Enhanced survival in perineural invasion of pancreatic cancer: an in vitro approach

Summary

Pancreatic cancer (PanCa) is characterized by perineural invasion (PNI), early lymph node and liver metastasis, and poor prognosis. PNI is one of the important causes of local recurrence. Little is known about the mechanism of PNI in PanCa. We presented a novel model system that may shed light on the mystery of PNI in PanCa. In this study, mouse dorsal root ganglia (DRGs) and human PanCa cell line (MIA PaCa-2) were cocultured in Matrigel matrix (BD Biosciences, San Jose, CA) to build this PNI model. MIA PaCa-2 cell line alone (control 1) or DRG alone (control 2) was cultured with Matrigel matrix as controls. Neurite outgrowth, cell colony growth, neurite-colony contact, and retrograde extension were observed under inverted microscopy and then were photographed and quantitated with the Optimas imaging system (Optimas Corp., Bothell, MA). At day 14, both the experimental and control 2 samples were harvested and subjected to total RNA isolation and fixed in paraffin-embedded blocks. Slides cut from paraffin blocks were studied with Ki-67 immunostaining and TUNEL assay. Gene profiling was performed using complementary DNA microarray. Overexpressed target genes were verified by quantitative reverse transcriptase polymerase chain reaction. The results showed that reciprocity was observed between neurites and MIA PaCa colonies with 24 hours of coculture. Neurite outgrowth was stimulated in the presence of pancreatic carcinoma cells, which showed 2-fold more area than did control 2. After 72 hours, MIA PaCa colonies cocultured with DRG exhibited 58% more colony area than did control 1. The Ki-67 index of the DRG/MIA PaCa cells (mean, 5.02%) was significantly higher than that in control 1 (mean, 1.18%) (P < .05); in contrast, the apoptotic index in the DRG/MIA PaCa cells was significantly lower (mean, 0.45%) than that in the control 1 (mean, 1.85%) (P < .001). Prosurvival genes MALT1 and TRAF were increased 2-fold in DRG/MIA PaCa compared with controls. We demonstrated that neural-epithelial interaction is a mutually beneficial process for the growth of nerves and PanCa cells. It is possible that oncogenes and growth factors might act synergistically in promoting proliferation and/or inhibiting apoptosis, a survival strategy crucial to the development of PNI in PanCa.

Introduction

In the United States, pancreatic cancer (PanCa) ranks as the 9th or 10th most commonly diagnosed cancer, but is the 4th or 5th leading cause of cancer death [1], [2]. The average 5-year survival of PanCa has been reported to be only 10% to 25% even after curative resection [3]. The dismal prognosis has been linked to local recurrence, lymph node metastasis, liver metastasis, peritoneal dissemination, and perineural invasion (PNI) [4], [5], [6], [7], [8]. PanCa is characterized by extremely high frequency of PNI (can be as high as 90%, even 100%) [5], [9]. Initial intrapancreatic PNI may lead to invasion of extrapancreatic nerve plexus, which is believed to be an important cause for positive surgical margin, and continuous extrapancreatic spread [5], [7], [10]. Recognition of the pathway of extrapancreatic spread and its significance in relation to PanCa progression has implications in surgical treatment of PanCa such as en bloc resection of the retropancreatic tissue involving the nerve plexus and fat tissue [7], [10]. PNI has been associated with poorer survival. It has been reported that the 5-year survival for patients with nodal metastasis but without PNI was 75%. In contrast, only 29% survival was reported in those with both nodal metastasis and PNI. Remarkably, patients with stage I disease without PNI had the most favorable 5-year survival of 89% [11]. Better understanding of the perineural molecular mechanisms could bring about effective control of PNI-related morbidities such as local recurrence and extrapancreatic spread.

It is generally believed that the vectorial balance between cell proliferation and cell death regulates net tumor progression. In PanCa, many genes and/or proteins might be involved in the process of promoting tumor growth while suppressing apoptotic mechanisms. The interactions between these molecules may create a favorable microenvironment for tumor cells to grow aggressively and neurotropically, a phenomenon also observed in other malignancies such as squamous cell carcinoma of the head and neck [12], bile duct carcinoma [13], and prostate carcinoma [14]. Indeed, our preliminary in vitro study of PNI in prostate cancer has demonstrated mutual communications between nerves and cancer cells, which might cause extensive and reciprocal nerve-epithelial interactions [15]. Furthermore, another study in human prostate specimens has shown that PNI is intimately associated with decreased apoptosis and increased proliferation [16]. Similar results have been observed as well in an in vitro study of PNI in prostate cancer [17]. To our knowledge, the in vitro study of proliferation and apoptosis and their associations with PNI in PanCa has not yet been reported.

To examine the interactions between nerve and cancer cells, we created a novel in vitro model system, which was applied to observe the growth of both nerve and PanCa cells. By using this coculturing model system of nerve and PanCa cells, we analyzed the proliferating activity and apoptotic rate of PanCa cells in this model system. We believe that a better understanding of molecular events associated with nerve-epithelial interaction could shed light on the mechanisms of PNI in PanCa.