Inhibition of autotaxin sensitizes colon cancer to radiation by suppressing LPAR2-AKT survival signaling.

Introduction
Colon cancer represents a major global health challenge, with colon adenocarcinoma constituting a significant proportion of cases. It is estimated that over 1.9 million new cases are diagnosed annually worldwide, and the disease remains a leading cause of cancer-related mortality, largely due to its high metastatic potential [1]. While surgical resection is the cornerstone of curative-intent treatment for localized colon cancer, a subset of patients faces poor outcomes due to local recurrence or metastasis [2]. In contrast to rectal cancer, for which neoadjuvant radiotherapy (RT) is a standard of care, RT plays a much more limited and selective role in the management of colon adenocarcinoma [3, 4]. Its application is generally reserved for specific clinical scenarios, such as palliative relief for symptomatic metastases, management of positive surgical margins, or ablation of isolated oligometastatic lesions [5]. The efficacy of RT in these contexts is often compromised by the development of radioresistance [6]. Therefore, elucidating the molecular mechanisms underlying radioresistance is crucial for enhancing the efficacy of RT in these selected colon cancer patients and improving their prognosis.
The resistance of colon cancer to RT arises from multiple factors. Key contributors include the activation of survival pathways such as PI3K/AKT [7] and MAPK [8, 9], promoting cell proliferation and inhibiting apoptosis. Additionally, cancer stem cells (CSCs) exhibit inherent resistance to radiation [10]. The tumor microenvironment, characterized by hypoxia and supportive stromal cells, further protects cancer cells from radiation damage [11, 12]. Genetic alterations, particularly in tumor suppressor genes such as p53 [13] and oncogenes such as KRAS [14], also contribute to resistance. Furthermore, the expression of DNA repair proteins such as O6-methylguanine-DNA-methyltransferase (MGMT) [15, 16] and RAD51 [17] enhances the capacity of cancer cells to repair radiation-induced DNA damage. Importantly, tumor cells develop radioresistance by secreting bioactive factors such as fatty acids, miRNA and exosomes [18–20], which activate survival pathways through paracrine and autocrine signaling to promote RT resistance. However, the specific effects of RT on factor release and downstream signaling alterations in colon cancer remain poorly understood.
Autotaxin (ATX), a secreted glycoprotein and member of the ENPP family (ectonucleotide pyrophosphatase/phosphodiesterase 2, ENPP2), functions by cleaving lysophospholipids into the lipid mediator LPA through its lysophospholipase D (lysoPLD) activity [21]. LPA is a bioactive lipid that critically regulates diverse cellular processes such as proliferation, migration, and survival [22]. In the context of colon cancer, numerous studies have indicated that ATX and its product, LPA, may contribute to tumor progression and the aggressive behavior of cancer cells [23]. ATX is implicated in inflammatory processes that can further promote cancer progression [24]. Inflammation within the tumor microenvironment can enhance ATX expression, establishing a feedback loop that exacerbates cancer development [25]. Investigating whether ATX influences the sensitivity of colon cancer cells to RT is essential, as this could yield new targets and treatment strategies for overcoming RT resistance.
This study demonstrates that the ATX-LPAR2 axis significantly contributes to tumorigenesis and radioresistance in colon cancer. Elevated ATX expression correlates with poor prognosis and is further increased by radiation, promoting proliferation through LPAR2-AKT signaling. Inhibition or gene knockdown of ATX reduced tumor volume and weight, simultaneously enhancing the efficacy of radiation treatment. These findings highlight ATX’s therapeutic promise in addressing radiation resistance and enhancing clinical efficacy for colon cancer patients.

Materials and methods

Cell culture
The mouse colon cancer cell line MC38 was obtained from the American Type Culture Collection (Manassas, VA, USA). HCT116 and SW480 cells were purchased from Procell Life Science & Technology (Wuhan, China). The cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) (KeyGENBioTECH, Nanjing, China), supplemented with 10% fetal bovine serum and 1% streptomycin/penicillin (Beyotime Biotechnology, Shanghai, China). All cell cultures were incubated at 37 °C in a humidified atmosphere containing 5% CO2.

Radiation for cell
MC38, HCT116, and SW480 cells were seeded in 6-well or 96-well plates at densities of 5 × 10⁵, 4 × 10⁵, 4 × 10⁵ or 4 × 10³, 3 × 10³, 3 × 10³ cells per well, respectively. After 24 h, the adherent cells were exposed to specific doses of inhibitors or irradiated at 2, 4–6 Gy (dose rate: 3 Gy/min) using an X-RAD 225XL irradiator (PXI Company, USA). Following a 48-hour incubation at 37 °C in a humidified 5% CO₂ atmosphere, both the cells and culture supernatants were collected for subsequent analysis.

Lipidomics analysis
MC38 cells were plated in 6-well plates at 5 × 10⁵ cells per well and allowed to adhere for 24 h. Upon reaching adherence, the cells were subjected to irradiation using the protocol described above. After irradiation, we collected the cells and quantified their metabolites via high-performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS) analysis following established methods [26]. The analytical sequence included a pooled quality control (QC) sample derived from all samples, which was injected at regular intervals to monitor instrument performance. The relative standard deviation (RSD) of peak areas for major lipids in the QC samples was below 15%, confirming data reliability. Lipid extraction and analysis were performed using a validated protocol with deuterated internal standards (including d4-LPA 16:0 and d4-LPA 18:1; Avanti Polar Lipids) for accurate quantification of LPA species. Multivariate data analysis was performed using MetaboAnalyst 6.0 [27]. Principal Component Analysis (PCA) was used for unsupervised pattern recognition (Fig. 1a). For biomarker selection, Partial Least Squares-Discriminant Analysis (PLS-DA) was applied. Variable Importance in Projection (VIP) scores from the validated PLS-DA model were used to identify the most significant lipids (Fig. 1b). All detected lipid species were confirmed using authentic standards. Metabolite abundance was normalized to the total cell count.

Quantitative real-time polymerase chain reaction (qRT-PCR)
The cells were treated as described above. Total RNA isolation was carried out with TRIzol reagent (Invitrogen, Carlsbad, CA, USA), adhering to the supplier’s instructions. Next, first-strand RNA was reverse transcribed into complementary DNA (cDNA) using the Reverse Transcription System (Vazyme Biotechnology, Nanjing, China). Quantitative real-time PCR was conducted on an ABI PRISM 7500 Sequence Detection System (New York, USA) to assess ATX and β-actin expression. The thermal cycling conditions comprised an initial denaturation at 95 °C for 30 s, succeeded by 40 cycles of denaturation at 95 °C for 5 s and primer annealing at 60 °C for 34 s. mRNA expression levels were determined relative to β-actin as the internal control. The primer sequences are listed in Supplementary Table S1.

Cell viability assessment
Cell viability was assessed using a Cell Counting Kit-8 (Beyotime Biotechnology, Shanghai, China). Cells were seeded in 96-well plates at a density of 4000 cells per well and subsequently treated with ATX (25, 50 or 100 pg/mL) (human: 93917ES60, mouse: 94546ES60, Yeasen Biotechnology, Shanghai, China) AKT inhibitor MK-2206 (10 nM) (HY-108232, MCE, NJ, USA) or the ATX inhibitor HA130 (50 nM) (HY-19329, MCE, NJ, USA) for 24, 48, and 72 h. Following treatment, 10 µL of CCK-8 solution was added to each well. After 2 h, the absorbance was measured at 450 nm using a microplate reader.

Colony formation assay
For the colony formation assay, cells with or without ATX/LPAR2 knockdown were plated in 6-well plates at a density of 1,000 cells per well. Twenty-four hours thereafter, cultures were treated with inhibitors or irradiated (6 Gy, 3 Gy/min; X-RAD 225XL). The medium was replaced with fresh inhibitor-containing medium every two days. Upon 7 days of incubation, cells were fixed with 4% paraformaldehyde for 20 min and stained with crystal violet for 30 min at room temperature prior to imaging.

TUNEL assay
Tumor cells were seeded in 96-well plates at a density of 6,000 cells per well. After 24 h, the cells were irradiated with a dose of 6 Gy, with or without HA130 (50 nM). After an additional 48 h, apoptosis was assessed using the One Step TUNEL Apoptosis Assay Kit (Beyotime Biotechnology, Shanghai, China) following the manufacturer’s instructions. The fluorescence intensity of TUNEL was quantified using the Cytation 3 Cell Imaging Multi-Mode Reader (BioTek, Winooski, VT) at excitation/emission wavelengths of 493/528 nm.

Western blotting
Cells were lysed using RIPA buffer (NCM Biotech, Suzhou, China) according to the manufacturer’s instructions. The resulting lysates were mixed with loading buffer (Beyotime Biotechnology, Shanghai, China) and heated in boiling water for 10 min. Samples were subsequently loaded onto a 12% bis-tris-acrylamide gel for electrophoresis, followed by the transfer of protein bands onto nitrocellulose membranes (PALL Corporation, Mexico). Subsequently, the membranes were blocked with 5% nonfat milk (VICMED, Xuzhou, China) at room temperature for 1 h. The membranes were then incubated overnight at 4 °C with primary antibodies: anti-ATX (1:1000; sc-374222, Santa Cruz Biotechnology, Dallas, TX, USA), anti-p-AKT (1:1000; ab81283, Abcam, Cambridge, UK), anti-total-AKT (#9272, Cell Signaling Technology, Danvers, MA, USA) anti- LPAR2(A14819, Abclonal, Wuhan, China) and anti-GADPH (1:10,000; A19056, Abclonal, Wuhan, China) anti-β-actin (1:10,000; ab6276, Abcam). Following this, the membranes were washed with TBST buffer. For detection, a horseradish peroxidase (HRP)-conjugated secondary antibody (1:5000; Proteintech, China) was applied to the membranes and incubated at 37 °C for 2 h. Immunopositive bands were visualized using an ECL chemiluminescent detection system (Thermo, Waltham, MA, USA), and the images were transferred to a Tanon imaging system (Shanghai, China). Densitometry analysis was performed using ImageJ Software (Bethesda, MD, USA).

LPAR2 or ATX knockdown
Pseudoviral particles for gene knockdown were produced by co-transfecting 293 T cells with pMD2.G, psPAX2, and plasmids encoding either control shRNA or shRNAs targeting LPAR2 or ATX (sc-39926-SH, sc-44906-SH, Santa Cruz Biotechnology, Inc.). The viral supernatant was harvested at 48 h post-transfection and used to infect HCT116 and SW480 cells. Knockdown efficiency was validated by western blot after a 3-day selection with puromycin.

Enzyme-linked immunosorbent assay (ELISA)
The concentrations of ATX and LPA in cell culture supernatants were measured using ELISA kits (CLOUD-CLONE CORP, Houston, USA), following the manufacturer’s guidelines.
To measure intratumoral cytokine concentrations, we collected tumor specimens and immediately stored them in liquid nitrogen. We quantified LPA levels using ELISA kits.

Mouse model
Male C57BL/6J mice and BALB/c nude mice (7–8 weeks old, 20–22 g) were obtained from the Xuzhou Medical University Animal Center and maintained under specific pathogen-free conditions with a 12/12-hour light/dark cycle and free access to food and water. Animal health was assessed daily.
A sample size of n = 4 per group was determined by power analysis based on pilot data (α = 0.05, β = 0.8). MC38 cells (5 × 10⁵ in 100 µL PBS) were inoculated subcutaneously into the right flank. When the average tumor volume reached approximately 100 mm³, mice were randomly allocated to one of four groups—Control, HA130, RT, or HA130 + RT—using a computer-generated randomization sequence. Investigators performing tumor volume measurements and endpoint analyses were blinded to group assignments throughout the study.
No animals were excluded from the analysis. Humane endpoints were predefined as follows: tumor volume > 1500 mm³, ulceration or infection of the tumor, or body weight loss > 20% with signs of distress.
Radiotherapy was delivered between days 10–12 post-inoculation. Mice were anesthetized with tribromoethanol (250 mg/kg, i.p.) and irradiated using an X-RAD 225XL system (225 kV, 13 mA, 3 Gy/min) with a 1.0 cm collimator. A daily dose of 6 Gy was administered to the tumor isocenter for three consecutive days (total 18 Gy). Dosimetry was verified with a calibrated dosimeter. Control and HA130-only mice underwent anesthesia and positioning without irradiation.
HA130 (2 µmol/kg; MCE) or placebo was administered intratumorally three times per week, starting on the first day of radiotherapy. Tumor volume was measured every three days using the formula: Volume = (Length × Width²)/2. On day 25, all mice were euthanized, and tumors were collected for weighing and analysis.
For the ATX knockdown model, HCT116 cells with stable ATX knockdown were injected subcutaneously in BALB/c nude mice (n = 6). Once tumors reached ~ 100 mm³, mice were randomized into control or RT groups and irradiated as described above. Tumor growth and body weight were monitored until endpoint.
All procedures were approved by the Animal Care and Use Committee of Xuzhou Medical University (Approval No. 202302T013).

Immunohistochemistry (IHC)
IHC staining was performed using the streptavidin-peroxidase (SP) method with a standard SP Kit (ZSGB-BIO, Beijing, China) to detect Ki67. Briefly, subcutaneous graft tumor tissues from mice subjected to the specified treatments were fixed in 4% paraformaldehyde for 12 h. The tumor tissues were then paraffin-embedded and sectioned into 5 μm-thick slices. The slide was incubated overnight at 4 °C with a monoclonal rabbit anti-Ki67 antibody (GB111499, ServiceBio, Wuhan, China) following dewaxing, antigen retrieval, and blocking steps. Subsequently, an HRP-conjugated goat anti-rabbit IgG antibody (ZSGB-BIO, Beijing, China) was applied at room temperature for 2 h, followed by the addition of diaminobenzidine (DAB) (ZLI-9018; ZSGB-BIO, Beijing, China) for visualization. Images were captured using the Olympus VS120 imaging system. The proportion of proliferative tumor cells was determined by calculating the ratio of Ki67-positive tumor cells to the total number of tumor cells in the tissue samples.

Immunofluorescence
Paraffin-embedded tumor samples were sectioned at 5 μm thickness and mounted on glass slides, which were dried overnight at 42 °C. The sections underwent deparaffinization in xylene, rehydration through an ethanol series, and antigen retrieval in 10 mM sodium citrate buffer (pH 6.0) at 95 °C. Nonspecific binding sites were blocked with 3% bovine serum albumin in PBS for one hour. Primary antibodie anti-ATX (1:300 dilution) was incubated overnight at 4 °C. Secondary antibodies (VICMED, Xuzhou, China; 1:200 dilution) and DAPI (Beyotime, Shanghai, China; 1:1000 dilution) were subsequently applied. Slides were imaged using a Zeiss LSM 880 confocal microscope (Carl Zeiss, Jena, Germany).

Online analysis
We utilized the Kaplan-Meier (KM) plotter analysis tool (https://kmplot.com/analysis/) to investigate the relationship between the expression of genes in the ATX-LPA metabolism pathway and prognosis in patients with colon cancer using the colon adenocarcinoma (COAD) cohort from the TCGA database. The analysis was restricted to the COAD dataset to ensure disease specificity. The following Affymetrix transcript probes were used: ENPP2 (1555902_s_at), LPAR2 (1554293_a_at), LIPC (1555808_s_at), LCAT (1553977_at), PLA2G4A (204667_x_at), and LPCAT1 (1559219_s_at). Patients were dichotomized into high and low expression groups based on the median expression level of each gene. Hazard ratios (HR) with 95% confidence intervals (CI) were calculated. P-values were adjusted for multiple testing across the six genes using the False Discovery Rate (FDR) method.

Statistical analysis
Unless otherwise indicated, results are expressed as mean ± SEM. All statistical analyses were performed using GraphPad Prism software, and all tests were two-sided. Data distribution was assessed for normality using the Shapiro-Wilk test, and homogeneity of variances was verified using Levene’s test. For comparisons between two groups, an unpaired Student’s t-test was applied. For comparisons involving multiple groups, one-way or two-way ANOVA was applied, as appropriate, followed by Tukey’s post hoc test for multiple comparisons. A threshold of P < 0.05 denoted statistical significance.

Results

Radiation treatment increases ATX expression and LPA generation in colon cancer cells
To investigate how radiotherapy influences lipid metabolism in colon cancer cells, we performed lipidomic analysis on irradiated cells by HPLC-MS/MS. Partial least-squares discriminant analysis of lipidomics data from RT-treated colon cells revealed distinct lipid metabolic clustering compared with controls (Fig. 1a). Variable importance in projection (VIP) analysis identified LPAs as the primary metabolites driving this separation (Fig. 1b). Phospholipid heat mapping demonstrated elevated LPA levels in RT-treated colon cells (Fig. 1c). The levels of 16:0 LPA, 18:2 LPA, 18:1 LPA and 18:0 LPA were all elevated in RT-treated cells (Fig. 1d, P < 0.01). ATX is a secreted glycoprotein that exhibits lysoPLD activity. It primarily catalyzes the conversion of lysophosphatidylcholine (LysoPC) to LPA, mediating diverse physiological and pathological pathways [22]. Therefore, we explored the effects of radiation treatment on the expression and secretion of ATX in colon cancer cells. MC38 mouse colon cancer cells, along with human HCT116 and SW480 colon cancer cells, were subjected to irradiation at doses of 0, 2, 4, and 6 Gy. After 48 h, both cells and supernatants were collected for analysis. RT-qPCR analysis revealed that the mRNA expression of ATX/ENPP2 was significantly upregulated in irradiated colon cancer cells (Fig. 1e-g). Furthermore, ATX protein expression was consistently upregulated in irradiated colon cancer cells in a dose-dependent manner (Fig. 1h-j).
Since ATX functions as a secreted protein [28], we assessed its presence in the secretome using ELISA. Analysis confirmed that radiation significantly enhanced ATX secretion in a dose-dependent manner (Fig. 1k-m). In parallel, levels of its metabolite, LPA, were also substantially increased in the supernatant (Fig. 1n-p). Collectively, these results demonstrate that radiotherapy induces both the expression and secretion of ATX, strongly suggesting that the ATX-LPAR axis may be a key determinant of radiotherapy efficacy in colon cancer.

ATX promotes colon cell growth through LPAR2-AKT activation
Based on the premise that radiotherapy induces ATX secretion, we asked whether this secreted ATX could drive colon cancer cell proliferation. Accordingly, cells were treated with recombinant ATX, which resulted in a significant time- and dose-dependent increase in proliferation, as measured by CCK-8 assay (Fig. 2a-f, P < 0.001). To assess the functional contribution of ATX, we inhibited its activity with HA130. Consistently, this treatment effectively reduced LPA levels (Figure S1a) and significantly suppressed colon cancer cell proliferation (Fig. 2g-i). Furthermore, a rescue experiment demonstrated that the addition of exogenous LPA reversed the anti-proliferative effect of HA130 (Fig. 2g-i), indicating that ATX promotes colon cancer cell proliferation primarily through its metabolite LPA.

Functioning as the key enzyme in LPA production, ATX generates LPA to activate LPA receptors (LPAR1-6). This activation triggers heterotrimeric G protein signaling and initiates downstream pathways [29]. Consequently, the ATX-LPA-LPAR axis plays a well-established role in promoting tumorigenesis across a spectrum of cancers [30]. To investigate the role of LPA receptors in colon cancer, we first screened their expression profiles using qRT-PCR. This analysis revealed that LPAR2 was the predominant subtype, exhibiting high basal levels in HCT116 and SW480 cells, while the expression of other LPARs remained largely unaffected by radiation treatment (Figure S1b, c).
A previous report indicated that LPA binding to the LPAR2 activates the AKT pathway, thereby promoting colon cancer progression [23]. Therefore, we measured the phosphorylation levels of AKT in colon cancer cells stimulated by ATX. The results demonstrated that ATX treatment significantly increased AKT phosphorylation levels in colon cancer cells (Fig. 2j, k). However, the increase in AKT phosphorylation induced by ATX was significantly inhibited by HA130 (Fig. 2l, m, P < 0.001). To determine whether AKT activation promotes proliferation in colon cancer, the AKT inhibitor MK-2206 was utilized. The data indicated that MK-2206 significantly inhibited ATX-induced proliferation (Fig. 2n-p, P < 0.01), suggesting that ATX promotes colon cell growth through AKT activation.

ATX-LPAR2 blocking enhances radiosensitivity in colon cancer cells in vitro
Given our findings that RT-induced ATX secretion promotes colon cancer cell proliferation, we hypothesized that ATX inhibition could radiosensitize these cells. To test this, we stably knocked down ATX in HCT116 and SW480 cell lines using shRNA (Figure S2a) and evaluated the impact of ATX loss on radiation response and cancer progression. The impact of ATX on radiosensitivity was assessed using a clonogenic assay. ATX knockdown significantly enhanced the radiation sensitivity of colon cancer cells compared to the scramble control (Fig. 3a, b). Consistent with this functional radiosensitization, ATX depletion substantially attenuated the RT-induced phosphorylation of AKT (Fig. 3c, d), corroborating the involvement of the AKT signaling pathway.

Combining HA130 with radiation confirmed its role in radiosensitization. HA130 effectively suppressed the radiation-induced elevation of ATX and LPA in MC38 cells (Figure S2b, c) and concurrently attenuated AKT phosphorylation (Fig. 3d). This suppression of pro-survival signaling translated into enhanced functional responses: HA130 potentiated the anti-proliferative (Figure S2e) and pro-apoptotic effects of radiation (Figure S2f), confirming that ATX inhibition sensitizes colon cancer cells to radiotherapy.
To define the role of the LPA-LPAR2 axis in radioresistance, we knocked down LPAR2 in HCT116 and SW480 cells (Figure S2g). LPAR2 loss potentiated radiation-induced cell death (Fig. 3e, f) and concurrently suppressed RT-induced AKT phosphorylation (Fig. 3g, h, P < 0.001), confirming that LPAR2 signaling through AKT protects colon cancer cells from radiotherapy.
To pharmacologically validate this finding, we employed the LPAR2 inhibitor H2L5186303 in MC38 cells. Its application produced effects comparable to those of the ATX inhibitor HA130 (Figure S2h-j), confirming that LPAR2 activity contributes to radioresistance. Collectively, our data demonstrate that pharmacological inhibition of the ATX-LPAR2 axis effectively suppresses colon cancer progression and enhances the efficacy of radiotherapy in vitro.

Pharmacological inhibition of ATX prevents the tumorigenesis and radioresistance of colon cancer cells in vivo
To further investigate the effects of ATX on colon cancer growth and its radiosensitivity, an in vivo tumor formation assay was conducted. As illustrated in Fig. 4a, the mice were divided into four groups and subjected to radiation treatment (6 Gy daily) starting on the 12th day and continuing for three consecutive days. The results indicated that RT resulted in a decrease in both tumor volume and weight (Fig. 4b and c). Furthermore, inhibition of ATX enhanced the suppressive effects of RT on tumor volume and weight (Fig. 4b, c, P < 0.01). Additionally, Ki67 staining was performed on paraffin-embedded tumor sections to evaluate the inhibitory effects of HA130 on tumor growth. The findings indicated that treatment with radiation significantly reduced the proportion of Ki67-positive hyperproliferative colon cancer cells compared to the control group (Fig. 4d). To determine whether ATX and LPA levels remained elevated in tumors, we analyzed both isolated tumor tissues and supernatants. Immunofluorescence and ELISA demonstrated that radiation significantly increased ATX and LPA levels in colon cancer tissues. HA130 effectively inhibited the ATX-mediated rise in LPA in irradiated samples (Fig. 4e, f). Furthermore, the combination of RT and HA130 demonstrates the ability to inhibit the proliferation of colon cancer cells.

Next, to further validate the impact of ATX on colon cancer growth and radiosensitivity, we established an in vivo tumor formation model using HCT116 cells with ATX knockdown. The results demonstrated that ATX knockdown significantly suppressed tumor volume and weight. Moreover, it enhanced the inhibitory effects of radiation on tumor progression (Fig. 4g, h, P < 0.001). These findings indicate that ATX knockdown not only curbs tumor growth but also promotes radiosensitivity in vivo. Taken together, these results suggest that pharmacological inhibition of ATX effectively restrains tumor growth while increasing radiosensitivity under in vivo conditions.

High expression of ATX-LPA genes is associated with worse prognosis in colon cancer patients
ATX serves as the pivotal enzyme responsible for catalyzing the hydrolysis of LPC into LPA, a bioactive lipid that interacts with specific receptors located on the plasma membrane [28]. The key enzymes of the ATX-LPA axis are illustrated in Fig. 5a. To investigate the relationship between the ATX-LPA metabolic pathway and prognosis in colon cancer patients, we utilized the Kaplan–Meier plotter online analysis [31], which calculates survival statistics from large patient cohorts. The platform stratifies patients by gene expression percentiles and compares survival outcomes through Cox proportional hazards modeling with KM curve visualization [32]. The results indicated that the expression levels of lipase C (LIPC, HR = 1.72, 95% CI: 1.29–2.28, FDR P < 0.001), lecithin cholesterol acyltransferase (LCAT, HR = 1.58, 95% CI: 1.19–2.10, FDR P = 0.002), phospholipase A2 (PLA2, HR = 1.69, 95% CI: 1.27–2.25, FDR P < 0.001), and ATX (HR = 1.65, 95% CI: 1.24–2.19, FDR P = 0.001) were negatively correlated with the prognosis of colon cancer patients (Fig. 5b to e). Conversely, the expression level of lysophosphatidylcholine acyltransferase (LPCAT), an enzyme that reduces LPA synthetic substrates, was positively correlated with prognosis in colon cancer patients (HR = 0.62, 95% CI: 0.47–0.82, FDR P = 0.001, Fig. 5f). LPA binding to the LPAR2 receptor activates the AKT pathway, which drives colon cancer progression [23]. Although a trend was observed, high expression of LPAR2 was not significantly associated with survival outcome after FDR correction (HR = 1.25, 95% CI: 0.94–1.66, FDR P = 0.12, Fig. 5g). These clinical associations support the potential relevance of the ATX-LPA axis, particularly ATX, in colon cancer progression, which aligns with our mechanistic findings.

Discussion
Tumor cells secrete soluble proteins, including VEGF, FGF, and PDGF [33–35], under pressure such as RT to promote cell growth and resist radiation damage. The findings of this study elucidate the pivotal role of ATX in the progression and radioresistance of colon cancer. The data presented herein suggest that the ATX-LPAR axis is not only a significant driver of tumorigenesis but also a crucial mediator of the response to RT. By pharmacologically inhibiting ATX, we demonstrate that tumor growth can be effectively suppressed and radiosensitivity markedly enhanced (Fig. 6).

Lipidomic profiling demonstrated distinct clustering of lipid metabolism in irradiated colon cancer cells, particularly showing elevated levels of LPA species (16:0, 18:0, 18:1, 18:2) following radiotherapy. VIP analysis identified these LPAs as primary contributors to metabolic separation. Given ATX’s catalytic role in converting LPC to LPA, we investigated its involvement in radiation-induced lipid remodeling. Both RT-qPCR and Western blot analysis revealed upregulated ATX mRNA and protein expression in irradiated MC38 cells, indicating that ATX-mediated LPA generation participates in the metabolic reprogramming induced by radiotherapy in colon cancer (Fig. 1). In addition, our analysis reveals a strong correlation between high expression levels of ATX and poor prognosis in colon cancer patients, as evidenced by Kaplan-Meier survival curves (Fig. 5). This finding aligns with previous studies that have implicated the ATX-LPAR signaling pathway in various malignancies [36–38]. Future studies measuring ATX and LPA levels in clinical cohorts from colon cancer patients undergoing radiotherapy are warranted to directly validate the prognostic value of this pathway. The enzymatic activity of ATX, which converts LPC to LPA, leads to the accumulation of LPA in the tumor microenvironment. LPA is known to promote cell proliferation, migration, and survival through interactions with specific G protein-coupled receptors, particularly LPAR2. The negative prognostic implications of elevated ATX expression suggest that targeting this pathway could be a strategic approach to improve clinical outcomes in colon cancer patients.
Our investigation into the effects of RT on ATX expression reveals that radiation treatment significantly increases both the mRNA and protein levels of ATX in colon cancer cells (Fig. 2). This observation suggests a feedback mechanism in which radiation not only exerts cytotoxic effects but also activates pathways that may promote tumor survival and proliferation through ATX secretion. The increase in ATX levels following radiation may contribute to the radioresistance observed in a subset of colon cancer patients, highlighting the need for therapeutic strategies that disrupt this cycle. In line with this findings, The secretion of VEGF, FGF, and PDGF by tumors in response to radiation helps safeguard the cancer vasculature from the cytotoxic effects of radiation and contributes to the enhanced radioresistance of endothelial cells [39]. Therefore, the use of inhibitors targeting VEGF, FGF, and PDGF in combination can significantly improve the efficacy of RT.
The data presented indicate that ATX promotes colon cancer cell proliferation through the activation of the LPAR2-AKT signaling pathway. We demonstrate that ATX administration leads to increased phosphorylation of AKT, a key player in cell survival and growth signaling pathways [7]. The inhibition of ATX with HA130 not only reduces cell proliferation but also diminishes AKT activation, reinforcing the notion that the ATX-LPAR2-AKT axis is a critical mediator of colon cancer cell growth. The use of the AKT inhibitor MK-2206 further substantiates these findings, as it effectively abrogates the proliferative effects induced by ATX. This underscores the potential of targeting the ATX-LPAR2-AKT signaling cascade as a therapeutic strategy to inhibit colon cancer progression.
The study provides compelling evidence that pharmacological inhibition of ATX enhances the radiosensitivity of colon cancer cells. The combination of HA130 and radiation treatment not only reduces cell proliferation but also significantly increases apoptotic cell death, as assessed by TUNEL assays. This synergistic effect suggests that ATX inhibition may sensitize colon cancer cells to radiation by impairing their ability to survive and proliferate in the face of treatment-induced stress. The ability of HA130 to augment the effects of RT offers a promising avenue for improving therapeutic outcomes for patients with radioresistant tumors. Our in vivo experiments further corroborate the significance of ATX in colon cancer progression and radioresistance. ATX inhibition in a mouse model resulted in a marked reduction in tumor volume and weight, alongside a significant enhancement of the anti-tumor effects of radiation. The Ki67 staining results indicate that ATX inhibition leads to decreased proliferation of tumor cells, reinforcing our in vitro findings. These results collectively suggest that targeting ATX may not only inhibit tumor growth but also enhance the efficacy of existing therapeutic modalities, such as RT.
We demonstrate that the ATX-LPAR2-AKT axis drives tumor progression and radiation resistance, indicating that ATX inhibition could effectively complement RT. Such combined therapy may improve tumor eradication while minimizing damage to healthy tissues, possibly lowering required radiation doses and side effects. Elevated ATX expression correlates with poor prognosis, suggesting its utility as a biomarker to identify patients likely to respond to ATX-targeted interventions. This stratification could optimize treatment selection, especially for aggressive or recurrent cases. Existing clinical investigations of ATX inhibitors like HA130 in other cancers may facilitate their rapid repurposing for colon cancer [23]. While promising, these findings require validation in patient-derived models and early-phase trials to evaluate drug interactions and establish optimal dosing. Successful translation of ATX inhibition with RT may redefine therapeutic approaches for radioresistant colon cancer, offering survival and quality-of-life benefits.
Despite the significant findings of this study, several limitations warrant consideration. The research primarily employed the MC38, HCT116 and SW480 colon cancer cell lines, which may not capture the full heterogeneity of human colon tumors. Future studies should validate these findings in orthotopic and patient-derived models to assess their broader applicability. Although the in vivo experiments used immunocompetent mice, the study did not examine how the immune microenvironment influences ATX-mediated radioresistance. While HA130 demonstrated effective ATX inhibition, its potential off-target effects and optimal therapeutic dosing require further characterization for clinical application. Resolving these limitations would strengthen both the validity and translational potential of our results.
In summary, our study highlights the critical role of the ATX-LPA signaling pathway in colon cancer progression and radioresistance. By inhibiting ATX, we can not only impede tumor growth but also enhance the sensitivity of colon cancer cells to RT. These findings underscore the potential of ATX as a therapeutic target and pave the way for future research aimed at improving treatment strategies for colon cancer patients. The integration of ATX inhibitors into clinical practice could represent a significant advancement in the management of this challenging malignancy.

Supplementary Information