MAMDC2 suppresses gastric cancer cells but promotes progression through cancer associated fibroblasts

Gastric cancer (GC) remains a leading cause of cancer-related mortality worldwide, and novel biomarkers and therapeutic targets are urgently needed. MAMDC2 has been implicated in several malignancies, but its biological function and cellular origin in GC remain largely unknown. We integrated bulk RNA-seq data from GEO and TCGA databases to evaluate the diagnostic and prognostic value of MAMDC2 in GC. Single-cell RNA sequencing and spatial transcriptomics were used to delineate its cellular origin within the tumor microenvironment. In vitro functional assays, including CCK-8, colony formation, flow cytometry, wound healing, and Transwell assays, were performed to assess the effects of MAMDC2 overexpression in GC cell lines. Conditioned medium (CM) experiments using MAMDC2-overexpressing fibroblasts were conducted to evaluate the paracrine effects of fibroblast-derived MAMDC2 on GC cell migration, invasion, and EMT-like changes. Western blotting was used to explore the underlying signaling pathways. MAMDC2 was downregulated in GC tissues and cell lines and showed diagnostic value. Clinically, high MAMDC2 expression was associated with poor overall survival and independently predicted adverse prognosis. Single-cell and spatial transcriptomic analyses consistently revealed that MAMDC2 is predominantly expressed in cancer-associated fibroblasts (CAFs) rather than in malignant epithelial cells. Functional experiments showed that forced MAMDC2 overexpression in GC cells suppressed proliferation, migration, and invasion while promoting apoptosis. Notably, conditioned medium from MAMDC2-overexpressing CAFs enhanced MGC-803 cell migration and invasion, accompanied by EMT-like marker changes (E-cadherin downregulation, Vimentin and N-cadherin upregulation), suggesting that CAF-derived MAMDC2 may promote GC progression in a paracrine manner. These paracrine findings are proof-of-concept, as they were obtained using one commercial CAF line and one recipient GC cell line. Western blotting further showed that MAMDC2 overexpression was associated with altered MAPK pathway activation, including decreased p-JNK and p-p38 and increased p-ERK, although these changes are correlative and require functional validation. MAMDC2 plays context-dependent roles in GC. At the cancer cell-autonomous level, MAMDC2 overexpression suppresses proliferation, migration, and invasion. However, MAMDC2 is predominantly expressed in CAFs, and CAF-derived MAMDC2 may promote GC cell migration, invasion, and EMT-like changes through paracrine signaling, offering a plausible explanation for the association between high MAMDC2 expression and poor prognosis. These findings suggest that MAMDC2 may act as a context-dependent regulator and a potential stromal-derived modulator of the GC microenvironment, rather than a classical cell-intrinsic tumor suppressor. Further validation using additional GC cell lines, primary CAFs, and in vivo models is warranted.