Colorectal cancer (CRC) remains one of the most prevalent and deadly malignancies globally, with a dismal 5-year survival rate of only 14% for patients with stage IV disease.

The tumour microenvironment (TME), particularly cancer-associated fibroblasts (CAFs)—the primary producers of extracellular matrix (ECM) components—plays a pivotal role in CRC progression, metastasis, and treatment resistance.

However, the molecular mechanisms underlying CAF-mediated ECM remodelling, especially collagen production, have long been incompletely understood.

Collagen constitutes ~90% of the ECM, and its excessive accumulation creates physical barriers to immune cell infiltration and drug delivery, fostering a protumorigenic TME.

Glycine is the most abundant amino acid in collagen (critical for its triple-helix structure and stability), but how CAFs acquire sufficient glycine to support collagen synthesis in CRC has remained unclear.

To address this gap, the research team isolated primary CAFs and normal fibroblasts (NFs) from surgically resected CRC tissues and paired normal adjacent tissues (NATs).

Metabolomic analyses revealed that CAFs exhibit significant activation of amino acid metabolism—with glycine levels 1.82-fold higher in CAFs and 1.45-fold higher in CAF-derived conditioned media (CAF-CM) compared to NFs and NF-CM.

Further investigation showed that this elevation stems from enhanced de novo glycine synthesis (not exogenous uptake), as CAFs displayed upregulated expression of key enzymes in this pathway: PHGDH, phosphoserine aminotransferase 1 (PSAT1), phosphoserine phosphatase (PSPH), and serine hydroxymethyltransferase 2 (SHMT2).

The team then explored the driver of this metabolic reprogramming.

They found that conditioned media from aggressive SW480 CRC cells (SW480-CM) recapitulated the activation of de novo glycine synthesis in CAFs, leading to increased glycine, proline, and alanine (all collagen-building amino acids) production.

Subsequent experiments identified transforming growth factor β1 (TGF-β1)—secreted at higher levels by CRC cells than CAFs—as the key signalling molecule.

Exogenous TGF-β1 treatment mimicked the effects of SW480-CM, upregulating de novo glycine synthesis enzymes and collagen production in CAFs; conversely, blocking TGF-β1 signalling with the TGF-β receptor I (TGF-βR1) inhibitor SB431542 or a neutralising anti-TGF-β1 antibody abrogated these effects.

Most importantly, the study validated PHGDH as a actionable target.

Both genetic knockdown of PHGDH (via siRNA) and pharmacological inhibition (using the specific inhibitor NCT503) significantly reduced TGF-β1-induced collagen I and collagen IV production in CAFs—effects confirmed by Western blot and immunofluorescence analyses.

Clinical relevance was further supported by tissue studies: Masson’s trichrome and immunohistochemical staining showed markedly higher collagen I/IV levels, as well as elevated PHGDH and CAF marker α-SMA expression, in human CRC tissues compared to NATs.

Publicly available datasets (PRJNA717755, PRJNA319481) also revealed positive correlations between TGF-βR1, de novo glycine synthesis enzymes, and collagen genes in CRC tissues and CAFs.

In summary, this study not only highlight the pivotal role of de novo glycine synthesis in CRC progression but also suggest a potential strategy for treating CRC by inhibiting PHGDH and provide a strong rationale for further study of PHGDH inhibition.

See the article, in MedComm – Oncology.

Source: Sichuan International Medical Exchange and Promotion Association