Mononuclear cells modulate the activity of pancreatic stellate cells which in turn promote fibrosis and inflammation in chronic pancreatitis

Christoph W Michalski 1 2 Andre Gorbachevski 1 Mert Erkan 2 Carolin Reiser 2 Stefanie Deucker 1 Frank Bergmann 0 Thomas Giese 4 Markus Weigand 3 Nathalia A Giese 1 Helmut Friess 2 Jrg Kleeff 2 0 Institute of Pathology, University of Heidelberg , Germany 1 Department of General Surgery, University of Heidelberg , Germany 2 Department of Surgery, Technische Universitat Munchen , Munich , Germany 3 Department of Anaesthesiology, University of Heidelberg , Germany 4 Institute of Immunology, University of Heidelberg , Germany Background: Interactions between mononuclear cells and activated pancreatic myofibroblasts (pancreatic stellate cells; PSC) may contribute to inflammation and fibrosis in chronic pancreatitis (CP). Methods: Markers of fibrosis and inflammation were concomitantly analysed by immunohistochemistry in chronic pancreatitis tissues. In vitro, PSC were stimulated with TNFalpha and LPS. Primary human blood mononuclear cells (PBMC) and PSC were cocultured, followed by analysis of cytokines and extracellular matrix (ECM) proteins. PBMC were derived from healthy donors and CP and septic shock patients. Results: In areas of mononuclear cell infiltration in chronic pancreatitis tissues, there was decreased immunoreactivity for collagen1 and fibronectin, in contrast to areas with sparse mononuclear cells, although PSC were detectable in both areas. LPS and TNFalpha induced collagen1 and fibronectin levels as well as the matrix degradation enzyme MMP-1. Coculture experiments with PSC and PBMC revealed increased fibronectin secretion induced by PBMC. In addition, donor and CP PBMC significantly induced an increase in IL-6, MCP-1 and TGFbeta levels under coculture conditions. Determination of the source of cytokines and ECM proteins by mRNA expression analysis confirmed PSC as major contributors of ECM production. The increase in cytokine expression was PBMC- and also PSC-derived. Conclusion: Mononuclear cells modulate the activity of pancreatic stellate cells, which may in turn promote fibrosis and inflammation. - Background The exact pathogenesis of chronic pancreatitis (CP) is unknown, but the disease results from an injury to a pancreatic cell population, which then activates pancreatic stellate cells (PSC) as a final common response. This process is associated with excessive accumulation of extracellular matrix (ECM) proteins by PSC and a massive infiltration of mononuclear cells [1-6]. Clinically, CP is accompanied by a severe pain syndrome and by a loss of pancreatic function (both exocrine and endocrine). PSC have been determined as the major source of ECM protein production, and have also been shown to produce cytokines and chemokines [7-11]. Thus, anti-fibrogenic therapies aim to reduce the activity of PSC in order to inhibit the accumulation of ECM proteins and also to prevent digestion of the basement membranes, which allows PSC to enter inflammatory areas and to excessively deposit ECM. In liver cirrhosis, tissue macrophages derived from blood mononuclear cells have been shown to influence fibrogenesis by activating hepatic myofibroblasts [12], while suppression of macrophage infiltration inhibits hepatic stellate cell activation and thus liver fibrogenesis. During fibrogenesis, local inflammation may not be initiated by apoptosis/necrosis of parenchymal cells but by resident and recruited inflammatory cells; these inflammatory cells, once in an active state, release cytokines which in turn activate stellate cells [13,14]. Recent studies have investigated the response of hepatic stellate cells to lymphocyte subsets, which can be divided into Th1 and Th2 predominant [15]. Different Th1 or Th2 subsets seem to modulate the fibrotic response towards anti- or pro-fibrogenesis, thus resulting in an apparent IL10 paradox: IL-10 is anti-fibrogenic but is produced during a pro-fibrotic Th2 immune response [16]. Mononuclear cells within fibrotic areas have been shown to secrete a wide variety of agents which induce matrix generation/ degradation [17-19], but there are no data on different responses of PSC to PBMC from various sources i.e., healthy donors and CP patients. In CP, a generalized immune response has been suggested due to increased expression of TNFalpha and its receptor on PBMC and which was associated with PSC cytotoxicity [20]. Furthermore, it has been shown that LPS-activated macrophages stimulate the synthesis of ECM proteins by PSC [21]. In rats, suppression of macrophage infiltration inhibited activation of hepatic stellate cells [12]. Similarly, it has been shown that both pro- and anti-inflammatory cytokines, such as PDGF and TGFbeta, activate PSC and thus contribute to pancreatic fibrogenesis [22]. TGFbeta was subsequently confirmed as a key regulator of ECM production and PSC proliferation due to its inhibition of MMPs in an autocrine fashion which enhanced fibrogenesis by reducing collagen degradation [23]. In terms of transcriptional regulation, nuclear factor kappaB (NFkB) is a well-characterized transcription factor which is activated by oxidative stress or by TNFalpha, and which activates key anti-fibrotic genes such as MMP-2. Interestingly, inhibition of NFkB by IkB has been shown to decrease IL-6 and ICAM-1 levels in rat hepatic stellate cells during experimental liver injury [24]. Thus in many ways, lymphocytes modulate fibrogenesis depending on their activation state, either towards anti- or pro-fibrosis. Clinically, it is hypothesized that immune suppression accelerates the development and progression of fibrosis, underlining the importance of immune regulation for stellate cell function [25-27]. Understanding the pathobiological mechanisms underlying activation states of immune cells and PSC, as well as their interactions which are associated with the development of chronic pancreatitis, is therefore a prerequisite for the design of novel therapeutic approaches. Here, we demonstrate the significance of an interaction between PBMC and PSC which influences fibrogenesis and inflammation in chronic pancreatitis. Patients, Materials and Methods Patients and tissue sampling Patients with chronic pancreatitis (n = 13), patients with sepsis (n = 4), and age- and sex-matched healthy adults (n = 10) were eligible for enrollment in the study. Blood samples were collected preoperatively from CP patients or on the intensive care unit from septic shock patients. Tissue samples were collected during pancreatic resections for CP. These were either immediately processed for isolation of primary human pancreatic stellate cells or snap frozen at -80C or formalin fixed and paraffin embedded. The use of human material for analysis was approved by the local ethics committee (University of Heidelberg, Germany), and written informed consent was obtained prior to the operation and the blood sampling. Immunohistochemistry Immunohistochemistry was performed as described previously [28,29]. Briefly, consecutive sections from chronic pancreati (...truncated)