LMTK2 Inhibits Metastasis of Colon Cancer Cells Through p38 MAPK

ABSTRACT

Metastasis is the leading cause of death in patients with colon cancer. Although Lemur tyrosine kinase 2 (LMTK2) has been implicated in cell proliferation, its role in metastasis is unclear. In this study, we found that low LMTK2 expression correlates with metastatic status (M1 stage) and a higher metastasis rate in a large colon cancer cohort. Additionally, LMTK2 expression is negatively correlated with pro‐metastatic genes and positively correlated with metastasis suppressors, a pattern that was validated in 50 clinical colon cancer specimens. Functional studies demonstrated that LMTK2 potently inhibits cell migration and invasion in vitro and suppresses metastasis in a mouse model in vivo. Mechanistically, LMTK2 disrupts the interaction between p38 MAPK and its upstream kinases MKK3/6, thereby inhibiting p38 activation and subsequent cell migration and invasion. Notably, the kinase‐deficient LMTK2 (K168R) mutant also suppresses p38 phosphorylation and impedes cell migration and invasion. These results uncover a novel anti‐metastatic function of LMTK2, which acts through inhibition of p38 independently of its kinase activity.

LMTK2 suppresses colon cancer metastasis by acting as a kinase‐independent scaffold that directly binds p38, blocking its interaction with MKK3/6 and subsequent phosphorylation, thereby inhibiting the downstream pro‐metastatic gene program.

Abbreviations

CDK5

Cyclin‐dependent kinase 5

CI

Confidence intervals

FDR

False discovery rate

GEO

Gene Expression Omnibus

GSEA

Gene Set Enrichment Analysis

GSK3β

Glycogen synthase kinase 3 beta

HE

Hematoxylin–eosin

HR

Hazard ratios

LMTK2

Lemur tyrosine kinase 2

MKK3/6

MAPK kinase 3 and 6

NES

Normalized enrichment score

PCA

Principal component analysis

PP1C

Protein phosphatase 1C

RFS

Recurrence‐free survival

RT‐qPCR

Quantitative real‐time PCR

shRNA

Short hairpin RNA

siRNA

small interfering RNA

1. Introduction

Colorectal cancer, including malignancies of the colon and rectum, was the third most commonly diagnosed cancer worldwide in 2025, with an estimated 154,270 new cases and 52,900 deaths [1]. Although the 5‐year survival rate for localized colorectal cancer is about 65.4%, this number drops to only 15.6% for cancers that have metastasized to the liver, lungs, or other organs [2]. Therefore, identifying novel key regulators that govern the metastatic process is crucial for developing new prognostic biomarkers and therapeutic strategies.

Lemur tyrosine kinase 2 (LMTK2), also known as serine/threonine‐protein kinase KPI‐2, is a membrane‐associated serine/threonine protein kinase. Like other members of the LMTK family, it contains N‐terminal transmembrane helices and a long C‐terminal cytoplasmic tail with serine/threonine/tyrosine kinase activity [3]. LMTK2 has been reported to bind to and phosphorylate threonine 320 of the protein phosphatase 1C (PP1C), leading to inactivation of PP1C [3]. LMTK2 kinase activity is regulated by Cyclin‐dependent kinase 5 (CDK5) [4] and has been shown to control Smad2 signaling by regulating PP1C and glycogen synthase kinase 3 beta (GSK3β) [5]. LMTK2 also plays a critical role in endosomal membrane trafficking [6, 7] and neuronal signaling [8, 9, 10, 11]. Its role in cancer, however, is complex and appears to be highly context‐dependent. Mutations in LMTK2 have been identified in a few cancer types [12, 13, 14, 15]. Several studies have suggested a pro‐tumorigenic role, linking LMTK2 to the regulation of cell proliferation and apoptosis pathways [16, 17, 18, 19]. Conversely, other evidence points to potential tumor‐suppressive functions, such as its involvement in inhibiting proliferation in gastric cancer and human glioblastoma [20, 21].

While its name implies kinase‐dependent actions, emerging evidence suggests that LMTK2 may also act as a scaffold protein, independently of its catalytic activity [6], revealing its complicated functions. Despite these insights, the specific function and mechanistic role of LMTK2 in colon cancer progression, particularly in the critical process of metastasis, remain entirely unexplored. We discover that low LMTK2 expression correlates with metastatic status (M1 stage) and a higher metastasis rate in a large colon cancer cohort. Subsequently, we demonstrate that knockdown of LMTK2 promotes the migration and invasion of colon cancer cells, whereas overexpression of LMTK2 has the opposite effect. In vivo animal studies show that knockdown of LMTK2 increases liver metastasis of colon cancer cells. Mechanistic studies demonstrate that LMTK2 suppresses the interaction between p38 MAPK and MAPK kinase 3 and 6 (MKK3/6), the upstream kinases of p38, leading to inhibition of p38 phosphorylation and, thereafter, cell migration and invasion. Our results suggest that LMTK2 acts as a negative regulator of colon cancer cell metastasis.

2. Materials and Methods

2.1. Bioinformatic Analysis of LMTK2 in Colon Cancer

2.1.1. Data Acquisition and Preprocessing

Raw gene expression data (GPL570 platform) and corresponding clinical information for the GEO dataset GSE39582 [22] were downloaded from the NCBI Gene Expression Omnibus (GEO; https://www.ncbi.nlm.nih.gov/geo/; accessed on 22 August 2025). The data were downloaded in MINiML format, which contains complete platform, sample, and series GSE records.

Data preprocessing was performed as previously described [23, 24]. Raw data were log2‐transformed and quantile‐normalized using the normalize.quantiles function from the preprocessCore package (version 1.68.0) [25] in R (version 4.3.1). Probe IDs were converted to gene symbols based on the platform annotation information. Probes matching multiple genes were excluded, and the average expression value was calculated for genes represented by multiple probes. Batch effects were removed using the removeBatchEffect function from the limma package (version 3.58.1) [26] with default parameters. Preprocessing quality was assessed by inspecting box plots (for normalization) and by comparing principal component analysis (PCA) plots before and after batch removal.

The final processed dataset comprised 566 human colon cancer samples. After excluding cases with incomplete M‐stage (metastasis stage) information, 543 patients were included for subsequent analyses. Patients were divided into high‐ and low‐expression groups based on the median expression level of LMTK2.

2.1.2. Differential Expression and Correlation Analysis

LMTK2 expression levels were compared between M0 (non‐metastatic) and M1 (metastatic) subgroups using the Wilcoxon rank‐sum test. Spearman correlation coefficients were calculated between LMTK2 and a predefined panel of metastasis‐associated genes using the cor.test function in R. Correlation results were visualized as a co‐expression heatmap using the ggplot2 package (version 3.4.4).

2.1.3. Pathway Enrichment Analysis

Gene Set Enrichment Analysis (GSEA) was performed using the clusterProfiler package (version 4.8.3) [27]. All protein‐coding genes were pre‐ranked based on their Pearson correlation coefficients with LMTK2. The analysis was conducted using the Canonical Pathways gene set collection (c2.cp.all.v2022.1.Hs.symbols.gmt), which contains 3050 gene sets and was obtained from the Molecular Signatures Database (MSigDB; https://www.gsea‐msigdb.org/gsea/msigdb). Significance was assessed using 1000 permutations, and gene sets with a false discovery rate (FDR) < 0.005 and |normalized enrichment score (NES)| ≥ 2.2 were considered significantly enriched.

2.1.4. Survival Analysis

The prognostic value of LMTK2 expression was evaluated according to M stage (M0 and M1) using the Kaplan–Meier plotter platform (KM plotter; https://kmplot.com/analysis/; accessed on 15 August 2025). Recurrence‐free survival (RFS) was analyzed for each subgroup. Patients were stratified by the median expression of LMTK2. Kaplan–Meier curves were generated, and the log‐rank test was applied to compare survival differences. Hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated using a univariate Cox proportional hazards model.

2.2. Tumor Tissues and Cell Culture

Fifty colon cancer tissues were obtained from patients who underwent surgical resection at the Affiliated Hospital of Qingdao University between 2020 and 2022. Tissues were collected immediately after resection, snap‐frozen in liquid nitrogen, and stored at −80°C until use. The study protocol was approved by the Ethics Committee of the Medical College of Qingdao University (approval no. QDU‐HEC‐2025555). Written informed consent was obtained from all participants. The cohort comprised 50 patients (30 males, 20 females; age range: 44 − 81 years; median age: 63.6 years) with histologically confirmed colon adenocarcinoma.

Human colon cancer cell lines RKO and HCT116 were purchased from Procell Life Science & Technology Co. Ltd. (Wuhan, China; Cat#CL‐0196 and Cat#CL‐0096). The cells were authenticated by short tandem repeat (STR) profiling upon purchase, and a certificate of analysis was provided. Cells were maintained in MEM (RKO) or McCoy's 5A medium (HCT116) (Procell; Cat#PM150410 and Cat#PM150710) supplemented with 10% fetal bovine serum (FBS; Procell; Cat#164210) at 37°C in a humidified atmosphere containing 5% CO₂.

2.3. RNA Isolation and Quantitative Real‐Time PCR

Total RNA was extracted from colon cancer tissues using TRIzol reagent (Ambion, Austin, TX, USA; Cat#15596026) according to the manufacturer's protocol. RNA concentration and purity were assessed using a NanoDrop One (Thermo Fisher Scientific, Waltham, MA, USA). Reverse transcription of the total RNA was performed using FastKing gDNA Dispelling RT SuperMix (Tiangen, Beijing, China; Cat#KR118) following the manufacturer's instructions.

Quantitative real‐time PCR (RT‐qPCR) was performed using Universal SYBR Green Fast qPCR Mix (ABclonal, Woburn, MA, USA; Cat#RK21203) on a QuantStudio 3 Real‐Time PCR System (Thermo Fisher Scientific). Each reaction (20 μL) contained 10 μL SYBR Green Master Mix, 0.4 μL of each primer (10 μM), 1 μL cDNA template, and 8.2 μL nuclease‐free water. The cycling conditions were: 95°C for 3 min; 40 cycles of 95°C for 5 s and 60°C for 34 s; followed by a melting curve analysis. All reactions were performed in triplicate. The expression levels of LMTK2 and a panel of metastasis‐associated genes (ZEB1, MMP9, VIM, CDH1, CDX2, TWIST1) were measured. Primer sequences are listed in Table S1. Relative gene expression was calculated using the 2^‐ΔΔCq method [28], with β‐actin as the endogenous control. The correlation between LMTK2 and other genes was assessed using Pearson correlation analysis.

2.4. Short Hairpin RNA (shRNA) and Small Interfering RNA (siRNA)

Short hairpin RNA (shRNA) knockdown plasmids targeting LMTK2 were constructed using the pLKO.1 vector. Complementary DNA oligonucleotides targeting LMTK2 were synthesized (Sangon Biotech, Shanghai, China), annealed, and inserted into pLKO.1 as previously described [18]. The targeting sequences for LMTK2 were:

Sense:

5′‐CCGGCAGGTACAAGGAGGATTATATCTCGAGATATAATCCTCCTTGTACCTGTTTTTG‐3′.

Antisense: 5′‐AATTCAAAAACAGGTACAAGGAGGATTATATCTCGAGATATAATCCTCCTTGTACCTG‐3′.

Scrambled control sequences were:

Sense: 5′‐CCGGCCTAAGGTTAAGTCGCCCTCGCTCGAGCGAGGGCGACTTAACCTTAGGTTTTTG‐3′.

Antisense: 5′‐AATTCAAAAACCTAAGGTTAAGTCGCCCTCGCTCGAGCGAGGGCGACTTAACCTTAGG‐3′.

Small interfering RNA (siRNA) oligos targeting LMTK2 and negative control siRNAs were purchased from GenePharma (Shanghai, China). siRNA sequences are listed in Table S1.

2.5. Plasmid Construction

The mammalian expression plasmid encoding N‐terminally 3 × FLAG‐tagged human LMTK2 (Flag‐LMTK2) was generated as previously described [18]. Briefly, the full‐length open reading frame of human LMTK2 (GenBank accession no. NM_014916) was amplified by high‐fidelity PCR using KOD‐Plus‐Neo DNA polymerase (TOYOBO, Osaka, Japan; Cat#KOD‐401) and cloned into the pcDNA3.1(+) vector. The integrity of the final construct was confirmed by Sanger sequencing (Sangon Biotech).

To generate the kinase‐dead mutant Flag‐LMTK2 (K168R), site‐directed mutagenesis was performed on the Flag‐LMTK2 plasmid as the template using the Mut Express II Fast Mutagenesis Kit V2 (Vazyme, Nanjing, China; Cat#C214‐02) according to the manufacturer's instructions. A specific mutagenic primer pair was designed to introduce the point mutation. The successful introduction of the mutation was verified by DNA sequencing (Sangon Biotech).

2.6. Transient Transfection of Cells

Transient transfection of RKO and HCT116 cells was performed using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA; Cat#11668019) according to the manufacturer's instructions. Briefly, cells were seeded in 6‐well plates and cultured overnight to reach 70%−80% confluence. For each well, 2.5 μg of plasmid DNA or 100 pmol siRNA was diluted in 250 μL Opti‐MEM I (Invitrogen; Cat#31985070), and 10 μL or 5 μL Lipofectamine 2000 was diluted in 250 μL Opti‐MEM I. After a 5 min incubation at room temperature, the diluted DNA (or RNA) and Lipofectamine 2000 were combined and incubated for 20 min at room temperature. The mixture was then added to the cells and incubated at 37°C for 6 h. The medium was then replaced with fresh complete culture medium, and cells were cultured for an additional 48 h.

2.7. Western Blot

Western blot analysis was performed using standard procedures. Cells were lysed in RIPA lysis buffer (Beyotime, Shanghai, China; Cat#P0013C) supplemented with protease inhibitor and phosphatase inhibitor. Protein concentrations were determined using the BCA Protein Assay Kit (Thermo Fisher Scientific; Cat#23225) according to the manufacturer's protocol.

Equal amounts of protein (40 μg) were separated by 10% SDS‐PAGE and transferred to PVDF membranes (Millipore, Burlington, MA, USA; Cat#IPVH00010). Membranes were blocked with 5% non‐fat milk for 1 h at room temperature, followed by incubation with primary antibodies overnight at 4°C. After washing three times with TBST, membranes were incubated with HRP‐conjugated secondary antibodies for 1 h at room temperature. Protein bands were visualized using chemiluminescence (ECL) substrate (Millipore; Cat#WBULS0500) and imaged using a Bio‐Rad chemiluminescence imaging system (Bio‐Rad, Hercules, CA, USA).

The following primary antibodies were used: Anti‐LMTK2 (Sigma‐Aldrich, St. Louis, MO, USA; Cat#SAB4500900; rabbit polyclonal, 1:1000), anti‐p38 (Cell Signaling Technology, Danvers, MA, USA; Cat#9212; rabbit polyclonal, 1:1000), anti‐phospho‐p38 (Thr180/Tyr182) (Cell Signaling Technology; Cat#9211; rabbit polyclonal, 1:1000), anti‐phospho‐MKK3 (Ser189)/MKK6 (Ser207) (Cell Signaling Technology; Cat#12280; rabbit monoclonal, 1:1000), anti‐β‐actin (Abmart, Shanghai, China; Cat#P30002; rabbit polyclonal, 1:2000), anti‐HA (Abmart; Cat#M20003; mouse monoclonal, 1:2000), anti‐Flag (Abmart; Cat#M20008; mouse monoclonal, 1:2000), anti‐MKK3/6 (Santa Cruz Biotechnology, Dallas, TX, USA; Cat#sc‐136,982; mouse monoclonal, 1:500), anti‐PP1A (Cell Signaling Technology; Cat#2582; rabbit polyclonal, 1:1000), and anti‐phospho‐PP1A (Thr320) (Cell Signaling Technology; Cat#2581; rabbit polyclonal, 1:1000). HRP‐conjugated Goat anti‐Rabbit IgG (H+L) (ABclonal; Cat#AS014) and HRP‐conjugated Goat anti‐Mouse IgG (H+L) (ABclonal; Cat#AS003) were used at a 1:3000 dilution.

2.8. Determination of Cell Migration and Invasion

Cell migration was determined using Transwell permeable support inserts with 8 μm porous membranes (Costar, Corning Inc., Corning, NY, USA; Cat#3422) in a 24‐well plate. 2 × 10⁵ cells suspended in serum‐free medium (200 μL) were added into the upper chamber, and complete medium (500 μL) was added into the lower chamber. Cells were cultured at 37°C in 5% CO₂. After 16 h, the cells remaining on the upper surface of the membrane were removed using a cotton swab, and those migrated to lower surface were fixed with 4% paraformaldehyde (Solarbio, Beijing, China; Cat#P1110) for 15 min at room temperature, stained using the Richard‐Allan Scientific 3‐Step Stain Set (Thermo Fisher Scientific; Cat#3300) according to the manufacturer's instructions, and counted under an inverted microscope (Olympus IX73, Tokyo, Japan) at 200 × magnification. Five random fields per membrane were counted, and each experiment was performed in triplicate.

To determine cell invasion, the Transwell inserts were pre‐coated with growth factor‐reduced Matrigel (Corning; Cat#356231) at 37°C for 4 h. The cells were then plated onto the inserts and allowed to invade for 24 h. The invaded cells on the lower membrane surface were fixed, stained, and quantified as described above.

2.9. Determination of Cancer Cell Metastasis in Vivo

Cancer cell liver metastasis assay was performed as described [29]. In brief, 2 × 10⁶ cells (in 50 μL PBS) were injected into the spleen. Ten minutes after injection, the spleens were removed. After 6 weeks, the mice were sacrificed, and liver metastasis of cancer cells was evaluated. The four‐week‐old male nude mice (Balb/c‐nu/nu) were obtained from the Shanghai Experimental Animal Center and maintained under pathogen‐free conditions. All animal experiments were performed in accordance with the guidelines of the Institutional Animal Care and Use Committee (Approval No. QDU‐AEC‐2025812).

2.10. Hematoxylin–Eosin (HE) and Immunohistochemistry (IHC) Staining

Tumor tissues were fixed in 4% paraformaldehyde for 24 h at room temperature, dehydrated through a graded ethanol series, cleared in xylene, and embedded in paraffin. Tissue sections (4 μm thick) were cut using a microtome, mounted on glass slides, and dried overnight at 37°C.

HE staining was performed using the HE Staining Kit (Solarbio; Cat#G1120) according to the manufacturer's instructions. Briefly, sections were deparaffinized, rehydrated, stained with hematoxylin for 5 min, differentiated if necessary, blued in tap water, counterstained with eosin for 2 min, dehydrated, cleared, and mounted with neutral balsam.

IHC staining was performed for phospho‐p38 (T180/Y182). Briefly, sections were deparaffinized, rehydrated, and subjected to antigen retrieval in citrate buffer (pH 6.0). Endogenous peroxidase was blocked with 3% H₂O₂, and sections were incubated with anti‐phospho‐p38 antibody overnight at 4°C. After washing, sections were incubated with an HRP‐conjugated secondary antibody, and signals were visualized using DAB substrate. Sections were counter‐stained with hematoxylin. The invasive front was defined as the tumor‐liver interface with infiltrating tumor cells, and the tumor core as the central area of the nodule.

2.11. Statistical Analysis

All statistical analyses were performed using R software (version 4.3.1; R Core Team, 2025) or GraphPad Prism (version 9.0; GraphPad Software, San Diego, CA, USA). Data are presented as mean ± standard deviation (SD).

Bioinformatic data were analyzed as described below. Categorical data were compared by the chi‐squared test. Continuous variables were compared by Student's t‐test. Gene correlations were evaluated using Spearman's rank correlation. In GSEA, gene sets with FDR q‐value < 0.005 were considered significantly enriched. Survival differences were assessed by log‐rank test, with p < 0.05 considered statistically significant. Experimental data were analyzed by unpaired two‐tailed t‐tests or two‐way ANOVA using GraphPad Prism 9.0. Statistical significance was defined as p < 0.05.

3. Results

3.1. LMTK2 Expression Is Associated With Metastatic Status in Colon Cancer Patients

The GSE39582 dataset (containing transcriptomic profiles of 543 CRC patients with complete M‐stage information) was used for survival and Gene Set Enrichment Analysis (GSEA) analyses. Patients were classified into M0 (non‐metastatic, n = 482) and M1 (metastatic, n = 61) subgroups according to the clinical annotation. LMTK2 high‐ and low‐expression groups were defined based on the median expression value of LMTK2 in the entire cohort. To gain a comprehensive understanding of the biological processes associated with LMTK2 expression, we performed GSEA. Using stringent thresholds (|NES| ≥ 2.2, FDR < 0.005), we identified 148 significantly enriched pathways, which were categorized into six functional themes (Figure 1A). Notably, pathways facilitating metastasis, including ECM organization and cell adhesion, were significantly suppressed (NES < 0) in tumors with high LMTK2 expression. Conversely, pathways governing cell proliferation and cycle were positively enriched (NES > 0), consistent with our previous study [18]. Among the top pathways (Figure 1B), Core Matrisome (NES = −3.10) and Extracellular Matrix Organization (NES = −2.99) were the most suppressed, while G2/M Checkpoint (NES = 2.74) was the most activated. This evidence suggests that LMTK2 functions as a metastasis suppressor in colon cancer.

FIGURE 1.

LMTK2 expression is associated with metastatic status. (A) Pathways enriched in GSEA based on correlation with LMTK2 expression were grouped into six major biological themes (|Normalized Enrichment Score, NES| ≥ 2.2, False Discovery Rate, FDR < 0.005). (B) The top three most significantly enriched pathways from each category are shown in (A). (C) The distribution of patients' metastasis status between LMTK2 high‐ and low‐expression groups, stratified by the median expression value. (D) LMTK2 expression levels between patients with (M1) and without (M0) distant metastasis. (E) Kaplan–Meier curves for recurrence‐free survival (RFS) based on LMTK2 expression levels within M0 and M1 patient subgroups. * p < 0.05.

Among the 543 colon cancer patients, those with high LMTK2 expression (based on median split) had a significantly lower frequency of M1 stage disease (8.3%) than those with low LMTK2 expression (14.0%, p < 0.05, Figure 1C). Also, LMTK2 expression was lower in the M1 group compared to the M0 group (p < 0.05, Figure 1D), suggesting a potential association between LMTK2 and tumor metastasis.

Prognostic analysis further uncovered a stage‐specific role for LMTK2 (Figure 1E). In non‐metastatic (M0) patients, high LMTK2 expression was significantly associated with poorer RFS. Conversely, in patients with distant metastases (M1), high LMTK2 expression correlated with longer RFS.

Using the GSE39582 cohort, we found that high LMTK2 expression was associated with poorer RFS in M0 patients (log‐rank p = 0.020) but better RFS in M1 patients (log‐rank p = 0.030) (Figure S1A). Cox regression analysis revealed a borderline‐significant interaction between LMTK2 expression and M stage (p = 0.059), supporting a stage‐dependent prognostic role (directional HRs: 1.51 in M0, 0.52 in M1). Subgroup‐specific GSEA (|NES| ≥ 2.2, FDR < 0.005) revealed that metastatic pathways were similarly suppressed by high LMTK2 in both M0 and M1 patients. However, pro‐proliferative pathways were significantly enriched only in M0 patients and absent in M1 patients (Figure S1B,C).

3.2. LMTK2 Correlates With Metastasis‐Associated Genes in Datasets and Clinical Tissues

To further investigate the association between LMTK2 and colon cancer metastasis, we evaluated the correlation between LMTK2 and well‐established metastasis‐associated genes. As shown in Figure 2A, LMTK2 expression was negatively correlated with pro‐metastatic genes such as SNAI2 (R = −0.32), CDH2 (R = −0.303), and VIM (R = −0.328), and positively correlated with the metastasis suppressor genes CDH1 (R = 0.247) and CDX2 (R = 0.229).

FIGURE 2.

LMTK2 Correlates with Metastasis‐Associated Genes. (A) Co‐expression heatmap between LMTK2 and a panel of key metastasis‐related genes in the GSE39582 colon cancer cohort. (B) Scatter plots of correlation between LMTK2 and VIM (C), ZEB1 (D), MMP9 (E), TWIST1 (F), CDH1 (G), and CDX2 in an independent cohort of 50 human colon cancer tissues using qRT‐PCR. The Pearson correlation coefficient (R) and corresponding p‐value are indicated on each plot. **p < 0.01, ***p < 0.001.

To experimentally validate the co‐expression patterns above, we quantified the mRNA levels of LMTK2 and six metastasis‐related genes in 50 clinical colon cancer tissues. These genes include pro‐metastatic drivers VIM, ZEB1, MMP9, and TWIST1, and metastasis suppressors CDH1 and CDX2, which are core regulators of the metastatic cascade [30, 31]. As shown in Figure 2B–E, LMTK2 expression exhibited negative correlations with the pro‐metastatic genes VIM (R = −0.52), ZEB1 (R = −0.48), MMP9 (R = −0.50), and TWIST1 (R = −0.37). Conversely, positive correlations were observed between LMTK2 and CDH1 (E‐cadherin; R = 0.47, Figure 2F) and CDX2 (R = 0.67, Figure 2G). These results provide evidence that LMTK2 may exert an anti‐metastatic function in human colon cancer.

3.3. LMTK2 Inhibits Colon Cancer Cell Metastasis

To investigate the role of LMTK2 in cancer cell migration and invasion, we performed Transwell assays. Our results demonstrated that depletion of LMTK2 enhanced HCT116 cell migration and invasion (Figure 3A,B). By contrast, overexpression of LMTK2 curbed the migration and invasion (Figure 3C,D) of the cells. Similar results were obtained when RKO cells were examined (Figure 3E–H). We also did an in vivo animal study (Figure 4A) and showed that depletion of LMTK2 promoted liver metastasis of HCT116 cells (Figure 4B–D), while overexpression of LMTK2 exhibited the opposite effects (Figure 4E,F). Collectively, these findings indicate that LMTK2 acts as a suppressor of colon cancer metastasis.

FIGURE 3.

LMTK2 inhibits migration and invasion of colon cancer cells. (A, B) Knockdown of LMTK2 promoted the migration (A) and invasion (B) of HCT116 cells. Cells were transfected with control or LMTK2 siRNA oligos, and cell migration and invasion were determined as described in Methods. (C, D) Overexpression of LMTK2 inhibited migration (C) and invasion (D) of HCT116 cells. (E, F) Knockdown of LMTK2 promoted the migration (E) and invasion (F) of RKO cells. (G, H) Overexpression of LMTK2 inhibited migration (G) and invasion (H) of RKO cells. Left panels, images of migrated or invaded cells; middle panels, the counting of migrated or invaded cells; right panels, LMTK2 knockdown or overexpression efficiency. *p < 0.05, **p < 0.01, ***p < 0.001. Black bar = 100 μm.

FIGURE 4.

Knockdown of LMTK2 promotes metastasis of HCT116 cells. (A) Schema of the animal experiments. HCT116 cells stably expressing control or sh‐LMTK2, and HCT116 cells stably overexpressing control or LMTK2 plasmids, were injected into the mouse spleen, and liver metastasis was determined as described in Methods. (B) The left panel shows representative mouse livers, and the right panel shows knockdown efficiency of LMTK2. (C) Number of metastatic foci per liver. (D) H&E stain of mouse livers. (E) The left panel shows representative mouse livers, and the right panel shows overexpression efficiency of LMTK2. (F) Number of metastatic foci per liver. Black bar = 50 μm. * p < 0.05, **p < 0.01.

3.4. LMTK2 Modulates Metastatic Potential Through Inactivating p38 MAPK

The p38 MAPK is involved in the regulation of cell migration and invasion [32, 33]. Our results showed that LMTK2‐knockdown enhanced the phosphorylation of p38 in both HCT116 and RKO cells (Figure 5A), and overexpression of LMTK2 did the opposite (Figure 5B). We thus presumed that p38 might be involved in the regulation of cell migration and invasion by LMTK2.

FIGURE 5.

LMTK2 inhibits cell migration and invasion through p38. (A) HCT116 and RKO cells were transfected with control or LMTK2 siRNA oligos. After 48 h, the cells were harvested for immunoblotting. (B) HCT116 and RKO cells were transfected with control or Flag‐LMTK2 plasmids. After 24 h, the cells were collected for immunoblotting. (C, D) HCT116 cells expressing control or LMTK2 siRNA oligos were employed for determination of cell migration (C) and invasion (D) in the presence or absence of p38 inhibitor SB203580 (10 μM). (E, F) HCT116 (E) or RKO (F) cells expressing control or LMTK2 siRNA oligos were employed for determination of metastasis‐related gene expression with or without p38 inhibitor SB203580 (10 μM) treatment. *p < 0.05, **p < 0.01, ***p < 0.001. Black bar = 100 μm.

To investigate whether LMTK2 modulated migration and invasion through p38, we depleted LMTK2 and determined migration and invasion of HCT116 cells in the presence of the p38 inhibitor SB203580. We found that pharmacological inhibition of p38 significantly repressed cell migration and invasion stimulated by LMTK2‐depletion (Figure 5C,D). SB203580 alone suppressed basal migration and invasion of the cells (Figure S2), consistent with previous reports of p38 promoting metastasis in certain contexts [32, 33].

We next sought to determine whether p38 mediated the transcriptional regulation of the metastasis‐related genes regulated by LMTK2. Remarkably, inhibition of p38 attenuated the expression of VIM, ZEB1, MMP9, and TWIST1 induced by LMTK2‐depletion (Figure 5E), while inhibition of p38 rescued the expression of CDH1 and CDX2 (Figure 5F). Moreover, IHC staining for phospho‐p38 on metastatic tissues from LMTK2‐knockdown mice showed that p‐p38‐positive cells were markedly enriched at the invasive front of metastatic lesions (Figure S3). We performed an animal experiment and showed that SB203580 treatment significantly inhibited HCT116 cell liver metastasis (Figure S4A,B). Collectively, these results indicate that LMTK2 restrains the metastatic potential of colon cancer cells via inhibition of p38.

3.5. LMTK2 Inhibits p38 Phosphorylation by Blocking the Interaction of p38 and MKK3/6

We next investigated the molecular mechanism by which LMTK2 inhibited phosphorylation of p38. Firstly, we determined whether the kinase activity of LMTK2 was required for this regulation. We constructed an LMTK2 (K168R) mutant that was reported to lose kinase activity [34] and demonstrated that the mutant was kinase‐dead (Figure S5A). We found that overexpression of LMTK2 (K168R) still suppressed p38 phosphorylation (Figure 6A). In line with the results, overexpression of LMTK2 (K168R) significantly inhibited the migration (Figure 6B) and invasion of the cells (Figure 6C). These results indicate that the LMTK2 kinase activity is not required for the regulation of p38 phosphorylation and cell migration and invasion.

FIGURE 6.

LMTK2 inhibits p‐p38 through MKK3/6. (A) HCT116 and RKO cells were transfected with control or Flag‐LMTK2 (K168R) plasmids. After 24 h, the cells were collected for immunoblotting. (B) HCT116 cells expressing control or Flag‐LMTK2 (K168R) plasmids were used for determination of cell migration (B) and cell invasion (C) as described in Methods. Left panel, cell images; middle panel, expression efficiency of Flag‐LMTK2 (K168R); right panel, the counting of migrated or invaded cells. (D) RKO cells were transfected with control or MKK3/6 siRNA oligos. After 48 h, the cells were harvested for western blot. (E) HCT116 and RKO cells were transfected with control, si‐LMTK2, or si‐LMTK2 and si‐MKK3/6 oligos. After 48 h, the cells were harvested for immunoblotting. *p < 0.05, **p < 0.01. Black bar = 100 μm.

MAPK kinase 3 and 6 (MKK3/6) are the kinases that phosphorylate p38 [35, 36]. Knockdown of MKK3/6 decreased phosphorylation of p38 (Figure 6D). We therefore determined whether LMTK2 modulated p38 through MKK3/6 and found that depletion of MKK3/6 (Figure S6) decreased p38 phosphorylation stimulated by LMTK2‐knockdown (Figure 6E), indicating that MKK3/6 is involved.

3.6. LMTK2 Inhibits MKK3/6‐p38 Interaction

Knockdown of LMTK2 stimulated p38 phosphorylation with little effect on the expression and phosphorylation of MKK3/6 (Figure 7A). We thus presumed that LMTK2 might inhibit p38 phosphorylation by interfering with the association between p38 and MKK3/6. As expected, our results demonstrated that LMTK2 knockdown enhanced the interaction between MKK3/6 and p38 (Figure 7B). Conversely, overexpression of LMTK2 reduced the association between p38 and MKK3/6 in a dose‐dependent manner (Figure S7A), providing reciprocal evidence that LMTK2 disrupts the formation of the p38–MKK3/6 complex. These results indicate that LMTK2 restricts the association of p38 and MKK3/6, thereby inhibiting the phosphorylation of p38.

FIGURE 7.

LMTK2 blocks the interaction between p38 and MKK3/6. (A) HCT116 and RKO cells were transfected with control or si‐LMTK2 oligos. After 48 h, the cells were harvested for western blot. (B) RKO cells were transfected with control or si‐LMTK2 oligos. After 48 h, the cells were collected, and cellular proteins were prepared for immunoprecipitation assay. (C) RKO cells expressing control or Flag‐LMTK2 plasmids were treated with or without TNFα (20 ng/mL) for 5 min, followed by immunoprecipitation analyses. (D) RKO cell lysates were immunoprecipitated with anti‐LMTK2 antibody or IgG, followed by immunoblotting with antibodies against p38 and LMTK2.

TNFα induces phosphorylation of p38 [37] and plays a crucial role in the promotion of cancer metastasis [38]. TNFα treatment enhanced the association between p38 and MKK3/6 and increased p38 phosphorylation. However, the TNFα‐induced interaction between p38 and MKK3/6 and the phosphorylation of p38 were suppressed by LMTK2 overexpression (Figure 7C). Overexpression of LMTK2 had little effect on TNFα‐induced MKK3/6 phosphorylation, which is consistent with the data that LMTK2 knockdown did not influence MKK3/6 phosphorylation (Figure 7A). To elucidate the molecular mechanism by which LMTK2 inhibits p38 phosphorylation, we investigated whether LMTK2 directly interacts with p38. As shown in Figure 7D, an interaction was observed between LMTK2 and p38. Furthermore, this interaction did not depend on the phosphorylation status of p38, as λ‐phosphatase treatment did not alter the binding between LMTK2 and p38 (Figure S7B). Moreover, Co‐IP assays revealed that LMTK2 (K168R) retained the ability to bind p38 comparably to LMTK2‐WT (Figure S5B), indicating that LMTK2 inactivates p38 independently of its kinase activity. Collectively, our results suggest that LMTK2 binds p38, thereby blocking its interaction with MKK3/6, which in turn inhibits p38 phosphorylation and ultimately suppresses cancer metastasis. A graphical abstract summarizing this study is provided in Figure 8.

FIGURE 8.

Graphical abstract summarizing the proposed mechanism. LMTK2 suppresses colon cancer metastasis by acting as a kinase‐independent scaffold that directly binds p38, blocking its interaction with MKK3/6 and subsequent phosphorylation, thereby inhibiting the downstream pro‐metastatic gene program.

4. Discussion

Metastasis is the major cause of mortality in colon cancer, making the discovery of metastasis‐modulatory genes and their mechanistic dissection a research priority [39]. Despite its established role in proliferation, the contribution of LMTK2 to metastatic progression has remained elusive. Here, we identify LMTK2 as a bona fide suppressor of colon cancer metastasis, capable of restraining migration and invasion in vitro and metastasis in vivo, at least in part through dampening p38 phosphorylation.

The role of LMTK2 in metastasis of colon cancer was unclear. Our GSEA analysis revealed marked negative enrichment of metastasis pathways, such as ECM organization and EMT, but positive enrichment of the cell cycle pathway (Figure 1A,B). The expression of LMTK2 was also closely associated with metastasis status (Figure 1C,D) and metastasis‐related gene expression, which were subsequently validated in our clinical colon cancer specimens (Figure 2). In vitro and in vivo experiments demonstrated that LMTK2 inhibits the metastatic potential of colon cancer cells. These data indicate that LMTK2 acts as a negative regulator of metastasis in colon cancer cells. Our previous results showed that LMTK2 promoted colon cancer cell proliferation [18]. These results suggest a ‘migration‐proliferation dichotomy’ for LMTK2. Notably, this dichotomy manifests as opposite prognostic effects depending on disease stage. As shown in Figure S1, our results suggest the opposite prognostic roles of LMTK2. In non‐metastatic (M0) patients, high LMTK2 appears to exert both pro‐proliferative and anti‐metastatic effects. The pro‐proliferative signature potentially promotes primary tumor growth and recurrence, which may lead to poorer RFS. In contrast, in metastatic (M1) patients, the pro‐proliferative capacity of LMTK2 seems attenuated, while its migration‐inhibitory function may become more prominent. This possible functional shift could restrict further dissemination of tumor cells and thus translate into a favorable prognostic factor. These results indicate that the functional impact of LMTK2 in primary lesions may vary according to the metastatic status of the patients. This phenomenon has been documented for other genes as well. For example, in pancreatic tumor cells, YTHDF2 has been shown to accelerate cell proliferation while suppressing migration and invasion [40]. The underlying mechanisms are not clear and warrant further investigation.

The results that LMTK2 restrains colon cancer cell metastasis are well in agreement with a recent report demonstrating an inhibitory role of LMTK2 on glioblastoma cell metastasis [21]. However, Zhao et al. [17]. demonstrated in hepatocellular carcinoma cells that LMTK2 was in favor of cell invasion. The opposing effects suggest that LMTK2 may function as either an anti‐metastatic or a pro‐metastatic regulator in a context‐dependent manner.

We found that LMTK2 restrained the interaction between p38 and MKK3/6, leading to suppression of p38 phosphorylation and subsequent inhibition of migration and invasion of colon cancer cells (Figures 5, 6, 7). Interestingly, the kinase‐deficient LMTK2 (K168R) mutant also inhibited phosphorylation of p38, migration, and invasion of colon cancer cells, indicating a kinase‐independent function of LMTK2 in controlling cancer cell metastasis. These results reveal a kinase‐independent function of LMTK2, broadening our understanding of this enzyme.

Interestingly, our results showed the rescue effect of SB203580 on LMTK2 mRNA levels following LMTK2 knockdown (Figure 5E), raising a possibility of a negative feedback loop within the LMTK2‐p38 axis. While LMTK2 suppresses p38 phosphorylation, p38 may in turn negatively regulate LMTK2 expression. This reciprocal regulation could serve to fine‐tune p38 signaling activity. The underlying molecular mechanism–whether transcriptional or post‐transcriptional–awaits further investigation. But such a feedback circuit would further reinforce the functional relevance of LMTK2 in modulating p38‐driven metastatic phenotypes.

While our study establishes a novel kinase‐independent, anti‐metastatic role for LMTK2 in colon cancer via p38 pathway inhibition, several questions remain unanswered. Mechanistically, the exact structural basis for LMTK2‐mediated disruption of the p38–MKK3/6 interaction requires further biochemical and structural analysis. Additionally, although we observed a correlation between LMTK2 expression and metastasis, the upstream effectors regulating its expression were not explored. Future work addressing these points will provide a more comprehensive understanding of LMTK2 in metastasis and may inform new therapeutic strategies targeting this pathway.

In this study, we demonstrate that LMTK2 functions to inhibit migration and invasion of colon cancer cells through the inhibition of p38 phosphorylation. Our findings reveal a previously unrecognized function of LMTK2 and the underlying molecular mechanism, highlighting the role of LMTK2 in restraining the metastatic potential of colon cancer.

Author Contributions

Jie Lun: methodology, data curation, validation, conceptualization, investigation, formal analysis, visualization, writing – original draft. Jinzhao Zhang: writing – original draft, methodology, data curation, conceptualization, investigation, validation, visualization, formal analysis. Jianxin Xu: visualization, formal analysis. Liyuan Jing: data curation, validation. Yanqing Qin: data curation, writing – review and editing. Jing Fang: funding acquisition, writing – review and editing, conceptualization, supervision, project administration, resources, methodology. Mengchao Yu: funding acquisition, writing – review and editing, project administration, conceptualization, methodology, supervision, resources.

Funding

This study was supported by the Natural Science Foundation of Qingdao (24‐4‐4‐zrjj‐160‐jch), the Shandong Provincial Medical and Health Science and Technology Project (202404080956), the Ministry of Science and Technology of China (2020YFA0803300), and the National Natural Science Foundation of China (82073061).

Ethics Statement

This study involving human cancer tissues was approved by the Ethics Committee of Qingdao University Medical College (approval no. QDU‐HEC‐2025555). This study does not involve clinical trials (N/A). The animal experiments were approved by the Ethics Committee of Qingdao University Medical College (approval no. QDU‐AEC‐2025812).

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1: The function of LMTK2 in different M‐stage patients in GSE39582.

Figure S2: p38 inhibitor alone suppresses basal migration, invasion, and pro‐metastatic gene expression.

Figure S3: p‐p38 accumulates at the invasive front upon LMTK2 knockdown.

Figure S4: p38 inhibition suppresses liver metastasis in vivo.

Figure S5: The K168R mutant is kinase‐dead but retains p38 binding capacity.

Figure S6: Validation of MKK3/6 knockdown efficiency.

Figure S7: LMTK2 disrupts the p38‐MKK3/6 complex and binds p38 independently of its phosphorylation status.

Table S1: Primer and siRNA sequences used in this study.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.