UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)
UvA-DARE (Digital Academic Repository)
Opening new doors: Hedgehog signaling and the pancreatic cancer stroma
Damhofer, H.
Publication date
2015
Document Version
Final published version
Link to publication
Citation for published version (APA):
Damhofer, H. (2015). Opening new doors: Hedgehog signaling and the pancreatic cancer
stroma.
General rights
It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s)
and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open
content license (like Creative Commons).
Disclaimer/Complaints regulations
If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please
let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material
inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter
to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You
will be contacted as soon as possible.
CHApTer
7
General Discussion & Outlooknon-canonical Hh signaling in pdAC
Most of the research in the last decades on the signaling mechanisms and components of the Hh pathway has focused on the transcriptional response of cells to Hh ligands, and how this signal orchestrates organization during development and homeostasis in the adult organism. Only recently did we start to realize that there is something beyond (or more literally ‘upstream’ of) the role of Hh in activating the Gli transcription factors and the expression of target genes. Both receptors Ptch and Smo were found to elicit diverse cellular responses independent of transcription. However, the mechanisms separating canonical/transcriptional and non-canonical/transcription-independent signaling are largely unknown. In chapter 2 of this thesis we shed light on how the subcellular localization of Smo can influence which Hh signaling route cells will use, and how the primary cilium skews this towards the transcriptional response.
This novel concept has several important implications. For instance, it implies that the primary cilium is not necessary for all the functions of Hh ligands and that non-canonical Hh signaling can still occur and play an active role in diseases that were previously thought unable of such signaling due to a defective or absent primary cilium (ciliopathies and malignancies, respectively). In addition, in many different experiments we found that the transcriptional response to Hh was much easier perturbed than the non-canonical one. We interpret this to mean that the non-canonical signaling is more robust and thus more easily retained in cells with (oncogenically) impaired or otherwise altered signaling.
The finding that chemotactic signaling is enhanced by the absence of a primary cilium is also interesting in the context of pancreatic cancer, as these cancer cells are generally thought to not be responsive to the Hh ligands they produce, and signaling is instead strictly paracrine to the stroma. This notion is based on the fact that these cells do not respond well to pathway inhibitors [1], and that they are devoid of a primary cilium and therefore incapable of transducing the signal to the Gli transcription factors [2]. However, all these studies only addressed the potential of transcriptional signaling responses and did not consider non-canonical effects. High local Hh concentration could provide tumor cells with an important survival and cell division cue as it was shown that Ptch can directly induce apoptosis and inhibit proliferation in absence of ligand, independent of Smo function, and these effects are blocked by the presence of ligand [3, 4].
Furthermore, due to the lack of a primary cilium, pancreatic cancer cells could be more prone to show a chemotactic response towards Hh ligand, and we did observe that PDAC cells migrate towards Hh ligand or Smo agonists (not shown in thesis). This could point to a role of the locally produced Hh in preventing tumor cell dissemination. This theory fits with the recent finding in mouse models of pancreatic cancer that show earlier tumor cell dissemination in tumors that lack Shh [5, 6]. The authors interpreted these findings to be a result of accompanied changes in the stroma by lack of canonical Hh signaling and disruption of tumor-restraining functions of the microenvironment, but effects of non-canonical Hh signaling on tumor and stromal cells were not excluded and could contribute to this phenotype. Interestingly, it was recently shown that cholangiocarcinoma cells also lack a primary cilium,
Chapter 7
7
have a chemotactic response towards Smo agonists and pharmacologic inhibition of Smo in
vivo in a syngeneic orthotopic rat model resulted in reduced metastasis formation, which was
speculated to be caused by non-canonical Hh signaling [7]. This indicates that chemotactic Hh signaling could have a broad implication in several malignancies, especially in tumor cells that have deficiencies in generating primary cilia.
Finally, Hh ligand produced by tumor cells could also serve as a chemoattractant for fibroblasts or other stromal cells, by activating their chemotactic Hh response. This could very well contribute to the desmoplastic phenotype of pancreatic tumors [8]. As both canonical as well as non-canonical signaling cascades are likely activated in parallel, it is still rather difficult to dissect the effects of disrupted signal in this complex system with several cell types present engaging in extensive crosstalk. One of the future challenges will be to develop models to reveal the relevance of non-canonical Hh signaling for developmental processes and diseases in vivo without confounding effects from transcriptional signaling. Tools to specifically address this could include blocking of specific non-canonical Hh signaling hubs, like the leukotriene synthesis machinery [9, 10], or the Src family of kinases [11].
Cell-bound Hh: a potent signaling molecule
Hh proteins are rather unusual signaling molecules due to their hydrophobic nature. Several different mechanisms have been proposed for how this lipid-modified morphogen can spread through tissue (see chapter 1), such as formation of multimers, association with lipoproteins and exosomes, release via Dispatched/Scube2, or shedding via ADAM metalloproteases. Most of these mechanisms have been studied in developmental systems or in vitro by over-expression. How these mechanisms relate to Hh produced by pancreatic cancer cells has not been addressed before. In chapter 4 we show that ADAMs are upregulated in pancreatic cancer, and that they can release Hh from tumor cells. Intriguingly, blockage of Hh release led to an unexpected increase in the signaling potency to fibroblasts in direct coculture experiments. In addition, the injection of pancreatic cancer cells with perturbed Hh dispersion (by ADAM12 or Dispatched silencing, unpublished observation) into the pancreata of mice resulted in an increased expression of Hh target genes in the murine tumor stroma. This implies that also in the context of tumor tissue, Hh retention actually leads to enhanced paracrine signaling activity.
This notion is especially interesting in regards to the recently described transport mechanisms of Hh along filopodia-like cell protrusions termed cytonemes [12, 13]. The observation made in our study would fit with a similar transport model in which accumulation of Hh molecules on the surface of producing cells would allow higher concentration of Hh on cell protrusions, and as a consequence enhanced activation of signal-receiving cells. Conversely, release of Hh by ADAMs or Dispatched would result in undirected dispersion of ligand, preventing accumulation of sufficient ligand to reach a threshold necessary to active the signaling pathway in the receiving cells. This is a new way of understanding the exchange of information between cells and it would be interesting to see if these findings extend to other signaling molecules as well. The initial discovery of cytoneme mediated
7
General Discussion & Outlooktransport was made in context of fibroblast growth factor signaling in Drosophila [14] and later extended to decapentaplegic/TGF-
β
signaling in the Drosophila wing disc [15]. This indicates that other signaling molecules beside Hh could also be carried by cell protrusions to their target cells. However, we are only at the beginning of exploring this new mode of signaling molecule dispersion and it could be argued that the impact of these findings, as well as our own, needs further characterization and exploration in models relevant for human disease rather than the development of insects or fishes.Other remaining questions pertain to how the different Hh release mechanisms described in literature relate to each other, and in which context cells use these different routes and what is the role of these different entities of Hh in development and disease. One very specific question that would be interesting but technically challenging to address is the presence and activity of cytonemes in tumors.
Hh signaling in pancreatic cancer stroma
The paracrine nature of (canonical) Hh signaling in pancreatic cancer was established a few years ago [1, 16] and has since then been confirmed in several pre-clinical models [5, 17, 18]. The aim of chapter 5 of this thesis was to discover the paracrine factors that are induced by tumor cell-derived Hh and which could potentially exert tumor-promoting functions on the epithelium. For this, we set up a novel model to study the transcriptional responses in the tumor versus the stromal compartment without the need for physical separation of these cells. We adapted a culturing method normally used in developmental stem cell biology, and forced human tumor cells to aggregate with mouse fibroblasts by culturing on a shaker. By then performing RNA-Seq analysis on these cultures, and stringent alignment of the reads to the human and the mouse genome, we were able to genetically dissect the two compartments. In doing so, we identified several potential candidates that were then stringently selected on their annotation as extracellular factors, their upregulation in cancer tissue, and association with poor clinical outcome in patients. Some of the identified proteins such as Plaur, Cda, Edil3 and Spock1 have been reported to promote different aspect of malignant progression in several cancers and are attractive candidates for further study in context of pancreatic cancer [19-21].
More mechanistic work on some of the stromal factors identified in our screen should help in understanding the complex role of Hh signaling in the pancreatic cancer stroma, especially in light of recent literature suggesting a rather tumor-restrictive function of Hh signaling in pancreatic cancer opposed to the previously suggested tumor-promoting effect, at least in genetically engineered mouse models of this disease [5, 6]. For example, several bone morphogenic proteins (BMPs) were found to be Hh regulated in our screen, although not necessarily highly upregulated in pancreatic cancer tissue. In a bladder cancer model, Bmp4 and Bmp5 were identified to be stromal-derived paracrine differentiation signals that are upregulated by tumor cell-derived Hh ligand and prevent dissemination of epithelial cells [22]. Bmp4 was also found to be regulated by Ihh in the intestinal mucosa and mesenchyme of intestinal adenomas [23, 24] and was able to induce differentiation of colorectal cancer
Chapter 7
7
stem cells, thereby increasing their response to chemotherapy [25]. Detailed analysis using mouse models of pancreatic cancer and patient-derived xenografts (PDXs) will be needed to functionally test the contribution of BMPs and other stromal differentiation cues to tumor initiation and progression. Preliminary analyses of our candidate list did reveal several BMP and Wnt paralogs to be potential stromal targets of Hh.
The newly developed PDXs and primary cell line models described in chapter 3 will be helpful in further investigating this tumor-stroma crosstalk, as well as elucidating the mechanisms and effects of new combinational therapies. Especially, further molecular characterization of PDXs and adaptation to an orthotopic grafting location together with improved imaging of experimental animals should aid in future biological and pharmacological studies.
identification of serum markers for stromal activation
The desmoplastic pancreatic cancer stroma is generally thought to provide a barrier preventing efficient delivery of therapeutics. Several cell types and ECM components have been shown to harbor tumor-promoting functions [26]. As a consequence, a lot of research has focused on developing agents to target this compartment and enhance the efficacy of chemotherapeutics [27]. However, it is still challenging to monitor these new targeting strategies, as no appropriate non-invasive stromal biomarker exists to date. Most of the molecular studies on stromal markers were performed on resected tumor material, which makes it difficult to apply to the majority of patients with locally advanced and metastatic disease, where there is often no access to sufficient tissue material to perform histological analysis.
In chapter 6 we found that ADAM12 is highly expressed in the stroma of pancreatic cancer patients and that it correlates with stromal (activation) markers and content. Also in the serum of pancreatic cancer patients, ADAM12 was found to be elevated and high serum levels correlated with poor clinical outcome. This suggests ADAM12 to be a promising serum-borne biomarker to monitor stromal targeting therapies, illustrated by one patient from our cohort with locally advanced pancreatic cancer who showed a very strong reduction in ADAM12 serum levels after combination therapy with nab-paclitaxel and gemcitabine.
Future work should focus on testing the feasibility of using ADAM12 serum levels to select patients for stromal targeting therapy, how ADAM12 serum levels change during the course of different treatment regimens, and how reductions in serum levels relate to response and patient outcome. We are currently investigating and expanding the opportunities to address these questions. Furthermore, it will be interesting to see if ADAM12 is upregulated in, and could be useful as a biomarker for, other diseases that involve stromal activation or a fibrotic response such as pulmonary fibrosis or prostate cancer.
COnLUding reMArKs
There are still a lot of things we do not understand about Hh signaling in general, and its role in pancreatic cancer specifically. The recently failed clinical trials of Hh inhibitors in pancreatic cancer patients attest to that. For example, we are only at the beginning
7
General Discussion & Outlookof exploring the possibilities and implications of non-canonical signaling routes in development and disease and even the form in which the signaling molecule is produced and distributed is still under debate.
In this thesis we provide data that contribute to our understanding of the fundamental mechanisms underlying this disease, but also provide tentative clinically applicable tools. The identification of the mechanisms governing non-canonical Hh signaling opens up previously unrecognized possibilities for Hh signaling in cells that were thought to be ‘unresponsive’ to Hh ligands. The finding that Hh ligand dispersal from pancreatic cancer cells is detrimental for signaling potency challenges the prevailing thought that Hh molecules need to be released to exert cellular responses. Additionally, new insight into Hh-regulated mediators of the pancreatic tumor stroma was gained by our transcriptional profiling approach, yielding interesting new leads. Especially exploring the new model systems such as PDXs and primary cell cultures described in this thesis provide excellent tools for future mechanistic studies as they can be manipulated ex vivo and reintroduced into animals. This will allow for example to address the role of paracrine signaling molecules in different cellular compartments by making use of the species differences between tumor and stroma cells as well as further study and characterize non-canonical Hh signaling events and filopodia-mediated transport of signaling ligands. Taken together, the new findings provided in this thesis will hopefully provide a basis for future studies to help us unravel some of the complexity of the Hh signaling pathway and pancreatic tumor biology.
reFerenCes
1 Yauch RL, Gould SE, Scales SJ, Tang T, Tian H, Ahn CP, Marshall D, Fu L, Januario T, Kallop D, et al: A paracrine requirement for hedgehog signalling in cancer. Nature 2008, 455:406-410. 2 Seeley ES, Carriere C, Goetze T, Longnecker DS,
Korc M: Pancreatic cancer and precursor pancreatic intraepithelial neoplasia lesions are devoid of primary cilia. Cancer Res 2009, 69:422-430. 3 Barnes EA, Kong M, Ollendorff V, Donoghue DJ:
Patched1 interacts with cyclin B1 to regulate cell cycle progression. EMBO J 2001, 20:2214-2223. 4 Thibert C, Teillet MA, Lapointe F, Mazelin
L, Le Douarin NM, Mehlen P: Inhibition of neuroepithelial patched-induced apoptosis by sonic hedgehog. Science 2003, 301:843-846. 5 Rhim AD, Oberstein PE, Thomas DH, Mirek ET,
Palermo CF, Sastra SA, Dekleva EN, Saunders T, Becerra CP, Tattersall IW, et al: Stromal elements act to restrain, rather than support, pancreatic ductal adenocarcinoma. Cancer Cell 2014, 25:735-747. 6 Lee JJ, Perera RM, Wang H, Wu DC, Liu XS, Han
S, Fitamant J, Jones PD, Ghanta KS, Kawano S, et al: Stromal response to Hedgehog signaling
restrains pancreatic cancer progression. Proc
Natl Acad Sci U S A 2014, 111:E3091-3100.
7 Razumilava N, Gradilone SA, Smoot RL, Mertens JC, Bronk SF, Sirica AE, Gores GJ: Non-canonical Hedgehog signaling contributes to chemotaxis in cholangiocarcinoma. J Hepatol 2014, 60:599-605.
8 Bailey JM, Swanson BJ, Hamada T, Eggers JP, Singh PK, Caffery T, Ouellette MM, Hollingsworth MA: Sonic hedgehog promotes desmoplasia in pancreatic cancer. Clin Cancer
Res 2008, 14:5995-6004.
9 Bijlsma MF, Borensztajn KS, Roelink H, Peppelenbosch MP, Spek CA: Sonic hedgehog induces transcription-independent cytoskeletal rearrangement and migration regulated by arachidonate metabolites. Cell Signal 2007, 19:2596-2604.
10 Bijlsma MF, Peppelenbosch MP, Spek CA, Roelink H: Leukotriene synthesis is required for hedgehog-dependent neurite projection in neuralized embryoid bodies but not for motor neuron differentiation. Stem Cells 2008, 26:1138-1145.
Chapter 7
7
11 Yam PT, Langlois SD, Morin S, Charron F: Sonic hedgehog guides axons through a noncanonical, Src-family-kinase-dependent signaling pathway. Neuron 2009, 62:349-362. 12 Bischoff M, Gradilla AC, Seijo I, Andres G,
Rodriguez-Navas C, Gonzalez-Mendez L, Guerrero I: Cytonemes are required for the establishment of a normal Hedgehog morphogen gradient in Drosophila epithelia. Nat Cell Biol 2013, 15:1269-1281.
13 Sanders TA, Llagostera E, Barna M: Specialized filopodia direct long-range transport of SHH during vertebrate tissue patterning. Nature 2013, 497:628-632.
14 Ramirez-Weber FA, Kornberg TB: Cytonemes: cellular processes that project to the principal signaling center in Drosophila imaginal discs.
Cell 1999, 97:599-607.
15 Hsiung F, Ramirez-Weber FA, Iwaki DD, Kornberg TB: Dependence of Drosophila wing imaginal disc cytonemes on Decapentaplegic.
Nature 2005, 437:560-563.
16 Tian H, Callahan CA, DuPree KJ, Darbonne WC, Ahn CP, Scales SJ, de Sauvage FJ: Hedgehog signaling is restricted to the stromal compartment during pancreatic carcinogenesis.
Proc Natl Acad Sci U S A 2009, 106:4254-4259.
17 Jung IH, Jung DE, Park YN, Song SY, Park SW: Aberrant Hedgehog ligands induce progressive pancreatic fibrosis by paracrine activation of myofibroblasts and ductular cells in transgenic zebrafish. PLoS One 2011, 6:e27941.
18 Li X, Wang Z, Ma Q, Xu Q, Liu H, Duan W, Lei J, Ma J, Wang X, Lv S, et al: Sonic hedgehog paracrine signaling activates stromal cells to promote perineural invasion in pancreatic cancer. Clin Cancer Res 2014, 20:4326-4338. 19 Li Y, Chen L, Chan TH, Liu M, Kong KL, Qiu
JL, Yuan YF, Guan XY: SPOCK1 is regulated by CHD1L and blocks apoptosis and promotes HCC cell invasiveness and metastasis in mice.
Gastroenterology 2013, 144:179-191 e174.
20 Aoka Y, Johnson FL, Penta K, Hirata Ki K, Hidai C, Schatzman R, Varner JA, Quertermous T: The
embryonic angiogenic factor Del1 accelerates tumor growth by enhancing vascular formation.
Microvasc Res 2002, 64:148-161.
21 He Y, Liu XD, Chen ZY, Zhu J, Xiong Y, Li K, Dong JH, Li X: Interaction between cancer cells and stromal fibroblasts is required for activation of the uPAR-uPA-MMP-2 cascade in pancreatic cancer metastasis. Clin Cancer Res 2007, 13:3115-3124. 22 Shin K, Lim A, Zhao C, Sahoo D, Pan Y,
Spiekerkoetter E, Liao JC, Beachy PA: Hedgehog signaling restrains bladder cancer progression by eliciting stromal production of urothelial differentiation factors. Cancer Cell 2014, 26:521-533.
23 van Dop WA, Heijmans J, Buller NV, Snoek SA, Rosekrans SL, Wassenberg EA, van den Bergh Weerman MA, Lanske B, Clarke AR, Winton DJ, et al: Loss of Indian Hedgehog activates multiple aspects of a wound healing response in the mouse intestine. Gastroenterology 2010, 139:1665-1676, 1676 e1661-1610.
24 Buller NV, Rosekrans SL, Metcalfe C, Heijmans J, van Dop WA, Fessler E, Jansen M, Ahn C, Vermeulen JL, Westendorp BF, et al: Stromal Indian hedgehog signaling is required for intestinal adenoma formation in mice.
Gastroenterology 2015, 148:170-180 e176.
25 Lombardo Y, Scopelliti A, Cammareri P, Todaro M, Iovino F, Ricci-Vitiani L, Gulotta G, Dieli F, de Maria R, Stassi G: Bone morphogenetic protein 4 induces differentiation of colorectal cancer stem cells and increases their response to chemotherapy in mice.
Gastroenterology 2011, 140:297-309.
26 Apte MV, Wilson JS: Dangerous liaisons: pancreatic stellate cells and pancreatic cancer cells. Journal of gastroenterology and
hepatology 2012, 27 Suppl 2:69-74.
27 Heinemann V, Reni M, Ychou M, Richel DJ, Macarulla T, Ducreux M: Tumour-stroma interactions in pancreatic ductal adenocarcinoma: rationale and current evidence for new therapeutic strategies. Cancer Treat Rev 2014, 40:118-128.
