Engineered mesenchymal stem cells expressing CXCR4 and LIGHT enhance Anti-Tumor activity in a subcutaneous gastric cancer xenograft model

Abstract

Gastric cancer remains a leading cause of cancer-related mortality, with limited treatment options and a poor prognosis. Mesenchymal stem cells (MSCs) possess inherent tumor-homing ability but exhibit modest efficacy against solid tumors. Enhancing both tumor targeting and immune activation through genetic modification may improve the therapeutic potential of these approaches. To evaluate the anti-cancer efficacy of human umbilical cord–derived MSCs (HUC-MSCs) genetically modified to co-express C-X-C chemokine receptor type 4 (CXCR4) and TNFSF14 (LIGHT) against gastric cancer in vitro and in vivo. HUC-MSCs were transduced via lentiviral vectors to generate dual-gene–engineered Tg-HUC-MSCs. In vitro, their effects on HGC-27 gastric cancer cells were assessed using the scratch wound migration assay and Annexin V/PI apoptosis analysis. In vivo, BALB/c nude mice bearing subcutaneous HGC-27 xenografts were treated with Tg-HUC-MSCs, unmodified HUC-MSCs, or a control. Tumor growth was monitored by caliper measurements. Excised tumors were analyzed histologically with hematoxylin–eosin staining and by Ki-67 immunohistochemistry. Tg-HUC-MSCs significantly inhibited HGC-27 cell migration and induced higher apoptosis rates than unmodified HUC-MSCs in vitro (p < 0.001). In vivo, Tg-HUC-MSCs markedly reduced tumor volume, increased necrotic areas, and decreased Ki-67 proliferation indices compared to controls (p < 0.001). Elevated CXCR4 and LIGHT expression correlated with improved tumor targeting and enhanced immune-mediated tumor cell killing. Dual-gene–engineered HUC-MSCs (CXCR4 + TNFSF14) demonstrated superior tumor-homing and anti-cancer activity against gastric cancer cells in vitro and in vivo, representing a promising cell-based therapeutic strategy for solid tumors.

Introduction

Cancer remains one of the leading causes of mortality across the globe, despite decades of intensive research and therapeutic advancements [1, 2]. While conventional approaches such as chemotherapy, radiotherapy, and targeted therapies have improved survival in certain malignancies, they are frequently associated with serious side effects, tumor resistance, and limited efficacy in advanced stages [3–5]. Gastric cancer, in particular, is a substantial therapeutic challenge owing to its aggressive course, late diagnosis, and poor five-year survival rate [6–8]. Gastric cancer ranks as the fifth most common malignancy and the third highest contributor to cancer-related deaths globally, highlighting the urgent need for more precise and effective treatment approaches [9–11]. Among its subtypes, poorly differentiated gastric carcinoma—represented by cell lines such as HGC-27—is particularly difficult to treat and often resistant to standard modalities.

In recent years, MSCs have gained attention as potential therapeutic carriers for cancer therapy due to their natural tumor-homing ability, low immunogenicity, and suitability for genetic modification [12]. Mesenchymal stem cells (MSCs), sourced from tissues such as adipose tissue, bone marrow, and umbilical cord, possess the ability to migrate toward tumor microenvironments (TMEs) and deliver therapeutic agents [13]. Genetic modification of MSCs permits the expression of therapeutic transgenes such as cytokines, pro-apoptotic genes, and immunomodulatory agents, enabling localized anti-tumor effects with reduced systemic toxicity compared with conventional cytokine therapies [14–16]. Preclinical studies have demonstrated that engineered MSCs can effectively migrate to tumor sites and deliver targeted therapeutic payloads, resulting in tumor growth inhibition in various solid tumor models.

HUC-MSCs are especially attractive owing to their non-invasive source, higher proliferation capacity, and robust immunomodulatory functions. Several preclinical studies have shown the utility of MSCs in delivering pro-apoptotic genes, cytokines, and oncolytic viruses directly to tumor sites [17]. However, clinical translation remains limited by its insufficient tumor targeting, risk of tumor promotion under certain conditions, and inadequate stimulation of anti-tumor immunity [18]. These limitations underscore the need for genetic enhancement of MSCs to fully realize their potential in cancer therapy.

One key avenue of improvement lies in enhancing the tumor-homing capabilities of MSCs through the overexpression of specific chemokine receptors. Among these, C-X-C chemokine receptor type 4 (CXCR4) has been widely studied for its role in the SDF-1/CXCR4 axis, which is instrumental in regulating cell migration and retention within the tumor microenvironment [17]. Many solid tumors, including gastric cancer, exhibit elevated levels of stromal cell-derived factor-1 (SDF-1), which attracts CXCR4-expressing cells to the tumor site. Engineering HUC-MSCs to overexpress CXCR4 has shown promise in improving their homing efficiency and persistence at the tumor location, thereby increasing the therapeutic payload delivered to malignant tissues.

While CXCR4 enhances localization, TNFSF14—also known as LIGHT (homologous to lymphotoxins, exhibits inducible expression)—acts as a potent immunomodulatory molecule. By competing with HSV glycoprotein D for Herpesvirus Entry Mediator (HVEM), a receptor expressed on T lymphocytes, LIGHT can effectively trigger anti-tumor immune responses [19]. LIGHT, a member of the TNF superfamily, binds to two key receptors—HVEM and Lymphotoxin Beta Receptor (LTβR)—thereby stimulating T cell activation, enhancing dendritic cell function, and promoting the destruction of tumor vasculature. It has also been shown to reshape the immunosuppressive tumor microenvironment and increase infiltration of cytotoxic T lymphocytes [20]. When expressed by MSCs, LIGHT has the potential to convert these otherwise immunosuppressive cells into agents of immune activation within the tumor niche. Thus, LIGHT represents a desirable therapeutic gene for inclusion in engineered MSC-based cancer therapies [20].

Based on these insights, we hypothesize that dual genetic modification of HUC-MSCs with CXCR4 and TNFSF14 can synergistically enhance tumor targeting and immunogenicity, thereby providing a potent therapeutic platform for gastric cancer. In this study, we established a lentiviral vector-based transduction system to co-express CXCR4 and TNFSF14 in HUC-MSCs and evaluated their anti-tumor effects against the HGC-27 gastric cancer cell line. This novel approach integrates the migratory advantage conferred by CXCR4 with the immune-stimulatory potential of LIGHT, potentially transforming MSCs into an efficient and targeted system for tumor immunotherapy. To our knowledge, this is among the first studies to investigate the combined therapeutic potential of CXCR4 and TNFSF14 in genetically engineered MSCs, specifically against gastric cancer.

Material and method

Lentiviral packaging

The CT core plasmid carrying the genes of interest (CXCR4 and TNFSF14) was used as the recombinant expression vector for lentiviral production (Fig. 1a). The construct was designed to express the genes under the regulation of this specific Elongation Factor 1 Alpha (EF1α) promoter. It included appropriate regulatory elements for enhanced transcription. Human embryonic kidney (HEK) 293 T cells were plated in T175 flasks and maintained in Dulbecco’s Modified Eagle Medium (DMEM; 22400-089, Gibco, USA) supplemented with 10% fetal bovine serum (FBS; 10091-148, Gibco, USA) under standard culture conditions (37 °C, 5% CO₂) [20]. For lentiviral packaging, 293 T cells were co-transfected with the CT core plasmid, along with the packaging plasmids pMDLg/pRRE, pRSV-Rev, and the envelope plasmid pMD2.G (Addgene, USA), with the help of polyethylenimine hydrochloride (PEI; 24765-1, Polysciences, USA) as the transfection reagent [21]. Lentiviral particle titers were determined by transduction of HEK293T cells and calculated to be 4.32 × 10⁸ transducing units (TU)/mL.

Fig. 1.

(a) Map of plasmid pEF1α-CXCR4-TNFSF14 showing the inserted CXCR4 and TNFSF14 genes with regulatory elements (b) Diagram showing HUC-MSCs engineered with lentiviral vectors carrying CXCR4 and TNFSF14 for tumor-targeted immunotherapy (c) Experimental timeline indicating the chronological sequence of procedures. Mice were first acclimatized for 7 days, followed by tumor cell inoculation (HGC-27) on Day 0. MSC treatments (unmodified or engineered Tg-HUC-MSCs) were administered on days 7, 11, and 15. Tumor growth was monitored using caliper measurements, and samples were collected at the end of the study on Day 34 for histological analysis

Lentiviral supernatants were collected at 24-, 48-, and 72-hours post-transfection. The pooled viral-containing medium was filtered and concentrated 100-fold using PEG6000 (528877-1KG, Millipore, USA) by centrifugation at 8000 × g for 15 min [22]. The resulting concentrated lentiviral particles were stored at − 80 °C until use for transduction of HUC-MSCs.

In vitro cell culturing and lentivirus-mediated transduction

HUC-MSCs were obtained from Milestone Biotechnologies (Shanghai, China) and processed in compliance with all required biosafety standards. The HUC-MSCs were cultured in low-glucose DMEM (22400-089, Gibco, USA) supplemented with 10% FBS and 1% penicillin-streptomycin (10091-148, Gibco, USA). Cells were kept at 37 °C in a humidified atmosphere containing 5% CO₂ [22]. When the cells reached approximately 70–80% confluence, lentiviral particles were introduced at a multiplicity of infection (MOI) of 10. After 24 h of transduction, the medium was exchanged with fresh complete medium, and the cells were further cultured to maintain an appropriate density. The culture medium was refreshed every 2 to 3 days to support optimal cell growth and transgene expression. HUC-MSCs used in all experiments were at passage 3 (P3) [22].

Flow cytometry analysis

Flow cytometry was conducted using a BD LSRFortessa cytometer, and the data were processed with FlowJo software (version 10) [23]. Transduced HUC-MSCs (Tg-HUC-MSCs) were stained with fluorophore-conjugated antibodies (CD90-FITC, CD73-PE, CD105-PerCp-Cy5.5, CD29-APC, CD45-PerCP-Cy5.5, CD34-FITC, CD14-APCH, APC-Cy7-HLA-DR, CD19LPE) targeting mesenchymal stem cell surface markers, including CD184 (CXCR4), CD258, CD44, CD105, CD29, and CD90, as well as negative markers CD38, CD117, CD45, and CD34 [24]. Isotype controls were included to set gates and validate specificity. Marker expression and transduction efficiency were quantified based on fluorescence intensity profiles.

Cytotoxicity assay

HGC-27 gastric cancer cells stably expressing luciferase were seeded into 96-well plates at a density of 5,000 cells per well and allowed to adhere for 24 h. Unmodified HUC-MSCs (positive control) or genetically engineered Tg-HUC-MSCs were then added at effector-to-target (E: T) ratios of 1:1, 5:1, and 10:1 [25]. Following co-culture, D-luciferin (100 µL, 150 µg/mL; Abcam, USA) was added to each well, and luminescence was measured using a microplate luminometer. Luciferase activity was recorded as relative light units (RLU). Target-only wells containing HGC-27 cells without MSCs were used to define the maximal luciferase signal (100% viability), while background luminescence accounted for less than 1% of the target-only signal. Tumor cell viability was calculated as (experimental RLU/maximal RLU) × 100, and percent cytotoxicity was expressed as 100 − viability (%) [26].

Scratch assay

To evaluate the effect of HUC-MSCs and Tg-HUC-MSCs on cancer cell migration, HGC-27 cells were added to 12-well culture plates and left for 24 h until approximately 90% confluence was achieved, forming a uniform monolayer [26]. A scratch was made in each well using a 200 µL pipette tip, followed by gentle washing with PBS to eliminate any detached cells. Transwell chambers containing either the + ve control (HUC-MSCs), experimental group (Tg-HUC-MSCs), or control (no MSCs) were placed above the scratched cancer cell layer. The co-cultures were incubated for 48 h, and images were captured at 0, 12, 24, and 36 h to monitor wound closure. The exposed wound area was quantified and analyzed using OLYMPUS cellSens Dimension 2.1 software (OLYMPUS, Japan) [27].

Apoptosis analysis

To evaluate apoptosis induction in cancer cells, HGC-27 cells were added to 12-well plates and left for 24 h. Transwell chambers containing the + ve control group (HUC-MSCs), the experimental group (Tg-HUC-MSCs), or the control group (no MSCs) were then placed above the cancer cell layer. After a further 24 h of co-culture, HGC-27 cells were harvested, and apoptosis was quantified using the Annexin V/PI staining protocol and flow cytometer [28].

Xenograft mouse model

BALB/c nude mice (aged 6–8 weeks) were purchased from Jiangsu Cyagen Biosciences Co., Ltd. (Shanghai, China) and maintained under specific pathogen-free conditions. For the xenograft model, HGC-27 gastric cancer cells were harvested and resuspended in sterile saline at a concentration of 1 × 10⁸ cells/mL. A total of 50 µL of this suspension, containing 5 × 10⁶ cells, was subcutaneously injected into the right flank of each mouse. Tumor formation was monitored and evaluated seven days after injection. Upon reaching an approximate tumor volume of 50 mm³, calculated using the formula (length × width²)/2, treatment began. Mice were randomly divided into three treatment groups: (1) control (no MSCs), (2) positive control (unmodified HUC-MSCs), and (3) experimental (Tg-HUC-MSCs). The engineered Tg-HUC-MSCs (CXCR4 + LIGHT) were administered through intratumoral injections at a dose of 1 × 10⁶ cells per mouse every four days for a total of three treatments (on days 7, 11, and 15). Tumor progression was assessed by measuring tumor size, body weight, and bioluminescent intensity every four days . The imaging data were captured using the IVIS system following intraperitoneal injection of D-luciferin (150 mg/kg; Abcam, USA) and analyzed using Living Image software (Caliper Life Sciences, USA). Mice were humanely sacrificed when tumor volumes exceeded 2000 mm³ or upon reaching other ethical endpoints, in accordance with guidelines approved by the Institutional Animal Care and Use Committee (IACUC) (Fig. 1c).

Histology and immunohistochemistry

At the experimental endpoint, tumors were surgically removed, and their size and weight were measured to evaluate differences among the three groups. Tumor volume was estimated using the formula (length × width²)/2. The removed tumor samples were fixed in 4% paraformaldehyde, embedded in paraffin, and sectioned at 5 μm. Hematoxylin and eosin (H&E) staining was then carried out to examine tumor morphology and necrosis [29]. Immunohistochemical (IHC) staining for Ki-67 was performed on tumor sections to evaluate cell proliferation. Positive cells of Ki-67 appeared as brown-stained nuclei, and the proliferation index was found by calculating the percentage of Ki-67–positive cells across five randomly chosen high-power fields for each sample.

Results

Verification of CXCR4 and LIGHT expressions in engineered HUC-MSCs

Flow cytometry confirmed stable cell surface expression of CXCR4 and LIGHT in Tg-MSCs after lentiviral transduction. Expression levels increased depending on dose, with Tg-MSCs transduced using 20 µL lentivirus showing the highest positivity (> 80%), significantly higher than both MSCs and untreated controls (p < 0.001) (Fig. 2). These data validated efficient dual-gene incorporation in engineered HUC-MSCs.

Fig. 2.

Flow cytometry analysis of LIGHT and CXCR4 surface expression in engineered HUC-MSCs. Representative flow cytometry dot plots showing cell surface expression of LIGHT (CD258/TNFSF14) and CXCR4 (CD184) in control MSCs and Tg-MSCs transduced with increasing lentiviral doses (CT-5, CT-10, CT-20). GFP-positive cells indicate successfully transduced MSCs. Upper panels display LIGHT (CD258) expression, and lower panels display CXCR4 (CD184) expression. Progressive increases in the percentage of GFP⁺/CD258⁺ and GFP⁺/CD184⁺ cells were observed with higher viral doses, confirming dose-dependent and stable surface expression of both transgenes

Tg-MSCs exhibit stronger cytotoxic effects against gastric cancer cells

Luciferase-based cytotoxicity assays demonstrated that Tg-MSCs induced a significant and dose-dependent reduction in luciferase signal in HGC-27 cells compared with unmodified MSCs and control groups. At effector-to-target ratios of 1:1, 5:1, and 10:1, Tg-MSC treatment resulted in progressive decreases in normalized luciferase activity, corresponding to tumor cell killing rates of approximately 20%, 60%, and greater than 70%, respectively. In contrast, unmodified MSCs produced only modest reductions in luciferase signal across all effector-to-target ratios. Statistical analysis confirmed that Tg-MSCs exhibited significantly greater cytotoxicity than MSCs or controls at all tested ratios (p < 0.001) (Fig. 3).

Fig. 3.

Luciferase-based cytotoxicity assay assessing the anti-tumor activity of Tg-MSCs against HGC-27 gastric cancer cells. HGC-27 cells stably expressing luciferase were co-cultured with unmodified MSCs or Tg-MSCs at effector-to-target ratios of 1:1, 5:1, and 10:1. Luciferase activity was measured as relative light units (RLU) and normalized to target-only controls (100% viability). Data are presented as percentage tumor cell killing derived from normalized luciferase signal (mean ± SD, n = 3). Tg-MSCs induced significantly higher cytotoxicity than unmodified MSCs at all ratios (p < 0.001, one-way ANOVA with Tukey’s post-test)

Tg-MSCs inhibit gastric cancer cell migration

Scratch wound assays revealed that Tg-MSCs markedly inhibited HGC-27 migration compared to MSCs and controls. At 12 h, wound closure in Tg-MSC–treated groups was reduced by ~ 65%, whereas MSC-treated cells showed only ~ 25% reduction (p < 0.001). At 24 h, wound closure remained incomplete in Tg-MSC groups, while control and MSC groups displayed near-complete closure (Fig. 4).

Fig. 4.

Scratch wound migration assay showing the effect of MSCs and Tg-MSCs on HGC-27 gastric cancer cell migration. Representative images were captured at 12 h and 24 h following scratch induction. Red dashed lines indicate wound boundaries. Tg-MSC treatment markedly delayed wound closure compared with control and MSC-treated groups, indicating inhibition of tumor cell migration

Tg-MSCs induce robust apoptosis in vitro

Annexin V/PI staining confirmed that Tg-MSCs significantly enhanced apoptosis of HGC-27 cells. Compared to MSCs and controls, Tg-MSCs increased both early and late apoptotic fractions (total apoptosis ~ 45% vs. ~18% in MSCs and ~ 10% in controls, p < 0.001). These findings indicate that Tg-MSCs strongly promote programmed tumor cell death (Fig. 5a and b).

Fig. 5.

(a) Flow cytometry plots of Annexin V/PI apoptosis assay showing increased apoptotic fractions in Tg-MSC–treated HGC-27 cells (b) Quantification of apoptosis percentages across control, MSC, and Tg-MSC groups. The apoptosis levels were quantified in the control, MHC, and Tg-MHC groups, with yellow representing control, red representing MHC, and green representing Tg-MHC. Data are presented as mean ± SD (n = 3). Statistical analysis was performed using one-way ANOVA with Tukey’s post-test, and significant differences were noted (p < 0.001)

In vivo Tumor suppression by Tg-MSCs

In BALB/c nude mice bearing HGC-27 xenografts, Tg-MSC treatment led to profound tumor suppression compared with both MSC-treated and control mice (Fig. 6a). Tumor volumes in Tg-MSC–treated animals were reduced by more than 60% at endpoint (p < 0.001) (Fig. 6b). Body weight changes in control, MSC-, and Tg-MSC-treated groups showed no significant differences (Fig. 6c). Importantly, MSCs alone showed minimal tumor growth inhibition, underscoring the superior efficacy of dual-gene–engineered MSCs.

Fig. 6.

(a) Representative tumor measurements across groups, confirming consistent tumor suppression by Tg-MSCs (b) Tumor growth curves showing significantly reduced tumor volumes in Tg-MSC–treated mice compared to controls and MSCs (p < 0.001) before and after taking out the outliers (c) Body weight changes in control, Ctrl-MSC, and Tg-MSC groups. No significant differences were observed, indicating that Tg-MSCs did not cause systemic toxicity. (d) Ki-67 immunohistochemistry of tumor xenografts from PBS-treated control, MSC-treated, and Tg–MSC–treated groups. Tumors from control animals exhibited extensive nuclear Ki-67 staining, consistent with high proliferative activity. MSC-treated tumors showed reduced and more heterogeneous Ki-67 expression. Tg-MSC–treated tumors demonstrated markedly diminished Ki-67 positivity, with large regions lacking nuclear staining. Scale bars: 200 µM (e) Bar graph showing the percentage of Ki-67–positive proliferating cells in tumor tissues from the control, MSC-treated, and Tg-MSC–treated groups. Data are presented as mean ± SD. Tg-MSC treatment resulted in a significant reduction in Ki-67–positive cells compared with unmodified MSCs, indicating decreased tumor cell proliferation. Statistical significance was determined using one-way ANOVA (p< 0.01) (f) Representative H&E-stained sections of the heart, kidney, liver, lung, and spleen across the three treatment groups. All organs displayed preserved tissue architecture with no histopathological abnormalities observed. Scale bars: 100 µM

Histological analysis reveals reduced proliferation and enhanced necrosis

Examination of main organs (heart, liver, kidney, spleen, and lung) using H&E staining showed no abnormal architecture or pathological lesions in any treatment group. Cardiac muscle fibers remained intact, hepatic lobular organization was preserved, renal glomeruli and tubules appeared normal, and lung alveoli, as well as splenic follicles, showed no structural alterations (Fig. 6f). These findings indicate that both MSC and Tg-MSC administration were well tolerated in vivo and did not induce detectable off-target tissue damage.

Ki-67 immunohistochemistry confirmed these findings. Control tumors exhibited strong and widespread nuclear positivity with a proliferation index of ~ 75%. MSC-treated tumors showed patchier Ki-67 expression, reduced to ~ 55%. Tg-MSC tumors demonstrated markedly diminished Ki-67 staining, with proliferation reduced to below 20% (p < 0.001) (Fig. 6d and e). These Findings indicate that Tg-MSCs exert superior anti-proliferative and cytotoxic effects compared with MSCs or controls.

Discussion

In contrast to conventional systemic therapies for gastric cancer, Tg-HUC-MSCs aim to enhance anti-tumor activity through localized tumor targeting. In this study, we demonstrated that dual-gene–engineered HUC-MSCs co-expressing CXCR4 and LIGHT significantly enhanced anti-tumor efficacy against gastric cancer, both in vitro and in vivo. Compared with unmodified HUC-MSCs, Tg-HUC-MSCs displayed superior tumor homing, induced robust apoptosis, inhibited migration, and suppressed tumor proliferation. These findings provide quantitative and qualitative evidence that genetic modification of HUC-MSCs with a chemokine receptor (CXCR4) and a costimulatory ligand (LIGHT) can simultaneously improve tumor targeting and immune-mediated tumor cell killing.

Recent studies have demonstrated the therapeutic potential of human umbilical cord–derived mesenchymal stem cells (HUC-MSCs) in solid tumor models. HUC-MSC-derived exosomes have been engineered to deliver antisense oligonucleotides targeting oncogenic microRNAs, resulting in reduced tumor growth and proliferation in colorectal cancer xenografts [30]. In addition, HUC-MSC-conditioned media has been shown to inhibit tumor cell proliferation, migration, and invasion while promoting apoptosis through modulation of the IL-6/JAK2/STAT3 signaling pathway in glioblastoma models [31]. These findings support the use of HUC-MSCs as functional and modifiable platforms for anti-tumor therapy in solid cancers.

Normally, HUC-MSCs exhibit intrinsic tumor tropism but limited efficacy, primarily due to insufficient tumor infiltration and weak immune activation [30]. By overexpressing CXCR4, we improved MSC migration toward CXCL12-rich tumor microenvironments, consistent with reports that CXCR4/CXCL12 signaling is critical for HUC-MSC trafficking in colorectal cancer [32]. LIGHT (TNFSF14) interacts with HVEM and LTβR receptors expressed on T cells, NK cells, and tumor cells, thereby enhancing T-cell proliferation, cytotoxicity, and cytokine release, while also promoting tumor cell apoptosis [33]. Prior studies have shown that LIGHT enhances infiltration of CD8 + T cells and NK cells into tumor sites, mainly reshaping the tumor microenvironment into a more immunogenic state [19]. Together, the dual expression of CXCR4 and LIGHT in MSCs provides both efficient tumor localization and potent immune activation, explaining the superior anti-tumor outcomes observed in our models.

Our in vitro findings support this mechanistic synergy: Tg-HUC-MSCs inhibited gastric cancer cell migration, promoted apoptosis, and exhibited stronger cytotoxicity than unmodified MSCs. The observed effects on tumor cell behavior in the Transwell co-culture model are likely due to paracrine signaling, as the MSCs and tumor cells did not directly interact in this setup. In vivo, xenograft models further validated these effects, with Tg-MSCs markedly suppressing tumor growth and reducing Ki-67 proliferation indices. Histopathological evidence of increased necrosis corroborated the functional data, highlighting the ability of Tg-MSCs to remodel tumor tissue architecture.

While genetically engineered mesenchymal stem cells (MSCs) offer significant therapeutic promise, their safety and potential off-target effects remain important considerations. Previous studies have shown that MSC-based therapies can display context-dependent immunomodulatory behavior, which may vary according to the tumor microenvironment and host immune status, highlighting the need for careful evaluation of immune-related effects [34]. In addition, systemic administration of MSCs has been associated with rare but notable adverse events, including ectopic tissue accumulation, thromboembolic complications, and unintended pro-tumorigenic signaling under specific conditions [35]. Nevertheless, advances in genetic engineering strategies, including controlled transgene expression and tumor-targeted delivery approaches, have substantially improved the safety profile of engineered MSCs in preclinical studies [36]. Importantly, engineered MSCs have been successfully used as targeted cellular vehicles for the delivery of therapeutic genes in solid tumor models, achieving effective tumor suppression with minimal systemic toxicity, supporting their translational potential when appropriately optimized [14].

In the present study, the primary focus was to establish proof-of-concept evidence that CXCR4- and LIGHT-engineered human umbilical cord–derived MSCs enhance tumor targeting and suppress tumor growth in vivo. Accordingly, interactions with the native gastric tumor microenvironment, immune-related effects in immunocompetent settings, dose optimization, and broader model generalizability will be addressed in future studies using more physiologically relevant systems, such as orthotopic or patient-derived xenograft models with extended follow-up. Nevertheless, our findings provide an experimental foundation supporting the feasibility of engineered HUC-MSCs as targeted anti-tumor agents and justify further safety- and translation-oriented research.

Conclusion

Dual-gene–engineered HUC-MSCs expressing CXCR4 and LIGHT demonstrated enhanced tumor tropism and significant tumor suppression compared to unmodified HUC-MSCs in gastric cancer models. This strategy enhances tumor targeting and may reduce immune evasion, laying the groundwork for future studies on cell-based therapies for solid tumors. Our findings suggest that engineering HUC-MSCs with CXCR4 and LIGHT may provide a novel approach to gastric cancer immunotherapy. However, further studies in immunocompetent models and long-term safety assessments are necessary to fully evaluate the immune activation and therapeutic potential of this strategy, as well as to explore combination therapies.

Author contributions

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Umer Anayyat, Xumin Wu, Zhiyu Liu, Cui Liao. The first draft of the manuscript was written by Umer Anayyat and revised by Muyun Liu, Xiaomei Wang and Jianbin Ye. All authors commented on previous versions of the manuscript. All authors read and approved of the final manuscript.

Funding

This work was supported by Innovation Capacity Building Project, National Engineering Research Center of Foundational Technologies for CGT Industry(NDRC-High-Technology[2023]No. 447 to Muyun Liu) and Shenzhen Non-invasive Quality Online Monitoring and Analysis Platform (F-2022-Z99-502233 to Muyun Liu).

Data availability

The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors have no relevant financial or non-financial interests to disclose.

Ethics approval

Human umbilical cord-derived MSCs were gratis provided by Shenzhen Beike Biotechnology Co., Ltd. Ethical approval and informed consent for this study were obtained from Safety and Ethics Committee at Shenzhen Beike Biotechnology Co., Ltd (approval number: BK-SL-202403 14 − 01 & BK-DWLL-2025-0013)According to the manufacturer, the MSCs were isolated and cultured from the Wharton’s ielly of umbilical cords of maternal volunteers.

Consent to participate

Informed consent was obtained from all individual participants included in the study.