Apparently, PMNs are attracted by the tumor cells via the chemokine-receptor axis CXCL16-CXCR6 [33]. In PDAC, PMN infiltration was associated with a distinct JQ1 clinical trial “micropapillary” growth
pattern, compatible with a PMN-mediated dispersal of the tumor cells [7]. Analyzing 112 pancreas biopsies, we found that prominent PMN infiltrates coincided with low E-cadherin expression, and since one single PMN contains about 1 pg elastase [34], it is possible that the infiltrating PMNs are responsible for the loss of E-cadherin. Loss of E-cadherin induces a more migratory phenotype, and an association between this migratory phenotype, evident as up-regulation of the pancreatic serine protease PRSS3 by pancreas tumor cells, and the occurrence of distant organ metastasis has been described by others [35], as has an epithelial-to-mesenchymal transition, which is also associated with enhanced migration and the generation of metastasis [36]. Although a cause-effect-chain cannot be established BGB324 conclusively, and some evidence is still correlative, our data support the concept that prominent PMN infiltrates favor the invasive growth and metastasis of PDAC cells. In our patients, we could not correlate the PMN infiltrate or the relative E-cadherin loss to any of the clinical
and pathological parameters. A study by Hong et al. with considerably more patients, however, showed that loss of E-cadherin was associated with poorer prognosis [37]. These authors also suggested that the microenvironment might affect the local E-cadherin expression, a presumption that perfectly fits into our concept that infiltrating PMNs degrade E-cadherin. In conclusion, we found that PMNs via elastase degrade E-cadherin on pancreatic tumor cells, resulting in an enhanced dyshesion, migration, and invasiveness of the tumor
cells, which — in turn — could contribute to tumor progression, metastasis, and poorer prognosis in PDAC. Peripheral blood from healthy human volunteers was obtained by puncture of peripheral veins and collected in heparin-NH4-coated check details tubes (Sarstedt, Nürnbrecht, Germany). PMNs were isolated by centrifugation on PolymorphPrep (Axis-Shield PoC AS, Oslo, Norway) which yielded an 85–95% pure PMN population. The PMNs were suspended in Hanks balanced salt solution and used within 1 h. Four human pancreatic cancer cell lines were used: MiaPaca-2, Su8686 (ATCC, Rockville, MD, USA), COLO-357, and T3M4 (provided by the European Pancreas Center, Heidelberg, Germany). Cells were grown in RPMI-1640 medium containing 10% fetal bovine serum, 100 U/mL penicillin, and 100 μg/mL penicillin-streptomycin (Invitrogen, Karlsruhe, Germany) and were incubated at 37°C in a 5% CO2 humidified atmosphere. T3M4 (5 × 104) were plated in specialized cell culture dishes with a mobile insert in one compartment (ibidi, Martinsried, Germany) for 24 h.