5 μM) and lactate for 12 h (n = 3). G, H Relative levels of lipid ROS and MDA were measured in LoVo CSCs treated with oxaliplatin(5 μM) and 2-DG for 12 h (n = 3). I, J Relative lipid ROS and MDA levels were measured in shNC and shLDHA LoVo CSCs treated with oxaliplatin (5 μM) for 12 h (n = 3). K, L Relative lipid ROS and MDA levels were measured in siNC and siP300 LoVo CSCs treated with oxaliplatin (5 μM) for 12 h (n = 3). M, N Relative lipid ROS and MDA levels were measured in shNC and shHDAC1 LoVo CSCs treated with oxaliplatin (5 μM) for 12 h (n = 3). Three biological replicates were shown. The presented data show the mean ± SD. Comparisons were conducted using one-way ANOVA with Tukey’s test. *p < 0.05, **p < 0.01, ***p < 0.001.

Fig. 6. Histone lactyation transcriptionally activates expression of GCLC.

A Distribution of H4K12la sites relative to the translation start site (TSS). B KEGG pathway analysis of H4K12la peaks. C Bioinformatic analysis revealed that GCLC is a downstream target of H4K12la. D IGV tracks for GCLC were generated using ChIP-seq analysis. Sites a-f are distributed across the GCLC genomic region, with sites b and c corresponding to H4K12la peaks. E CUT&Tag-qPCR assay was used to analyze the status of H4K12la in the GCLC genomic region in LoVo CSCs. F Western blot was performed to evaluate the expression of GCLC, Pan-Kla, H4K12la, and GPX4 in LoVo CSCs treated with NALA (10 μM) or Ferr-1(1 μM) plus oxaliplatin (Oxa). G, H Relative lipid ROS and MDA levels were measured in LoVo CSC streated with NALA (10 μM) or Ferr-1(1 μM) plus oxaliplatin (Oxa) for 12 h. I Western blotting was performed to evaluate the expression of GCLC, Pan Kla, H4K12la, and GPX4 in LoVo CSCs treated with 2-DG (10 μM) or Ferr-1(1 μM) plus oxaliplatin (Oxa). J, K Relative lipid ROS and MDA levels were measured in LoVo CSCs treated with 2-DG (10 μM) or Ferr-1(1 μM) plus oxaliplatin (Oxa) for 12 h. Three biological replicates were shown. The presented data show the mean ± SD. Comparisons were conducted using one-way ANOVA with Tukey’s test. *p < 0.05, **p < 0.01, ***p < 0.001.

Fig. 7. Inhibition of GCLC improves sensitivity to chemotherapy in CCSCs.

A Western blot analysis was performed to test the expression of GCLC, Pan Kla, H4K12la, and GPX4 in LoVo CSCs overexpressing GCLC and treated with or without oxamate (10 mM). B, C Relative lipid ROS and MDA levels were measured in LoVo CSCs overexpressing GCLC, and treated with or without oxamate (10 mM). Cells were treated with oxaliplatin (5 μM) for 12 h to induce ferroptosis. D Western blot was performed to evaluate the expression of GCLC, Pan Kla, H4K12la, and GPX4 in LoVo CSCs treated with NALA and/or BSO. E, F Relative lipid ROS and MDA levels were measured in LoVo CSCs treated with oxaliplatin (5 μM) for 12 h. G Western blot analysis was performed to test the expression of GCLC, Pan-Kla, H4K12la, and GPX4 in LoVo and SW-620 CSCs; S, sensitive; R, resistant. H–J Nude mice were injected subcutaneously with LoVo CSCs and then received injections of oxaliplatin at 10 mg/kg twice a week, BSO at 300 mg/kg once a day; Tumor volumes were measured every 3 days (n = 5) and growth curves were plotted (H); Harvested tumors were photographed (I); and weights (J) were measured (n = 5). K Schematic representation of the proposed model. Three biological replicates were shown. Data are presented as mean ± standard deviation. Comparisons were conducted using one-way ANOVA with Tukey’s test, and statistical analyses were performed using Student’s t-test. *p < 0.05, **p < 0.01, ***p < 0.001, ns indicates no significant difference.