Surgical Oncology xxx (2015) 1e7

Contents lists available at ScienceDirect

Surgical Oncology journal homepage: www.elsevier.com/locate/suronc

Review

The role of pancreatic stellate cells in pancreatic cancer John A.G. Moir a, b, *, Jelena Mann b, Steve A. White a a b

Freeman Hospital, Department of HPB and Transplant Surgery, Newcastle upon Tyne, United Kingdom Institute of Cellular Medicine, Fibrosis Lab, Newcastle upon Tyne, United Kingdom

a r t i c l e i n f o

a b s t r a c t

Article history: Accepted 11 May 2015

Background: The prognosis of pancreatic cancer remains desperately poor, with little progress made over the past 30 years despite the development of new combination chemotherapy regimens. Stromal activity is especially prominent in the tissue surrounding pancreatic tumours, and has a profound influence in dictating tumour development and dissemination. Pancreatic stellate cells (PaSCs) have a key role in this tumour microenvironment, and have been the subject of much research in the past decade. This review examines the relationship between PaSCs and cancer cells. Methods: A comprehensive literature search was performed of multiple databases up to March 2014, including Medline, Pubmed and Google Scholar. Results: A complex bidirectional interplay exists between PaSCs and cancer cells, resulting in a perpetuating loop of increased activity and an overriding pro-tumorigenic effect. This involves a number of signalling pathways that also impacts on other stromal components and vasculature, contributing to chemoresistance. The Reverse Warburg Effect is also introduced as a novel concept in tumour stroma. Conclusion: This review highlights the pancreatic tumour microenvironment, and in particular PaSCs, as an ideal target for therapeutics. There are a number of cellular processes involving PaSCs which could hold the key to more effectively treating pancreatic cancer. The feasibility of targeting these pathways warrant further in depth investigation, with the aim of reducing the aggressiveness of pancreatic cancer and improving chemodelivery. © 2015 Elsevier Ltd. All rights reserved.

Keywords: Pancreatic cancer Pancreatic stellate cells Tumour stroma

Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Pancreatic cancer e targeting the stroma . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The stromal microenvironment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Composition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The role of the stroma in fibrosis and cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Pancreatic stellate cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . PaSC interactions contributing to PDAC progression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Genome instability and mutation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Sustained proliferative signalling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Invasion and metastasis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tumour promoting inflammation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Angiogenesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Deregulated metabolism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

00 00 00 00 00 00 00 00 00 00 00 00 00

* Corresponding author. Freeman Hospital, Department of HPB and Transplant Surgery, Newcastle upon Tyne, NE7 7XL, United Kingdom. Tel.: þ44 07792405284. E-mail address: [email protected] (J.A.G. Moir). http://dx.doi.org/10.1016/j.suronc.2015.05.002 0960-7404/© 2015 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Moir JAG, et al., The role of pancreatic stellate cells in pancreatic cancer, Surgical Oncology (2015), http:// dx.doi.org/10.1016/j.suronc.2015.05.002

2

J.A.G. Moir et al. / Surgical Oncology xxx (2015) 1e7

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Authorship statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conflict of interest statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Introduction Pancreatic cancer e targeting the stroma Pancreatic cancer continues to be one of the most aggressive and devastating diseases with very few long-term survivors. Factors contributing to this poor prognosis include late clinical presentation, the increased propensity to metastasise, and the high level of resistance tumours exhibit towards chemotherapy and radiotherapy. The most common malignant pancreatic tumour is a pancreatic ductal adenocarcinoma (PDAC), which is of exocrine origin and accounts for over 95% [1]; therefore discussion within this review is specifically centred upon PDAC. It is essential that an improved understanding is acquired regarding the underlying cellular biology and genetic interactions that contribute to the aggressiveness of the disease. The aim is to discover new molecular pathways that may be targeted with therapeutics, thereby not only improving survival outcomes in the palliative setting, but also downsizing tumours with neoadjuvant treatment that may potentially facilitate curative resection. Tumour stroma is defined as the interstitial tissue surrounding a malignant lesion, consisting of a wide variety of inflammatory, vascular and neural components which interact with tumours resulting in a desmoplastic reaction (DR) [2] which encourages tumour invasion, metastasis, and chemoresistance [3e5]. Stromal activity is most linked with epithelial tumours (breast, prostate, ovarian, colorectal), and pancreatic cancer perhaps demonstrates the most prominent ‘stromal’ reaction [5]. Furthermore it has been observed that stromal reactions can precede cancer [6], confirming the importance of premalignant lesions such as pancreatic intraepithelial neoplastic lesions (PanINs). Pancreatic stellate cells (PaSCs) are major players in the stromal ‘desmoplastic’ reaction (DR) associated with PDAC. Furthermore, of all the components involved in this process, PaSCs seem to have the most significant role in generating a feedback loop between the stroma and cancer, and therefore have an essential role in tumour development. This review will examine current evidence of the cellular biology surrounding the stromal microenvironment, and the key role of PaSCs in the DR. The stromal microenvironment Composition The stromal components surround the central core of a malignant pancreatic ductal adenocarcinoma. The stroma is made up of a variety of distinctly different cellular substrates resulting in a dense desmoplastic reaction (DR) which accounts for up to 90% of the total tumour volume [7]. This stroma can typically be seen surrounding tumours on radiological imaging (computed tomography) in the form of ill-defined inflammatory tissue, and this suggests an interesting concept in tumour staging as large sections of “activated” stroma may be left behind following resection, which may encourage proliferation of extremely small numbers of tumour cells that may have been left in situ (R1 margin), as well as acting as a

00 00 00 00

barrier to chemotherapy, hence reducing the likelihood of eliminating this residual disease with adjuvant treatment. It was previously thought that this DR had a protective effect [8]. However it has now been shown that once the stromal elements are activated they take on promalignant attributes, and possibly even initiate carcinogenesis [3]. The DR is composed of a variety of cellular and non-cellular components, as described in Table 1. Essentially the cellular components interact with each other and the adjacent cancer cells, resulting in the deposition of an abundant extracellular matrix (ECM). The role of the stroma in fibrosis and cancer In the context of a normal pancreas, the stroma can be deemed a protective environment, providing crucial signalling which maintains tissue architecture and suppresses the malignant phenotype. In normal epithelial tissue, non-activated fibroblasts are the main players in the secretion and organisation of fibrous proteins, in particular type 1 collagen which constitutes 90% of protein content. When an injury is sustained to the ECM, the classic wound healing process is activated resulting in vascular damage and the formation of a fibrin clot. Activated monocytes then differentiate into macrophages, and subsequently release various growth factors, cytokines and proteases (e.g. metalloproteinases - MMPs). This results in both angiogenesis, and most importantly fibroblast proliferation, which leads to increased deposition of ECM proteins and resultant fibrosis. Normal tissue has a feedback mechanism to maintain homeostasis and regression in fibrosis. However, as supported by clinical observation such as in cases of chronic pancreatitis and liver cirrhosis, repeated injury results in continuous ECM synthesis, and ultimately irreversible damage. This continual damage to the ECM can be viewed as an aging process, whereby basement membranes become thin, and suboptimal cross-linking of collagen occurs. This results in a weak and fragile ECM with disorganisation promoting age-related diseases such as cancer [13]. The various ECM components as highlighted in Table 1 also have a pro-tumorigenic role in their own right. Fibrous proteins such as fibronectin and laminin stimulate reactive oxygen species (ROS) production through NAPDH oxidase, resulting in increased pancreatic cancer cell survival [14]. Matricellular proteins secreted from both stromal and cancer cells have also been shown to have a key role in facilitating tumour progression through downstream signalling that results in increased proliferation and metastasis of cancer cells. They have also been shown to directly impact on aberrant ECM remodelling, further promoting the pro-tumorigenic environment in the stroma [15]. Pancreatic stellate cells Stellate cells have been found in a variety of organ locations, including liver, kidney, intestine and spleen [16]. Pancreatic stellate cells were first isolated in 1998 by Apte et al. [17] and since then they have been shown to have a key role in health by maintaining tissue architecture through the regulation of ECM protein synthesis and degradation. In normal tissue PaSCs exist in a quiescent state, comprising 4e7% of pancreatic parenchyma [18]. These cells

Please cite this article in press as: Moir JAG, et al., The role of pancreatic stellate cells in pancreatic cancer, Surgical Oncology (2015), http:// dx.doi.org/10.1016/j.suronc.2015.05.002

J.A.G. Moir et al. / Surgical Oncology xxx (2015) 1e7

3

Table 1 Cellular and non-cellular components of the desmoplastic reaction within the stromal microenvironment. Cellular components

Non-cellular components

Fibroblasts and PaSCs

Most prominent role in the desmoplastic reaction

Fibrous proteins

Endothelial cells

Hypoxia, secondary to fibrosis and anti-angiogenic components, results in increased tumour growth and metastasis [9] Role in PDAC in keeping with the fact chronic pancreatitis is a strong risk factor Tumour associated macrophages, neutrophils and T cells localise to stroma, secreting growth factors and cytokines which stimulate the DR [10] Presence of leucocytes and pro-inflammatory markers (IL-6,8,10) associated with worse survival [11,12] Abundant nerve supply in the pancreas, making pain associated with disease a major challenge

Proteoglycans

Inflammatory cells

Nerve cells

contain an abundance of vitamin A lipid droplets within their cytoplasm and are therefore considered an adipogenic phenotype. In the normal pancreas PaSCs become activated in response to any form of insult, triggering fibrosis through various stromal interactions and thus allowing the wound healing process to occur. On resolution the PaSCs return to their quiescent state until they are called upon again in their damage limitation role. PaSCs have various other roles in normal pancreas functioning. Quiescent PaSCs have a positive effect on epithelial integrity through the integrin b1-dependent maintenance of the basement membrane, thus demonstrating a role in acinar functionality [19]. An additional role in exocrine function has been demonstrated, whereby the gastrointestinal hormone cholecystokinin (CCK) induces acetylcholine secretion by PaSCs, which in turn stimulates amylase secretion by acinar cells [20]. Furthermore CCK has been shown to have direct activating effect on PaSCs, and induces collagen synthesis [21]. The exact source of activated PaSCs (myofibroblast phenotype) is still a matter of much debate; however a consensus exists that in the context of benign inflammation or malignancy, the surrounding cancer, immune or endothelial cells release various growth factors and inflammatory cytokines that in turn activate these PaSCs through paracrine signalling networks [7]. When studying PaSCs it is important to realise the inherent differences between and activated and quiescent phenotype, particularly for the purposes of cell isolation and culture, and these are listed in Table 2. A further consideration to take into account when culturing PaSCs is the fact plastic itself stimulates activation and thus does not replicate the normal stromal microenvironment. A study comparing culture on plastic, matrigel (a basement membrane-like compound that mimics ECM in health) and collagen I (which mimics the ECM in diseased conditions) revealed significant differences in PaSC gene expression patterns [22].

Table 2 Table demonstrating differing characteristics of activated and quiescent PaSCs. Characterisation feature [22]

Activated PaSC

Quiescent PaSC

Appearance

Star shaped

aSMA expression Desmin expression Vimentin expression GFAP expression Adherence in tissue culture ECM protein secretion Receptor expression Density

Positive (in >90%) Positive in 20e40% Positive Positive þþþ þþþ þþþ þþþ

Spindle shaped and smaller Negative Negative Positive Negative þ þ þ þ

Including collagen, elastin, fibronectin, desmin and laminin Hydrating and buffering in interstitial space

The true hallmark of an activated PaSC is aSMA expression, as demonstrated in Table 1. As such a prognostic marker has been developed termed the Activated Stroma Index (ASI) which is the ratio of aSMA to collagen [23]. If the ASI is high, this therefore indicates increased PaSC activity and thus increased stromal interactions. The ASI has been identified as an independent prognostic marker in multivariate survival analysis when compared with the nodal status of the pancreatic cancer; however no correlation was found with tumour size or grade.

PaSC interactions contributing to PDAC progression Fig. 1 demonstrates potential pathways that exist between PaScs and cancer cells, emphasising the complex bidirectional relationship. However the question remains as to which of the various pathways exert the most influence on disease progression. Is it the direct interplay between cancer cell and PaSC, and if so which pathway exerts the greatest effect? Is it the effect on the ECM, with irreversible and disorganised deposition promoting carcinogenesis? Or is tumour aggressiveness more related to the disruption in vasculature that ensues, contributing to the hypoxic environment that not only drives PDAC but also increases chemoresistance. The likelihood is these concepts combine together to promote cancer progression. An exciting prospect exists whereby if the PaSCs can be targeted, one can exert an influential effect on the key and complex interplay between the cancer cells, ECM and tumour vasculature. The pro-tumorigenic effects of PaSCs will now be discussed with respect to the relevant “Hallmarks of Cancer” [24].

Genome instability and mutation PDAC is an extremely heterogeneous cancer, and a number of mutations drive its aggressive nature, most commonly the KRAS oncogene (in approximately 95% of cases), and the tumour suppressor genes such as INK4A/CDKN2A (85%), TP53 (80%) and SMAD4 (50%) [25].These mutations underpin cancer initiation, development and invasion, however a supportive microenvironment is required allow sustained proliferation. This can involve genetic alterations in various pathways which are inextricably associated with the PaSC-cancer cell relationship. For example TGFb has a complex role and genetic analysis has revealed a mutation in at least one of the TGFb family members in 100% of pancreatic cancers [26]. Furthermore additional mutations may be contributing to an increase in tumour invasiveness; in a cancer cellfibroblast co-culture it was found that the up-regulated genes COX-

Please cite this article in press as: Moir JAG, et al., The role of pancreatic stellate cells in pancreatic cancer, Surgical Oncology (2015), http:// dx.doi.org/10.1016/j.suronc.2015.05.002

4

J.A.G. Moir et al. / Surgical Oncology xxx (2015) 1e7

Fig. 1. Interactions of PaSCs in the stromal microenvironment. EGF e epidermal growth factor, FGF e fibroblast growth factor, HGF e hepatocyte growth factor, MMP e metalloproteinases, PDGF e platelet derived growth factor, TGF b1 e transforming growth factor b1, TNF e tumour necrosis factor a, VEGF e vascular endothelial growth factor.

2, HAS2 and MMP-1 were all associated with invasive potential [27]. Sustained proliferative signalling Cancer cells possess an innate ability to undergo sustained and belligerent proliferation, overriding the body's defence mechanisms through the deregulation of various signalling pathways (Fig. 1). Firstly cancer cells secrete growth factors which stimulate PaSC activation. These include TGF-b, HGF, FGF and EGF [18]. TGF-b has a particularly strong influence; both in vitro and in vivo coculture studies have demonstrated that expression induces proliferation of fibroblasts, and results in a more significant DR with excessive ECM protein deposition [28,29]. Sonic hedgehog (SHH) signalling has also been found to have a key role in both inflammation and malignancy through its effect on the DR. SHH has recently been found in precursor and primary pancreatic lesions, whilst being absent in the normal pancreas, thus implicating this pathway in the initiation and progression of pancreatic cancer [30]. It has been shown that SHH stimulates PaSCs by a paracrine mechanism, and furthermore promotes EMT, proliferation and invasive capacity [31]. However attempts to utilise SHH inhibition as a treatment approach have yielded less favourable results, potentially through PaSC depletion, emphasising the degree of caution required when targeting this pathway. Invasion and metastasis The process of invasion and metastasis has been sequentially broken down into the cellular processes that occur as cancer cells move into the local blood vessels and lymphatics, then extravasate into distant tissues and colonise, with progression of micrometastases into macroscopic tumours. Studies have demonstrated that when cancer cells and PaSCs combine, there is increased activity; a co-culture study in an in vivo subcutaneous mouse model showed the combination of cancer cells and PaSCs resulted in increased tumour size compared to cancer cells alone [32]. With respect to dissemination, epithelial mesenchymal transition (EMT) has been importantly linked to this process. EMT allows cells to

dissociate from their origin, losing their epithelial features, and is thereafter heavily linked with tumour proliferation and metastases [33]; PaSCs have been shown to have a role in this key developmental regulatory programme. In vivo orthotopic models have shown when pancreatic cancer cell lines are combined with PaSCs there is a more pronounced stromal reaction, increased invasion, and increased metastasis [34]. This is supported by the fact PaSCs have been shown to induce EMT in cancer cells themselves, resulting in dissociation and increased migration, further suggesting a role in metastasis [35]. The presence of aSMA positive cells (a hallmark of the activated phenotype) and even PaSCs themselves in metastases also suggests a role in facilitating seeding and proliferation of cancer cells at distant sites, and this concept certainly necessitates more investigation [34,36]. PDGF, secreted from cancer cells, has a potential role in this migratory process as it has been shown to have a chemotactic effect on PaSCs [37]. A further effect of PaScs is their ability to suppress hydrogen peroxide induced apoptosis, and therefore prolong pancreatic cancer cell survival [34]. Tumour promoting inflammation It has become accepted that infiltrating immune cells are present in all neoplastic lesions, in both a tumour preventative and promoting capacity. In the setting of a chronic non-healing wound, inflammatory cells release a host of growth factors, chemokines and cytokines with an overriding pro-tumorigenic effect. In PDAC it has been shown that leucocyte infiltration, even at the earliest stages, orchestrates an immune reaction that is immunosuppressive [11].This topic could be a review in itself, and is of particular interest with respect to immunotherapy, such as CD40 agonists which can induce PDAC tumour regression through T cell mediated pathways [38]. With respect to immune cells, as part of the inflammatory response PaSCs become activated and in turn stimulate the recruitment and localisation of macrophages, neutrophils, T cells and mast cells to the cancer stroma. In turn tumour associated macrophages secrete growth factors (PDGF, TGFB) and cytokines (IL6) which result in an increase in the DR through PaSC activation

Please cite this article in press as: Moir JAG, et al., The role of pancreatic stellate cells in pancreatic cancer, Surgical Oncology (2015), http:// dx.doi.org/10.1016/j.suronc.2015.05.002

J.A.G. Moir et al. / Surgical Oncology xxx (2015) 1e7

and effects on other stromal consituents [39]. All inflammatory cells secrete a plethora of cytokines, in particular IL-6 which activates STAT3 (signal transducer and activator of transcription 3) to promote tumorigenesis in tissue with Kras mutations [35]. As well as recruiting copious numbers of inflammatory cells which further enhance the pool of growth factors and cytokines which promote the DR, PaSCs also secrete galectin-1 which causes apoptosis of T cells and thus inhibits anti-tumour immunity [40]. As mentioned the ECM provides a supportive environment for carcinogenesis, chiefly through the effect of collagen, resulting in reduced physical compliance through an induction of an inflammatory environment, and thereby promoting carcinogenic initiation. In the context of malignancy, it is observed that the stromal matrix, and not the tumour itself, produces the majority of collagen (in particular types 1 and 3), with the principle source being PaSCs which transcribe and secrete the majority of the fibrous proteins constituting the ECM [41]. PaSCs therefore potentially hold the key to developing targeted anti-stromal therapy for the treatment of both benign and malignant conditions of the pancreas. Collagen type 1 has been shown to increase the proliferation, migration and resistance to apoptosis of PDAC cancer cells, mediated via integrin receptor signalling [42]. Various matricellular proteins secreted by PaSCs have also been implicated in increasing cancer cell invasiveness and migration, including SPARC [36] and periostin [43]. Overall the excessive ECM also creates a dense fibrotic paratumoral environment which is hypovascular and hypoxic, meaning there is reduced tumour perfusion of therapeutics, such as gemcitabine [44]. This is due to anomalous collagen formation and associated proteoglycan interactions in the ECM causing a physiological barrier to drugs [45]. It has even been shown that ECM proteins confer resistance in their own right [46]. As well as excessive ECM deposition resulting in protumorigenic effects, ECM turnover and associated degradation also has a key role in PaSC and cancer cell activity, with MMPs and proteases being key players in this process. Co-culture experiments have shown cancer cells exert a paracrine effect on adjacent PaSCs resulting in MMP secretion [47]. MMPs stimulate the degradation of collagens type I, III and IV [48]; however the effect is most marked on type IV, resulting in basement membrane disruption and thereby promoting tumour progression [49]. As a result of the ECM degradation a host of growth factors are released, including TGFB, PDGF and FGF [18]. Once again this represents a selfsustaining process whereby these growth factors affect cancer cell activity, as well as stimulating PaSC activation, further contributing to an over-active and protumorigenic ECM. Angiogenesis The main vascular cells in the stromal microenvironment are fully differentiated endothelial cells which are stimulated by VEGF. In order for tumours to grow they require new blood vessels and VEGF, along with other growth factors such as PDGF and FGF, which stimulate angiogenesis allowing tumour progression [50]. These new blood vessels tend to be disorganised and haemorrhagic, and in combination with the associated fibrotic stroma this results in a suboptimal vascular supply and overriding hypoxic conditions, which in turn stimulates further VEGF secretion [50]. Therefore VEGF over-expression is linked with both poor prognosis and increased metastatic potential [2]. VEGF is secreted by PaSCs, as well as fibroblasts, the ECM and cancer cells [51]; therefore antiPaSC therapies designed to cut off production at the source, rather than the current tactic of targeting VEGF receptors, represents a novel approach. The desmoplastic reaction itself also has a key role in supporting tumour growth through distorted blood supply and an overriding

5

hypoxic environment. A combination of reduced blood flow and increased metabolic activity in the surrounding stroma results in hypoxia which PDAC thrives in, contributing to the aggressiveness of disease [52]. It is already well known that hypoxia is strongly linked with both increased tumour growth and metastasis in PDAC [53]. The DR, which can now clearly be seen to be chiefly regulated and stimulated by PaSCs, leads to an increase in ECM proteins (particularly collagen) and an increase in interstitial fluid pressure (IFP), resulting in this disorganised vasculature which supports tumorigenesis [5]. Furthermore genetic factors are implicated, as evidence suggests hypoxia stabilises transcription factor HIF-1a and therefore up-regulates genes including glucose transporters (Glut1), glycolytic enzymes (Aldolase A) and angiogenic factors; this in turn fuels the interplay between cancer cells and their associated fibroblasts [54]. This is supported by studies showing HIF1a expression is associated with chemoresistance and increased tumour invasion [53]. A recent study has demonstrated that a further selfperpetuating loop exists, whereby PaSCs promote hypoxia, and in turn this hypoxic environment further stimulates PaSC activation. It has been shown that cancer-associated PaSCs in hypoxia enhance the invasion of cancer cells more strongly than those in normoxia, with carbonic anhydrase 9 (CA9) being a useful hypoxic-regulated marker (as well as HIF-1 a) [55]. As expected VEGF expression was increased (4.5 fold) in hypoxia as compared to normoxia, however the role of connective tissue growth factor (CTGF) was elucidated. CTGF, which is secreted by PaSCs, is already known to be expressed under hypoxia, and has been shown to protect cancer cells from apoptosis in an autocrine manner [56]. It has now been demonstrated that CTGF has a key role in the hypoxic environment, with knockdown experiments revealing CTGF secretion was increased in hypoxia, resulting in an increase in PaSC-induced cancer cell invasion in hypoxia as compared to normoxia; these findings were confirmed in both in vivo and in vitro experiments. Deregulated metabolism A topic that has received less attention with respect to the tumour microenvironment and fibroblasts in particular is the adjustment in energy metabolism that occurs. This may be because traditionally the Warburg Effect has been widely accepted as the mechanism that drives cancer cell metabolism and subsequent anabolic growth. The Warburg effect was a term first coined in the 1920s and states that cancer cells undergo a metabolic shift whereby they start to produce energy via aerobic glycolysis, which involves the conversion of glucose into lactate. It is based on the principle that “the prime cause of cancer is the replacement of respiration of oxygen in the normal body cells by fermentation of sugar”, thereby generating energy for cancer cell proliferation [57]. The Warburg effect has always been somewhat of a paradox, as mitochondrial oxidative phosphorylation produces 34 more ATPs per molecule of glucose than aerobic glycolysis. However a recent concept, termed the “reverse Warburg effect” (RWE), potentially revolutionises the way we look at cancer cell behaviour and the key role the stroma plays in its development. The RWE, which has been studied in breast cancer (however never directly in pancreatic cancer), demonstrates that stromal cancer associated fibroblasts (CAFs), which resemble the activated PaSC form, are the main site of aerobic glycolysis. Hydrogen peroxide is secreted by cancer cells and this triggers oxidative stress in neighbouring fibroblasts, resulting in activation of CAFs, glycolysis, autophagy and overall catabolism [58]. As a result the CAFs convert glucose to L-lactate, and in turn oxidative epithelial cancer cells act as parasites by hijacking this lactate and using it to generate larger amounts of ATP via oxidative phosphorylation in mitochondria, and

Please cite this article in press as: Moir JAG, et al., The role of pancreatic stellate cells in pancreatic cancer, Surgical Oncology (2015), http:// dx.doi.org/10.1016/j.suronc.2015.05.002

6

J.A.G. Moir et al. / Surgical Oncology xxx (2015) 1e7

thus promoting anabolic growth [59]. Essentially CAFs are acting as an energy substrate factory and in turn feeding cancer cells, thus causing a perpetuating cycle of increased activity and proliferation (Fig. 2). It can therefore be observed that as long as cancer cells produce hydrogen peroxide, they can commandeer neighbouring fibroblasts and turn them into an energy substrate factory to feed their growth. This may explain why cancer cells so readily metastasise to the liver, where there is an abundance of fibroblasts to take advantage of, and high levels of ketones. This increased invasive potential is supported by an in vivo study demonstrating that lactate itself promotes tumour growth and metastasis, independent of angiogenesis [60]. It has even been suggested that surgeons and oncologists should reconsider the use of Hartmann's solution in the fluid therapy of cancer patients due to the potentially detrimental effects of the lactate content [60]. It has been shown that breast CAFs express MCT4, a monocarboxylate transporter involved in lactate efflux with cancer cells further inducing its expression [61]. This concept supports the theory of the RWE, as rather than being glycolytic, cancer cells instead import lactate from surrounding CAFs using these up-regulated transporters. We can therefore hypothesise that PaSCs may also be implicated in the movement of lactate between the stroma and epithelial tumour, and interrupting this food chain may induce a tumour suppressive effect.

one must be wary when determining which therapeutic approach is most appropriate. Detailing treatment strategies would be a further review in itself, however we have demonstrated the multitude of avenues one may take in attempting to target the bidirectional relationship between PaSCs and cancer cells, whether that be directly targeting PaSCs, interrupting associated signalling pathways with both cancer cells and stromal components, or influencing the metabolic symbiosis. Further studies will help delineate the most influential pathways and thus most appropriate and effective therapeutic approach in the hope of improving outcomes in this devastating disease.

Conclusions

[1] Hidalgo M. Pancreatic cancer. N Engl J Med 2010 Apr 29;362(17):1605e17. [2] Maitra A, Hruban RH. Pancreatic cancer. Annu Rev Pathol 2008;3:157e88. [3] Bissell MJ, Radisky D. Putting tumours in context. Nat Rev Cancer 2001;1: 46e54. [4] Heldin CH, Rubin K, Pietras K, Ostman A. High interstitial fluid pressure e an obstacle in cancer therapy. Nat Rev Cancer 2004 Oct;4(10):806e13. [5] Vonlaufen A, Phillips PA, Xu Z, Goldstein D, Pirola RC, Wilson JS, et al. Pancreatic stellate cells and pancreatic cancer cells: an unholy alliance. Cancer Res 2008;68(19):7707e10. [6] Moinfar F, Man YG, Arnould L, Bratthauer GL, Ratschek M, Tavassoli FA. Concurrent and independent genetic alterations in the stromal and epithelial cells of mammary carcinoma: implications for tumorigenesis. Cancer Res 2000;60:2562e6. [7] Neesse A, Michl P, Frese KK, Feig C, Cook N, Jacobetz MA, et al. Stromal biology and therapy in pancreatic cancer. Gut 2011 Jun;60(6):861e8. [8] Lunevicius R, Nakanishi H, Ito S, Kozaki K, Kato T, Tatematsu M, et al. Clinicopathological significance of fibrotic capsule formation around liver metastasis from colorectal cancer. J Cancer Res Clin Oncol 2001;127:193e9. [9] Chang Q, Jurisica I, Do T, Hedley DW. Hypoxia predicts aggressive growth and spontaneous metastasis formation from orthotopically grown primary xenografts of human pancreatic cancer. Cancer Res 2011;71(8):3110e20. [10] Negus RP, Stamp GW, Hadley J, Balkwill FR. Quantitative assessment of the leukocyte infiltrate in ovarian cancer and its relationship to the expression of C-C chemokines. Am J Pathol 1997 May;150(5):1723e34. [11] Clark CE, Hingorani SR, Mick R, Combs C, Tuveson DA, Vonderheide RH. Dynamics of the immune reaction to pancreatic cancer from inception to invasion. Cancer Res 2007 Oct 1;67(19):9518e27. [12] Ebrahimi B, Tucker SL, Li D, Abbruzzese JL, Kurzrock R. Cytokines in pancreatic carcinoma: correlation with phenotypic characteristics and prognosis. Cancer 2004;101(12):2727e36. [13] Frantz C, Stewart KM, Weaver VM. The extracellular matrix at a glance. J Cell Sci 2010;123(24):4195e200. [14] Edderkaoui M, Hong P, Vaquero EC, Lee JK, Fischer L, Friess H, et al. Extracellular matrix stimulates reactive oxygen species production and increases pancreatic cancer cell survival through 5-lipoxygenase and NADPH oxidase. Am J Physiol Gastrointest Liver Physiol 2005 Dec;289(6):G1137e47. [15] Wong GS, Rustgi AK. Matricellular proteins: priming the tumour microenvironment for cancer development and metastasis. Br J Cancer 2013 Mar 5;108(4):755e61. [16] Geerts A. History, heterogeneity, developmental biology, and functions of quiescent hepatic stellate cells. Semin Liver Dis 2001;21:311e35. [17] Bachem MG, Schneider E, Gross H, Weidenbach H, Schmid RM, Menke A, et al. Identification, culture, and characterization of pancreatic stellate cells in rats and humans. Gastroenterology 1998;115:421e32. [18] Erkan M, Adler G, Apte MV, Bachem MG, Buchholz M, Detlefsen S, et al. StellaTUM: current consensus and discussion on pancreatic stellate cell research. Gut 2012 Feb;61(2):172e8. [19] Riopel MM, Li J, Liu S, Leask A, Wang R. b1 integrin-extracellular matrix interactions are essential for maintaining exocrine pancreas architecture and function. Lab Invest 2013 Jan;93(1):31e40. [20] Phillips PA, Yang L, Shulkes A, Vonlaufen A, Poljak A, Bustamante S, et al. Pancreatic stellate cells produce acetylcholine and may play a role in

It is clear that PaSCs have a profoundly influential role within the tumour microenvironment, exerting effects on both cancer cells and other stromal components. Whilst originally existing to provide a protective mechanism in response to injury, there has been surmounting evidence demonstrating a key role in promoting cancer activity, invasion and dissemination. Nonetheless, PaSCs still seem to retain protective elements, either through themselves or the effect they exert on other stromal components, and therefore

Fig. 2. Reverse Warburg Effect flow chart. Cancer cells stimulate activate the surrounding stroma resulting in aerobic glycolysis in fibroblasts. The resultant nutrients are hijacked by the cancer cells to allow perpetuating anabolic growth. H2O2 e hydrogen peroxide, ROS e reactive oxygen species, aa e amino acids.

Authorship statement Study concepts: John A.G. Moir, Jelena Mann, Steve A. White Study design: John A.G. Moir, Jelena Mann, Steve A. White Definition of intellectual content: John A.G. Moir Manuscript review: John A.G. Moir, Steve A. White Conflict of interest statement None. References

Please cite this article in press as: Moir JAG, et al., The role of pancreatic stellate cells in pancreatic cancer, Surgical Oncology (2015), http:// dx.doi.org/10.1016/j.suronc.2015.05.002

J.A.G. Moir et al. / Surgical Oncology xxx (2015) 1e7

[21]

[22]

[23]

[24] [25] [26]

[27]

[28]

[29]

[30] [31]

[32]

[33]

[34]

[35]

[36]

[37]

[38]

[39]

[40]

[41]

[42]

pancreatic exocrine secretion. Proc Natl Acad Sci U. S. A 2010 Oct 5;107(40): 17397e402. Berna MJ, Seiz O, Nast JF, Benten D, Bl€ aker M, Koch J, et al. CCK1 and CCK2 receptors are expressed on pancreatic stellate cells and induce collagen production. J Biol Chem 2010 Dec 10;285(50):38905e14. Apte MV, Yang L, Phillips PA, Xu Z, Kaplan W, Cowley M, et al. Extracellular matrix composition significantly influences pancreatic stellate cell gene expression pattern: role of transgelin in PSC function. Am J Physiol Gastrointest Liver Physiol 2013 Sep 15;305(6):G408e17. Erkan M, Michalski CW, Rieder S, Reiser-Erkan C, Abiatari I, Kolb A, et al. The activated stroma index is a novel and independent prognostic marker in pancreatic ductal adenocarcinoma. Clin Gastroenterol Hepatol 2008 Oct;6(10):1155e61. Hanahan D, Weinberg RA. “Hallmarks of Cancer: the next generation”. Cell 2011;144(5):646e74. Hansel DE, Kern SE, Hruban RH. Molecular pathogenesis of pancreatic cancer. Annu Rev Genomics Hum Genet 2003;4:237e56. Jones S, Zhang X, Parsons DW, Lin JC, Leary RJ, Angenendt P, et al. Core signaling pathways in human pancreatic cancers revealed by global genomic analyses. Science 2008;321:1801e6. Sato N, Maehara N, Goggins M. Gene expression profiling of tumor-stromal interactions between pancreatic cancer cells and stromal fibroblasts. Cancer Res 2004;64(19):6950e6. Lohr M, Schmidt C, Ringel J, Kluth M, Müller P, Nizze H, et al. Transforming growth factor-beta1 induces desmoplasia in an experimental model of human pancreatic carcinoma. Cancer Res 2001;61(2):550e5. Armstrong T, Packham G, Murphy LB, Bateman AC, Conti JA, Fine DR, et al. Type I collagen promotes the malignant phenotype of pancreatic ductal adenocarcinoma. Clin Cancer Res 2004;10(21):7427e37. Kayed H, Kleeff J, Osman T, Keleg S, Buchler MW, Friess H. Hedgehog signaling in the normal and diseased pancreas. Pancreas 2006;32:119e29. Bailey JM, Swanson BJ, Hamada T, Eggers JP, Singh PK, Caffery T, et al. Sonic hedgehog promotes desmoplasia in pancreatic Cancer. Clin Cancer Res October 1, 2008;14:5995. Bachem MG, Schünemann M, Ramadani M, Siech M, Beger H, Buck A, et al. Pancreatic carcinoma cells induce fibrosis by stimulating proliferation and matrix synthesis of stellate cells. Gastroenterology 2005;128:907e21. Lee JM, Dedhar S, Kalluri R, Thompson EW. The epithelial- mesenchymal transition: new insights in signaling, development, and disease. J Cell Biol 2006;172:973e81. Vonlaufen A, Joshi S, Qu C, Phillips PA, Xu Z, Parker NR, et al. Pancreatic stellate cells: partners in crime with pancreatic cancer cells. Cancer Res 2008;68:2085e93. Apte MV, Wilson JS, Lugea A, Pandol SJ. A starring role for stellate cells in the pancreatic cancer microenvironment. Gastroenterology 2013 Jun;144(6): 1210e9. Xu Z, Vonlaufen A, Phillips PA, Fiala-Beer E, Zhang X, Yang L, et al. Role of pancreatic stellate cells in pancreatic cancer metastasis. Am J Pathol 2010;177:2585e96. Phillips PA, Wu MJ, Kumar RK, Doherty E, McCarroll JA, Park S, et al. Cell migration: a novel aspect of pancreatic stellate cell biology. Gut 2003;52: 677e82. Vonderheide RH, Bajor DL, Winograd R, Evans RA, Bayne LJ, Beatty GL. CD40 immunotherapy for pancreatic cancer. Cancer Immunol Immunother 2013 May;62(5):949e54. Apte MV, Haber PS, Darby SJ, Rodgers SC, McCaughan GW, Korsten MA, et al. Pancreatic stellate cells are activated by proinflammatory cytokines: implications for pancreatic fibrogenesis. Gut 1999 Apr;44(4):534e41. Tang D, Yuan Z, Xue X, Lu Z, Zhang Y, Wang H, et al. High expression of Galectin-1 inpancreatic stellate cells plays a role in the development and maintenance of an immunosuppressive microenvironment in pancreatic cancer. Int J Cancer 2012;130:2337e48. Apte MV, Park S, Phillips PA, Santucci N, Goldstein D, Kumar RK, et al. Desmoplastic reaction in pancreatic cancer: role of pancreatic stellate cells. Pancreas 2004;29:179e87. Grzesiak JJ, Bouvet M. The alpha2beta1 integrin mediates the malignant phenotype on type I collagen in pancreatic cancer cell lines. Br J Cancer 2006;94(9):1311e9.

7

[43] Kanno A, Satoh K, Masamune A, Hirota M, Kimura K, Umino J, et al. Periostin, secreted from stromal cells, has biphasic effect on cell migration and correlates with the epithelial to mesenchymal transition of human pancreatic cancer cells. Int J Cancer 2008;122(12):2707e18. [44] Olive KP, Jacobetz MA, Davidson CJ, Gopinathan A, McIntyre D, Honess D, et al. Inhibition of Hedgehog signalling enhances delivery of chemotherapy in a mouse model of pancreatic cancer. Science 2009 Jun 12;324(5933):1457e61. [45] Netti PA, Berk DA, Swartz MA, Grodzinsky AJ, Jain RK. Role of extracellular matrix assembly ininterstitial transport in solid tumors. Cancer Res 2000;60: 2497e503. [46] Miyamoto H, Murakami T, Tsuchida K, Sugino H, Miyake H, Tashiro S. Tumorstroma interaction of human pancreatic cancer: acquired resistance to anticancer drugs and proliferation regulation is dependent on extracellular matrix proteins. Pancreas 2004;28:38e44. [47] Schneiderhan W, Diaz F, Fundel M, Zhou S, Siech M, Hasel C, et al. Pancreatic stellate cells are an important source of MMP-2 in human pancreatic cancer and accelerate tumor progression in a murine xenograft model and CAM assay. J Cell Sci 2007;120(Pt 3):512e9. [48] Shek FW, Benyon RC, Walker FM, McCrudden PR, Pender SL, Williams EJ, et al. Expression of transforming growth factor-beta 1 by pancreatic stellate cells and its implications for matrix secretion and turnover in chronic pancreatitis. Am J Pathol 2002;160(5):1787e98. [49] Guturu P, Shah V, Urrutia R. Interplay of tumor microenvironment cell types with parenchymal cells in pancreatic cancer development and therapeutic implications. J Gastrointest Cancer 2009;40(1e2):1e9. [50] Carmeliet P. VEGF as a key mediator of angiogenesis in cancer. Oncology 2005;3(69 Suppl):4e10. [51] Masamune A, Kikuta K, Watanabe T, Satoh K, Hirota M, Shimosegawa T. Hypoxia stimulates pancreatic stellate cells to induce fibrosis and angiogenesis in pancreatic cancer. Am J Physiol Gastrointest Liver Physiol 2008 Oct;295(4):G709e17. [52] Komar G, Kauhanen S, Liukko K, Sepp€ anen M, Kajander S, Ovaska J, et al. Decreased blood flow with increased metabolic activity: a novel sign of pancreatic tumor aggressiveness. Clin Cancer Res 2009;15:5511e7. [53] Kindler HL, Niedzwiecki D, Hollis D, Sutherland S, Schrag D, Hurwitz H, et al. Gemcitabine plus bevacizumab compared with gemcitabine plus placebo in patients with advanced pancreatic cancer: phase III trial of the Cancer and Leukemia Group B (CALGB 80303). J Clin Oncol 2010;28(22):3617e22. [54] Akakura N, Kobayashi M, Horiuchi I, Suzuki A, Wang J, Chen J, et al. Constitutive expression of hypoxia-inducible factor-1alpha renders pancreatic cancer cells resistant to apoptosis induced by hypoxia and nutrient deprivation. Cancer Res 2001;61:6548e54. [55] Eguchi D, Ikenaga N, Ohuchida K, Kozono S, Cui L, Fujiwara K, et al. Hypoxia enhances the interaction between pancreatic stellate cells and cancer cells via increased secretion of connective tissue growth factor. J Surg Res 2013 May;181(2):225e33. [56] Bennewith KL, Huang X, Ham CM, Graves EE, Erler JT, Kambham N, et al. The role of tumor cell-derived connective tissue growth factor (CTGF/CCN2) in pancreatic tumor growth. Cancer Res 2009;69:775. [57] Warburg O. On the origin of cancer cells. Science (New York, NY) 1956;123: 309e14. [58] Lisanti MP, Martinez-Outschoorn UE, Lin Z, Pavlides S, Whitaker-Menezes D, Pestell RG, et al. Hydrogen peroxide fuels aging, inflammation, cancer metabolism and metastasis: the seed and soil also needs “fertilizer.” Cell Cycle 2011;10:2440e9. [59] Pavlides S, Whitaker-Menezes D, Castello-Cros R, Flomenberg N, Witkiewicz AK, Frank PG, et al. The reverse Warburg effect: aerobic glycolysis in cancer associated fibroblasts and the tumor stroma. Cell cycle 2009;8: 3984e4001. [60] Bonuccelli G, Tsirigos A, Whitaker-Menezes D, Pavlides S, Pestell RG, Chiavarina B, et al. Ketones and lactate “fuel” tumor growth and metastasis: evidence that epithelial cancer cells use oxidative mitochondrial metabolism. Cell Cycle 2010;9:3506e14. [61] Whitaker-Menezes D, Martinez-Outschoorn UE, Lin Z, Ertel A, Flomenberg N, Witkiewicz AK, et al. Evidence for a stromal-epithelial “lactate shuttle” in human tumors: MCT4 is a marker of oxidative stress in cancer-associated fibroblasts. Cell Cycle 2011;10:1772e83.

Please cite this article in press as: Moir JAG, et al., The role of pancreatic stellate cells in pancreatic cancer, Surgical Oncology (2015), http:// dx.doi.org/10.1016/j.suronc.2015.05.002