Positive Feedback Loop of Histone Lactylation-Driven HNRNPC Promotes Autophagy to Confer Pancreatic Ductal Adenocarcinoma Gemcitabine Resistance

Abstract

Gemcitabine resistance remains a primary determinant of poor survival outcomes in pancreatic ductal adenocarcinoma (PDAC) patients, underscoring the urgent need to elucidate its molecular mechanisms and develop effective countermeasures. Here, gemcitabine-resistant pancreatic cancer cell lines and patient-derived xenograft (PDX) models are established, followed by high-throughput sequencing, which identified heterogeneous nuclear ribonucleoprotein C (HNRNPC) as a significantly upregulated factor in chemoresistant tumors. Silencing of HNRNPC expression substantially restores sensitivity to gemcitabine treatment in vitro and vivo. Mechanistically, multi-omics analysis reveals that histone H3 lysine18 lactylation (H3K18la) drives HNRNPC overexpression. HNRNPC stabilizes TNF receptor-associated factor 6 (TRAF6) transcripts in an N6-methyladenosine(m6A) -dependent manner, thereby activating autophagy to mediate gemcitabine resistance. Concurrently, HNRNPC orchestrates a metabolic reprogramming cascade by similarly stabilizing aldehyde dehydrogenase 1 family member A3 (ALDH1A3) mRNA, which enhances glycolysis and H3K18la levels, establishing a self-reinforcing histone lactylation-HNRNPC positive feedback loop. Notably, pharmacological inhibition of ALDH1A3 using 673A effectively disrupted this regulatory circuit and exerts a synergistic effect with gemcitabine in PDX. These findings not only delineate a histone lactylation-driven positive feedback loop sustaining chemoresistance through HNRNPC-mediated autophagy activation, but also develop the potential of 673A as a promising clinical candidate for overcoming gemcitabine resistance in PDAC treatment.

Keywords: HNRNPC; N6‐methyladenosine; autophagy; chemoresistance; histone lactylation.

Figure 1

HNRNPC is upregulated in gemcitabine‐resistant PDAC and correlates with poor gemcitabine therapy response. A) Schematic diagram of the gemcitabine‐resistant (GR) PDAC PDX models establishment and the RNA sequencing of PDAC patients and PDX. B) Schematic diagram of the induction and RNA sequencing of gemcitabine‐resistant PDAC cell lines. C) A Venn diagram intersecting with multiple datasets demonstrated the potential genes related to gemcitabine‐resistant PDAC. D) Kaplan–Meier survival curves revealed the correlation between HNRNPC expression and overall survival of gemcitabine‐treated PDAC patients from the TCGA dataset. E) Representative images of IHC staining for HNRNPC in different passages of PDX that underwent PBS or gemcitabine. IHC scores were calculated by multiplying the staining intensity grade (negative, 0; mild, 1; moderate, 2; severe, 3) by the percentage of positively stained tumor area (≤ 25%, 1; >25% and ≤ 50%, 2; >50% and ≤ 75%, 3; >75%, 4). F) The statistical analysis of IHC staining for HNRNPC scores in different passages of PDX that underwent PBS or gemcitabine. ^* p < 0.05 according to Student's t‐test. G) Representative images of IF staining for HNRNPC in gemcitabine‐sensitive or resistant PDAC patients. IF scores were measured by fluorescence intensity with ImageJ. H) Representative images of IHC staining for HNRNPC in gemcitabine‐sensitive or resistant PDAC patients. IHC scores were calculated by multiplying the staining intensity grade (negative, 0; mild, 1; moderate, 2; severe, 3) by the percentage of positively stained tumor area (≤ 25%, 1; > 25% and ≤ 50%, 2; >50% and ≤ 75%, 3; >75%, 4). I) The statistical analysis of IHC staining for HNRNPC scores in gemcitabine‐sensitive or resistant PDAC patients is shown. ^* p < 0.05 according to Student's t‐test. J) The correlation between HNRNPC expression and response to gemcitabine was analyzed using Fisher's exact test. K) The expression of HNRNPC in two PDX models was detected by qPCR, and the tumor images of each corresponding group under the treatment of PBS (sample size, n = 5) or gemcitabine (sample size, n = 5), with the tumor growth inhibition ratio displayed. ^* p < 0.05; ^**** p < 0.0001 according to Student's t‐test. The data are shown as mean ± SD.

Figure 2

HNRNPC expression is activated by histone lactylation. A) Diagram of the glycolysis pathway and the regulation. B) Heatmap in the TCGA dataset depicted the correlation between HNRNPC and metabolic enzymes of the glycolysis pathway in PDAC patients. C) Scatter plots showed the correlation between HNRNPC and HK1/LDHA, analyzed with the TCGA dataset. D) Representative images of IHC staining for HNRNPC/H3K18la in gemcitabine‐sensitive or resistant PDAC patients. IHC scores were calculated by multiplying the staining intensity grade (negative, 0; mild, 1; moderate, 2; severe, 3) by the percentage of positively stained tumor area (≤ 25%, 1; > 25% and ≤ 50%, 2; >50% and ≤ 75%, 3; >75%, 4). E) IHC scores staining for HNRNPC/H3K18la in gemcitabine‐sensitive or resistant PDAC patients. ^* p < 0.05 according to Student's t‐test. F) Scatter plots showed the correlation between HNRNPC and H3K18la IHC scores. G) H3K18la enrichment peaks on the HNRNPC promoter region in PDAC cells (GS) identified by CUT&Tag. H) HNRNPC mRNA expression level evaluated with qPCR under 2DG/Oxamate treatment in PDAC cells (GS). ^* p < 0.05; ^** p < 0.01; ^*** p < 0.001; ^**** p < 0.0001 according to Student's t‐test. I) HNRNPC mRNA expression level evaluated with qPCR under Nala treatment in PDAC cells (GS). ^* p < 0.05; ^** p < 0.01; ^*** p < 0.001; ^**** p < 0.0001 according to Student's t‐test. J) HNRNPC and H3K18la levels identified with western blot subjected to Nala. K) HNRNPC and H3K18la levels identified with western blot subjected to 2DG/Oxamate. L) ChIP‐qPCR analysis of H3K18la enrichment on the HNRNPC promoter region applied with vehicle/2‐DG/Oxamate/Nala in PDAC cells (GS). ^* p < 0.05; ^** p < 0.01; ^*** p < 0.001; ^**** p < 0.0001 according to Student's t‐test. The data are shown as mean ± SD. Abbreviation: GS, gemcitabine‐sensitive; GR, gemcitabine‐resistant.

Figure 3

HNRNPC knockdown enhances gemcitabine efficacy against PDAC in vitro and in vivo. A) HNRNPC expression after HNRNPC knockdown in PDAC cells detected by qPCR. ^* p < 0.05; ^** p < 0.01; ^*** p < 0.001; ^**** p < 0.0001 according to Student's t‐test. B) The IC50 value following HNRNPC knockdown in PDAC cells. C) Apoptosis assay in PDAC cells treated with 1 µM gemcitabine. D) Tumor spheres formation of PDAC cells under the treatment of 1 µM gemcitabine. E) CDX tumors from BxPC‐3(GR) (sample size, n = 5 each group). F) Tumor growth curves of each group. ^*** p < 0.001 according to Student's t‐test. G) Tumor weight of each group. ^**** p < 0.0001 according to Student's t‐test. H) The final weight of mice in each group was normalized to the control group. ns (no significance) according to Student's t‐test. I) Representative images of IHC staining for Ki67 and statistical analysis in tumors from each group. ^*** p < 0.001 according to Student's t‐test. J) Representative images of TUNEL and statistical analysis in tumors with different treatments. ^* p < 0.05 according to Student's t‐test. The data are shown as mean ± SD.

Figure 4

HNRNPC activates autophagy by stabilizing TRAF6 mRNA through m6A‐dependent mechanisms in PDAC. A) Heatmap of PDAC cells RNA sequencing. B) GO enrichment analysis of PDAC cells RNA sequencing differential expression genes. C) Western blot assessed the autophagy markers level in PDAC cells treated with gemcitabine at 1 µM. D) Representative images of autophagy formation (shown as red/yellow dots) documented by mRFP‐GFP‐LC3 puncta assay and statistical analysis in PDAC cells applied with 1 µM gemcitabine. ^* p < 0.05; ^** p < 0.01; ^*** p < 0.001; ^**** p < 0.0001 according to Student's t‐test. E) Representative transmission electron microscopy (TEM) images of autophagy level (indicated with arrows) in PDAC cells in the presence of 1 µM gemcitabine. F) A Venn diagram intersecting with multiple datasets implied TRAF6 as the target of HNRNPC to regulate autophagy in PDAC. G) RIP‐sequencing discovered the m6A modification and HNRNPC binding on TRAF6 mRNA. H) RIP‐qPCR detected TRAF6 mRNA m6A modification and HNRNPC binding. ^* p < 0.05; ^** p < 0.01; ^*** p < 0.001; ^**** p < 0.0001 according to Student's t‐test. I) qPCR analysis of TRAF6 mRNA levels after actinomycin D treatment in PDAC cells. ^** p < 0.01; ^*** p < 0.001 according to Student's t‐test. The data are shown as mean ± SD.

Figure 5

TRAF6 is responsible for HNRNPC‐mediated gemcitabine resistance in PDAC. A) TRAF6 expression after TRAF6 knockdown in PDAC cells was verified by qPCR. ^* p < 0.05; ^** p < 0.01; ^*** p < 0.001; ^**** p < 0.0001 according to Student's t‐test. B) The IC50 value following TRAF6 knockdown in PDAC cells. C) TRAF6 mRNA level confirmed by qPCR in PDAC cells, as TRAF6 is overexpressed. ^*** p < 0.001; ^**** p < 0.0001 according to Student's t‐test. D) The IC50 value along with TRAF6 overexpression in PDAC cells. E) Representative images of the Edu assay evaluating proliferation in PDAC cells in the presence of 1 µM gemcitabine. F) Western blot assessed autophagy marker levels in PDAC cells treated with gemcitabine at 1 µM. G) Representative images of autophagy formation (shown as red/yellow dots) monitored by mRFP‐GFP‐LC3 puncta assay along with statistical analysis in PDAC cells applied with 1 µM gemcitabine. ^* p < 0.05; ^** p < 0.01; ^*** p < 0.001; ^**** p < 0.0001 according to Student's t‐test. H) Representative transmission electron microscopy (TEM) images of autophagy level (indicated with arrows) in PDAC cells treated with 1 µM gemcitabine. The data are shown as mean ± SD.

Figure 6

HNRNPC maintains ALDH1A3 mRNA stability, reinforcing H3K18la to establish a positive feedback loop. A) Venn diagram intersection of HNRNPC knockdown (KD), m6A‐RIP, and HNRNPC‐RIP sequencing data indicated ALDH1A3 as the target of HNRNPC. B) qPCR quantified ALDH1A3 mRNA expression in HNRNPC knockdown PDAC cells. ^* p < 0.05; ^** p < 0.01; ^*** p < 0.001; ^**** p < 0.0001 according to Student's t‐test. C) Western blot confirmed ALDH1A3 expression in HNRNPC‐knockdown. D) RIP‐sequencing revealed the m6A modification and HNRNPC binding on ALDH1A3 mRNA. E) RIP‐qPCR detected ALDH1A3 mRNA m6A modification and HNRNPC binding. ^* p < 0.05; ^** p < 0.01; ^*** p < 0.001; ^**** p < 0.0001 according to Student's t‐test. F) qPCR analysis of ALDH1A3 mRNA level after actinomycin D treatment in PDAC cells. ^* p < 0.05; ^** p < 0.01; ^*** p < 0.001; ^**** p < 0.0001 according to Student's t‐test. G) HNRNPC expression level quantified with qPCR in PDAC cells after treatment with 673A. ^* p < 0.05; ^** p < 0.01; ^*** p < 0.001; ^**** p < 0.0001 according to Student's t‐test. H) HNRNPC expression and H3K18la level confirmed by western blot in PDAC cells subjected to 673A. I) ChIP‐qPCR analysis of H3K18la enrichment on the HNRNPC promoter region applied with vehicle or 673A in PDAC cells. ^* p < 0.05; ^** p < 0.01; ^*** p < 0.001; ^**** p < 0.0001 according to Student's t‐test. J) Representative images of mIF staining for ALDH1A3, H3K18la, and HNRNPC in gemcitabine‐sensitive or resistant PDAC patients. IF scores were measured by fluorescence intensity with ImageJ. K) Correlation analysis between ALDH1A3 and H3K18la/HNRNPC in PDAC patients by IHC. The data are shown as mean ± SD.

Figure 7

Pharmaceutical 673A targets the feedback signal to potentiate gemcitabine efficacy against gemcitabine‐resistant PDAC. A) Synergy effect validated by the Loewe model in combination of gemcitabine and 673A in PDAC cells. B) Apoptosis assay in PDAC cells administered at gemcitabine (1 µM), 673A (30 µM), or combination. C) PDX models bearing tumor received vehicle, gemcitabine, 673A, or combination, tumors were displayed (sample size, n = 5 each group). D) Tumor growth curves of each group. ^*** p < 0.001 according to Student's t‐test. E) Tumor weight of each group. ^** p < 0.01; ^*** p < 0.001 according to Student's t‐test. F) The normalized weight of mice in each group. ^* p < 0.05; ^** p < 0.01; ^*** p < 0.001; ^**** p < 0.0001; ns (no significance) according to Student's t‐test. G) Representative images of IHC staining for Ki67 and statistical analysis in PDX tumors from each group. ^** p < 0.01 according to Student's t‐test. H) Representative images of TUNEL and statistical analysis in PDX tumors with different treatments. ^*** p < 0.001 according to Student's t‐test. I) Schematic representation of the function and potential mechanism of HNRNPC in gemcitabine‐resistant PDAC. The data are shown as mean ± SD. ^* p < 0.05; ^** p < 0.01; ^*** p < 0.001 according to Student's t‐test.