Pancreatic cancer (PC) is highly lethal, with a five-year survival rate of ~10%1. PC is characterised by robust stromal activation, leading to pro-tumourigenic extracellular matrix deposition. Recently, we have shown that targeting desmoplasia can improve chemotherapy efficacy and impair metastasis in preclinical models2-4. As such, we aimed to use proteomics to dissect the matrisomal signatures of pancreatic tumours derived from the highly-metastatic KPC (Pdx1-Cre; LSL-K-rasG12D/+; LSL-p53R172H/+) and poorly-metastatic KPflC (Pdx1-Cre; LSL-K-rasG12D/+; LSL-p53fl/+) mouse models. We hypothesised that these tumours would have distinct matrisomes, with these changes revealing key matrix proteins involved in PC progression and spread.
Pancreatic tissue from wildtype, KPC, and KPflC mice were collected at early (6-weeks), mid (10-weeks) and end-stage disease (12+ weeks) and were decellularised5. Data-independent acquisition liquid-chromatography tandem mass spectrometry (LC-MS/MS) was used to identify differentially abundant proteins.
LC-MS/MS revealed an increased abundance of Nidogen-2 (NID2) in KPC tumours at mid-stage disease compared to KPflC. Interrogation of single cell-RNASeq murine and human PC datasets revealed that NID2 is enriched in PC tumours, especially in stromal cell populations. Immunofluorescence and western blotting show that NID2 is significantly enhanced in KPC cancer-associated fibroblasts (CAFs) compared to cancer cells.
3D organotypic matrices generated with NID2 knockdown (KD) CAFs had reduced desmoplasia, shown via second harmonic generation (SHG) imaging and Picrosirius Red/birefringence analysis. These NID2 KD matrices were also significantly softer, as shown via rheometer analysis. 3D invasion assays revealed that depletion of CAF-derived NID2 significantly impeded the invasion of cancer cells and enhanced standard-of-care Gemcitabine/Abraxane chemotherapy.
Subcutaneous and orthotopic co-seeding experiments using NID2 KD CAFs in combination with cancer cells show that reduced CAF-derived NID2 significantly impeded tumour growth and promoted vascular remodelling, thereby enhancing the efficacy of Gemcitabine/Abraxane. Strikingly, in orthotopic models, mice bearing NID2 KD tumours had significantly less metastasis to the liver and had increased survival.
In conclusion, temporally resolved LC-MS/MS proteomics of clinically relevant mouse models of pancreatic cancer has revealed NID2 as a new stromal co-target in this aggressive disease. Future work involves pharmacologically targeting NID2 using domain-specific blocking antibodies to effectively target NID2 in the pancreatic tumour microenvironment.