Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 Feb 4;24(2):230-241.
doi: 10.1158/1535-7163.MCT-23-0522.

Pancreatic CAF-Derived Autotaxin Drives Autocrine CTGF Expression to Modulate Protumorigenic Signaling

Affiliations

Pancreatic CAF-Derived Autotaxin Drives Autocrine CTGF Expression to Modulate Protumorigenic Signaling

Fanny Volat et al. Mol Cancer Ther. .

Abstract

Autotaxin (ATX), encoded by ENPP2, is a clinical target in pancreatic ductal adenocarcinoma (PDAC). ATX catalyzes the production of lysophosphatidic acid (LPA), an important regulator within the tumor microenvironment (TME), yet the protumorigenic action of the ATX/LPA axis in PDAC remains unclear. In this study, by interrogating patient samples and cell line datasets, we show that the PDAC TME, rather than cancer cells, is responsible for the majority of ENPP2 expression and highlight a key role for cancer-associated fibroblast (CAF)-derived ATX in autocrine and paracrine protumorigenic signaling. Using the clinical-stage ATX inhibitor, IOA-289, we identified connective tissue growth factor (CTGF), also known as CCN2, as a downstream mediator of ATX signaling in the PDAC CAF-derived cell line, 0082T. Genetic ablation or pharmacologic inhibition of ATX in 0082T CAFs reduced CTGF secretion via modulation of LPA/LPA receptor signaling. Despite the loss of ATX function, extracellular levels of LPA were paradoxically increased, indicating a role for ATX beyond its enzymatic activity and suggesting a role for its LPA chaperone function in the LPA/LPA receptor signaling in CAFs. As CAFs are the main source for CTGF in the PDAC TME, these findings suggest a role for ATX in promoting a protumorigenic microenvironment via modulation of CAF secretion not only via its LPA-producing activity but also via its LPA chaperone function, providing a potential mechanism for the antitumor effects of ATX inhibition.

PubMed Disclaimer

Conflict of interest statement

Conflicts of interest statement

This study includes results that arise from a collaboration between Cancer Research Horizons (CRH) and iOnctura. RM, FV, AP, LA, ETS, NP, AF, AC, PS, HS and SF are employees of CRH. LZM was a Ph.D. student under the supervision of Michael J. Wakelam and HJS and funded by CRH for this work. MAD, ZJ, GDC, and LV are employees and shareholders of iOnctura.

Figures

Figure 1
Figure 1. Supporting cells of the PDAC TME, such as CAFs, are the main contributors of ATX.
A, ENPP2 mRNA expression (normalised log2 RSEM) in normal tissue compared to primary tumour across several cancer types, displayed in order of difference in ENPP2 expression. RSEM: RNA-Seq by Expectation-Maximization. Data from UCSC Xena TCGA TARGET GTEx (21). Significance determined by two-way ANOVA and defined as a p value <0.05. RSEM: RNA-Seq by Expectation-Maximization. B, Grouping of PDAC TCGA samples by ESTIMATE scores (29). Tumour-rich samples (circles) were defined as lowest quartile immune and lowest quartile stromal and TME-rich samples (triangles) were defined as those with highest quartile stromal and highest quartile immune scores. Other samples were labelled as ‘ungrouped’. C, ENPP2 expression in tumour-rich (n=24) and TME-rich (n=27) samples within the TCGA pancreatic cancer data set. Significance determined by t-test, ‘****’ indicates p value <0.0001. D, ENPP2 expression in cells of TME and tumour cells in the pancreatic scRNA-seq dataset from (26). ENPP2-expressing cells defined as cells with non-zero ENPP2 expression value. Number of cells in TME or tumour compartment highlighted by ‘n’. Cancer abbreviations as follows: ACC-Adrenocortical Carcinoma; BLCA-Bladder Cancer; BRCA-Breast Invasive carcinoma; CESC-Cervical squamous cell carcinoma and endocervical adenocarcinoma; COAD-Colon adenocarcinoma; ESCA-Esophageal carcinoma; GBM-Glioblastoma multiforme; HCC-Hepatocellular Carcinoma; KICH-Kidney Chromophobe; KIRC-Kidney renal clear cell carcinoma; KIRP-Kidney renal papillary cell carcinoma; LGG- Low grade glioma; LUAD-Lung adenocarcinoma; LUSC-Lung squamous cell carcinoma; OV-Ovarian serous cystadenocarcinoma; PDAC-Pancreatic ductal adenocarcinoma; PRAD-Prostate adenocarcinoma; SKCM-Skin Cutaneous Melanoma; THCA-Thyroid carcinoma, UCS-Uterine Carcinosarcoma; UCEC-Uterine Corpus Endometrial Carcinoma.
Figure 2
Figure 2. ATX expression in cancer and fibroblasts cell lines.
A, ATX (˜100 kDa) detection by western blot in the CM generated from the CAF cell line 0082T, NTC (non-targeting control) and ENPP2 KO, and PDAC cell lines PANC-1, MIA PaCa-2, PA-TU-8988T, BxPC-3 and AsPC-1 cultured in serum free (SF) condition for 48 hours. Recombinant Autotaxin (rATX) and SF DMEM are used as controls. Representative western blot of N=3 independent biological replicates. Ponceau staining is shown as a loading control. B, ENPP2 mRNA expression in 0082T CAFs and PDAC cell lines cultured in SF condition for 48 hours. RPL13A was used as housekeeping gene. N=3 biological replicates. C, ATX detection by western blot in the cell lysates from the CAF cell line 0082T and PDAC cell lines MIA PaCa-2, BxPC-3 and Capan-2 cultured in their respective complete media for 48 hours. Recombinant Autotaxin (rATX) and 2% FBS DMEM are used as controls. αTubulin is used as a loading control. D, ATX detection by western blot in CM generated from co-culture of CAF 0082T (NTC or ENPP2 KO) with PDAC cell lines (PANC-1 or MIA PaCa-2) at a 1:1 ratio in SF condition for 48 hours. SF medium and 10% FBS DMEM are used as controls. Representative western blot of N=3 independent biological replicates performed with 3 independent 0082T KO pools. E, ENPP2 expression (log2RPKM – RNA-seq) in cancer cell lines derived from the most common cancer types (ranked by mean expression) and in fibroblasts cell lines. Datasets From the cancer cell line encyclopaedia (CCLE). Cancer cell line abbreviations and numbers : AST – Astrocytoma (n=23); BLCA-Bladder Cancer (n=35); BRCA-Breast Invasive carcinoma (n=53); COAD-Colon adenocarcinoma (n=60); ENDC – Endometrial carcinoma (n=24); ESCA-Esophageal carcinoma (n=25); EWS- Ewing Sarcoma (n=25); GBM-Glioblastoma multiforme (n=34); HCC-Hepatocellular Carcinoma (n=23); HNSC - Head and Neck squamous cell carcinoma (n=14); KRCC – Kidney renal cell carcinoma (n=31); LUAD-Lung adenocarcinoma (n=79); LUSC-Lung squamous cell carcinoma (n=27); NB-Neuroblastoma (n=23); OSCC-Oral squamous cell carcinoma (n=15); OV-Ovarian serous cystadenocarcinoma (n=17); PDAC-Pancreatic ductal adenocarcinoma (n=46); PM – Pleural mesothelioma (n=17); SCLC- Small cell lung cancer (n=51); SKCM-Skin Cutaneous Melanoma (n=57). nd, not detected. KO, ENPP2 knockout using sgRNA-2. NTC, Non-targeting control. L, Ladder. RPKM: Reads Per Kilobase of transcript, per Million mapped reads
Figure 3
Figure 3. ATX inhibition modulates CTGF expression in PDAC 0082T CAFs via LPAR signaling.
A, LPAR1-6 expression in cells of TME and tumour cells in the pancreatic scRNA-seq dataset from (26). B, LPA receptors expression in transcript per million (TPM) in 0082T cells treated with 0.1% DMSO for 24 hours (N=4). C, Venn diagram from 0082T RNA-seq data showing overlap of significant differentially expressed genes (adjusted p value by Wald test of <0.05) between 0.1% DMSO vs 12 μmol/Lol/L IOA-289 and 0.1% DMSO vs 12 μmol/Lol/L PF-8380 treatments. D-E, Relative mRNA expression of PLIN2 (D) and CTFG (E) in 0082T cells treated with 0.1% DMSO (vehicle control), 12 μmol/L IOA-289 or 12 μmol/L PF-8380 for 48 hours (N=3). RPL13A was used as housekeeping gene. Significance determined by one-way ANOVA and post-hoc Dunnett comparisons, “*” indicates p value <0.05. F, CTGF secretion by 0082T cells treated with 0.1% DMSO, 12 μmol/Lol/L IOA-289, 12 μmol/Lol/L PF-8380, or 12 μmol/Lol/L LPAR1-3 antagonist Ki16425 for 48 hours, measured by ELISA (N=3). Significance determined by one-way ANOVA and post-hoc Dunnett comparisons, “****” indicates p value <0.0001. G, CTGF expression in cells of TME and tumour cells in the pancreatic scRNA-seq dataset from (26).
Figure 4
Figure 4. ATX inhibition modulates LPA/LPC levels in conditioned media (CM) of 0082T CAFs.
A, Total LPA concentration in the CM of 0082T and PANC-1 cells treated with 0.1% DMSO, and control after 48 hours conditioning and with or without CFI (N=3). B, Total LPA concentration in the CM of 0082T cells and PANC-1 cells treated with 0.1% DMSO or ATX inhibitors 12 μmol/Lol/L IOA-289 or 12 μmol/Lol/L PF-8380 for 48 hours conditioning with CFI (N=3). Significance determined by two-way ANOVA and post-hoc Dunnett comparisons. ‘ns’ denotes non-significance and ‘*’ indicates p value <0.05. C, Heatmap showing concentration of individual LPA species in the CM of 0082T and PANC-1 cells, and in control after 48 hours of conditioning, with (+) and without (-) CFI when treated with 0.1% DMSO, 12 μmol/Lol/L IOA-289 or 12 μmol/Lol/L PF-8380 (N=3). D-E-F, Concentration of 18:1 LPA (D), 18:2 LPA (E) and 22:6 LPA (F) in the CM of 0082T and PANC-1 cells after treatment with 0.1% DMSO or ATX inhibitors 12 μmol/Lol/L IOA-289 or 12 μmol/Lol/L PF-8380 for 48 hours of conditioning and with CFI (N=3). Significance determined by two-way ANOVA and post-hoc Dunnett comparisons. ‘ns’ denotes non-significance. ‘*’ and ‘**’ indicates p values <0.05, and <0.01 respectively. G-H-I, Concentration of 18:1 LPC (G), 18:2 LPC (H) and 22:6 LPC (I) in the CM of 0082T and PANC-1 cells, and in control after treatment with 0.1% DMSO or ATX inhibitors 12 μmol/Lol/L IOA-289 or 12 μmol/Lol/L PF-8380 for 48 hours of conditioning and with CFI (N=3). Significance determined by two-way ANOVA and post-hoc Dunnett comparisons. ‘ns’ denotes non-significance. ‘*’ and ‘**’ indicates p value <0.05, and <0.01 respectively.
Figure 5
Figure 5. ATX genetic depletion in 0082T CAFs mimics the effects of IOA-289 inhibitor.
A, Validation of ATX depletion by western blot in CM generated from 4 independent 0082T ENPP2 KO pools compared to their corresponding 0082T non-targeting controls (NTC). B, Gene expression profile in 0082T NTC and 0082T ENPP2 KO after 48-hour treatment with 0.1% DMSO (vehicle control), 12 μmol/L IOA-289, relative to vehicle control NTC (N=4). RPL13A was used as housekeeping gene. C, CTGF mRNA expression in 0082T NTC and corresponding ENPP2 KO (N=4) after 48-hour treatment with 0.1% DMSO (vehicle control) or 12 μmol/L IOA-289, relative to vehicle control NTC. RPL13A was used as housekeeping gene. Mean ± SEM. Significance determined by a two-way ANOVA and post-hoc Fischer comparisons and ‘***’, and ‘****’ indicates p value =0.0001 and <0.0001, respectively. D-E-F, Concentration of 18:1 LPA (D), 18:2 LPA (E) and 22:6 LPA (F) in the CM of 0082T NTC and 0082T ENPP2 KO cells after treatment with 0.1% DMSO or 12 μmol/Lol/L IOA-289 for 48 hours of conditioning and CFI (N=4). Mean ± SEM. Significance determined by a two-way ANOVA, with ‘*’, ‘***’, indicates p value <0.05 and <0.0005, respectively. G-H, Relative CTGF mRNA expression (G) and protein secretion (H) in NTC 0082T (N=3) and corresponding ENPP2 KO (N=3 KO using sgRNA-2 and N=2 using sgRNA-1) cells after treatment with 0.1% DMSO (vehicle control), 12 μmol/L IOA-289 or 12 μmol/L Ki16425 for 48 hours. GAPDH and HPRT1 were used as housekeeping genes. Mean ± SEM. Significance determined by a two-way ANOVA and post-hoc Turkey comparisons. ‘*’, ‘**’, ‘***’, and ‘****’ indicates p value <0.05, <0.005, <0.0005, and <0.0001 respectively.

References

    1. Tokumura A, Majima E, Kariya Y, Tominaga K, Kogure K, Yasuda K, et al. Identification of Human Plasma Lysophospholipase, D, a Lysophosphatidic Acid-producing Enzyme, as Autotaxin, a Multifunctional Phosphodiesterase*. J Biol Chem. 2002 Oct 18;277(42):39436–42. - PubMed
    1. Chen J, Li H, Xu W, Guo X. Evaluation of serum ATX and LPA as potential diagnostic biomarkers in patients with pancreatic cancer. BMC Gastroenterol. 2021 Feb 10;21:58. doi: 10.1186/s12876-021-01635-6. - DOI - PMC - PubMed
    1. Kondo M, Ishizawa T, Enooku K, Tokuhara Y, Ohkawa R, Uranbileg B, et al. Increased serum autotaxin levels in hepatocellular carcinoma patients were caused by background liver fibrosis but not by carcinoma. Clin Chim Acta. 2014 Jun 10;433:128–34. - PubMed
    1. Shao Y, Yu Y, He Y, Chen Q, Liu H. Serum ATX as a novel biomarker for breast cancer. Medicine (Baltimore) 2019 Mar 15;98(13):e14973. doi: 10.1097/MD.0000000000014973. - DOI - PMC - PubMed
    1. Vít O, Petrák J. Autotaxin and Lysophosphatidic Acid Signalling: the Pleiotropic Regulatory Network in Cancer. Folia Biol (Praha) 2024 Apr 7;69(5–6):149–62. - PubMed

MeSH terms

Substances