Abstract
Objectives
Pancreatic ductal adenocarcinoma (PDAC) is thought to derive from different precursor lesions including the recently identified atypical flat lesions (AFL). While all precursor lesions and PDAC share ductal characteristics, there is an ongoing debate about the cellular origin of the different PDAC precursor lesions. In particular, pancreatic acinar cells have previously been shown to display a remarkable plasticity being able to undergo ductal dedifferentiation in the context of oncogenic stimuli.
Methods
Histological analyses were performed in a murine PDAC model that specifically expresses oncogenic Kras in adult pancreatic acinar cells. Occurrence, characterization, and lineage tracing of AFLs were investigated.
Results
Upon expression of oncogenic Kras in adult pancreatic acinar cells, AFLs with typical morphology and expression profile arise. Lineage tracing confirmed that the AFLs were of acinar origin.
Conclusions
Using a murine PDAC model, this study identifies pancreatic acinar cells as a cellular source for AFLs.
Keywords: Pancreatic cancer, AFL, IPMN, PanIN
Introduction
Pancreatic ductal adenocarcinoma (PDAC) is among the deadliest human cancers and is characterized by a poor survival rate. One of the main reasons for the dismal survival rate is that the diagnosis is often made at an advanced stage, making surgical intervention impossible. Therefore, there is an urgent need to better understand how PDAC initiates in order to develop novel diagnostic and screening tools. PDAC is known to arise from different precursor lesions, namely pancreatic intraepithelial neoplasia (PanIN), intraductal papillary mucinous neoplasm (IPMN), and mucinous cystic neoplasm (MCN) [1,2]. Recently, additional possible alternative PDAC precursors described as atypical flat lesions (AFLs) have been identified in a murine model of PDAC and in patients with familial pancreatic cancer [3]. AFLs are defined as ductal lesions that (i) arise in areas of acinar-to-ductal metaplasia (ADM), (ii) are highly proliferative, (iii) display a tubular structure, (iv) show cytological atypia, and (v) are typically surrounded by a loose but highly cellular stroma.
Although all these precursor lesions share a ductal morphology, studies in genetically engineered mouse models of PDAC have shown that selectively targeting the principal driver mutation of PDAC, an oncogenic version of Kras (KrasG12D), to adult pancreatic acinar but not ductal or centroacinar cells results in the conversion of acinar to duct-like cells and the formation of PanIN lesions through a process termed acinar-to-ductal reprogramming (ADR) [4]. In contrast, further genetic changes, in addition to expression of oncogenic Kras, were necessary to elicit IPMN and PDA formation from duct cells in a transgenic mouse model [5]. Therefore, ductal PDAC precursor lesions can arise from different cell types, including duct and acinar cells.
AFLs have previously been shown to arise in mouse models of PDAC that target KrasG12D to embryonic pancreatic progenitor cells using Ptf1aCre [3]. Ptf1a directs Cre activity during pancreatic organogenesis to all exocrine compartments, namely pancreatic ductal, centroacinar, and acinar cells [6]. Given this broad expression profile, a conclusion regarding the cellular origin of AFLs is not possible in this model. Using a mouse model that recombines KrasG12D specifically in adult pancreatic acinar cells but not in ductal or centroacinar cells, we show that AFLs with typical morphological, cytological and expression features arise from acinar cells. Thus, this study further highlights the importance of pancreatic acinar cells as a cellular source for PDAC.
Materials and methods
Animal experiments
The following mice were used to generate the experimental animals for this study and were maintained on a mixed genetic background: Ptf1aCreER [7], LSL-KrasG12D [8], and R26RYFP [9]. All studies on mice presented here were approved by the UCSF Institutional Animal Care and Use Committee (IACUC).
Tamoxifen was administered by oral gavage to 5 to 6-week-old mice as previously described [10] and mice were sacrificed at the indicated time points after tamoxifen induction.
Microscopy and immunohistochemistry
For quantification of ADM and AFLs, random whole pancreatic sections stained with hematoxylin and eosin H&E were analyzed. For immunohistochemical staining, sections of paraffin embedded pancreata were rehydrated and antigen retrieval was performed using Antigen Unmasking Solution (Vector Laboratories). Overnight incubation with the following primary antibodies was done at 4°C: goat anti-CPA1 (R&D), rabbit anti-CK19 (Abcam), rabbit anti-GFP (Santa Cruz Biotechnology), mouse anti-Ki67 (BD), rabbit anti-P53 (Vector Laboratories), goat anti-Amylase (Santa Cruz Biotechnology), rabbit anti-p44/p42 MAPK (Cell Signaling) and FITC-conjugated DBA-lectin (Vector Laboratories). Biotin-conjugated secondary antibodies were incubated for 1 h at room temperature, following development with ABC and DAB kits (both Vector Laboratories). Nuclear counterstaining was performed using haematoxylin. For immunofluorescence, sections were incubated with fluorophore-conjugated secondary antibody for 1 h at room temperature. Slides were mounted with DAPI hardset antifade mounting medium.
Statistics
The statistics for the correlation of ADM and AFLs were performed using linear regression in Prism V6.0 (GraphPad Software Inc.). Quantification of YFP-positive cells was done using ImageJ. Five random fields from each mouse pancreas (n = 3 mice) were chosen for quantification. T-test was performed using Microsoft Excel.
Results
Atypical flat lesions arise from pancreatic acinar cells
To determine whether AFLs can arise from pancreatic acinar cells, KrasG12D was specifically expressed in adult pancreatic acinar cells using the Ptf1aCreER allele. In contrast to Ptf1aCre, this model allows Cre-mediated recombination upon tamoxifen administration exclusively in adult pancreatic acinar cells as previously reported [4,7]. First, we confirmed that Ptf1aCreER specifically recombines in acinar but not centroacinar or ductal cells, in our hands, using the lineage tracing allele R26RYFP (Supplementary Fig. 1). Next, we analyzed pancreatic abnormalities in Ptf1aCreER; LSL-KrasG12D mice. Of note, we identified areas of ADM and PanINs and also observed AFLs within ADM areas (Fig. 1A and B). These AFLs showed the typical morphological features: ductal appearance, tubular structures, and the presence of a surrounding loose stroma. Lineage tracing of adult acinar cells in Ptf1aCreER; LSL-KrasG12D; R26RYFP mice was performed that revealed YFP-positivity of AFLs and, thus, further supports the acinar origin of the lesions (Fig. 1C–F). Previously, AFLs have been localized within areas of ADM [3]. AFLs in Ptf1aCreER; LSL-KrasG12D mice were also associated with ADM (Fig. 1A, C) and the amount of ADM correlated with the amount of AFLs in the pancreas (Fig. 1G).
Fig. 1. Atypical flat lesions arise from pancreatic acinar cells.
(A) H&E overview: AFL lesions (highlighted square) are identified in areas of ADM of Ptf1aCreER; LSL- KrasG12D mice. (B) Higher magnification of the AFL highlighted in (A). (C) H&E staining of an AFL lesion (arrow) in Ptf1aCreER; LSL-KrasG12D; R26RYFP mice, higher magnification in (D); (E) YFP staining of AFLs from Ptf1aCreER; LSL-KrasG12D; R26RYFP mice showing YFP-positivity of AFL lesions (arrow), higher magnification in (F). Note that ADM (*) and PanINs (arrowhead) are also YFP-positive. (G) Correlation of the number of AFLs with the ADM area per whole pancreatic cross-sections in Ptf1aCreER; LSL- KrasG12D and Ptf1aCreER; LSL-KrasG12D; R26RYFP mice. Mice were analyzed at 5 months (A–F) and at 6 weeks and 4–5 months (G) after tamoxifen administration. Scale bars 500 μM (A), 400 μM (C, E), and 100 μM (B, D, F).
Characterization of acinar derived atypical flat lesions
Further characterization of the acinar derived AFLs in Ptf1aCreER; LSL-KrasG12D mice revealed expression of cytokeratin 19 but not carboxypeptidase A1 (CPA1), in agreement with the ductal phenotype of these lesions (Fig. 2A and B). Supporting the role as a PDAC precursor, cytological atypia was observed in cells forming AFLs (Fig. 2C). Furthermore, proliferation as shown by frequent expression of Ki-67 was present in AFLs in Ptf1aCreER; LSL-KrasG12D mice (Fig. 2D). In addition, P53 and pMAPK were broadly expressed in the nuclei of AFL cells (Fig. 2E and F).
Fig. 2. Characterization of acinar derived atypical flat lesions.
(A–F) Characterization of acinar derived AFLs in Ptf1aCreER; LSL-KrasG12D and Ptf1aCreER; LSL-KrasG12D; R26RYFP mice. (A) CK19 staining reveals broad expression of the ductal marker in AFLs. (B) The acinar marker CPA1 is not expressed in AFLs. (C) Evidence of cellular atypia in acinar derived AFLs (arrow, H&E staining). (D) AFLs have a proliferative character as evidenced by an increased expression of Ki67 in the nuclei (arrow). P53 (E) and pMAPK (F) are broadly expressed in nuclei of AFLs. (A–F) Scale bars represent 100 μM; mice were analyzed at 5 months after tamoxifen administration.
Discussion
It is well established by pathological progression models that PDAC can arise from IPMN and PanIN lesions. While the former is thought to derive from ductal cells, animal models suggest an acinar origin for PanINs [4]. The recently identified AFLs are suggested to be alternative precursors for PDAC [2]. AFLs are defined as proliferative ductal lesions in areas of ADM with a peculiar surrounding stroma. The presence within areas of ADM also suggests an acinar origin for AFLs. However, this has not yet been experimentally proven. The selective recombination of KrasG12D in adult pancreatic acinar cells in combination with lineage tracing described here supports the origin of AFLs from pancreatic acinar cells. In line with an acinar origin, previous studies have shown that recombination of oncogenic Kras, selectively in pancreatic ductal cells, does not lead to a neoplastic transformation [4,11]. However, it is unclear whether specific combinations of oncogenic insults in ductal cells could result in the formation of AFLs. Up to now, AFLs have only been shown in adult mice with expression of KrasG12D during pancreatic embryogenesis and in families with a strong family history of PDAC [3]. To date, it was also unclear whether oncogenic insults in adult pancreatic cells can result in AFLs. Our data shows for the first time that expression of KrasG12D exclusively in the adult pancreas is sufficient to induce AFLs.
While the clinical significance of AFLs remains to be elucidated, there is an increasing body of evidence supporting the notion that acinar cells serve as the cellular source for PDAC. Acinar cells have a remarkable plasticity in the context of oncogenic insults as they undergo ductal dedifferentiation and can form ADM, AFLs, and PanINs, all of which are presumed PDAC precursors [2]. To date, it is unclear whether the latter two develop from ADM or if these are independent routes of acinar cells on the way to become PDAC. Future studies need to show to what extent different precursors contribute to pancreatic carcinogenesis and if it is of relevance with regards to human tumor biology and therapy.
Supplementary Material
Supplementary Figure 1 Specific recombination in adult acinar cells. (A, B) YFP staining in Ptf1aCreER; LSL-KrasG12D; R26RYFP mice reveals specific expression of YFP in acinar (arrows) but not ductal (arrow heads) or centroacinar (*) cells. (C, D) Immunofluorescence costaining for (C) amylase (red) or (D) DBA lectin (red), YFP (green), DAPI (blue) in Ptf1aCreER; LSL-KrasG12D; R26RYFP mice reveals specific expression of YFP in acinar (arrows) but not ductal (arrow heads) or centroacinar (*) cells. (E) Quantification of YFP and amylase or DBA lectin double positive area in relation to total amylase or DBA lectin positive area. (A-D) Scale bars represent 100 μM; mice were analyzed at 6 weeks after tamoxifen administration.
Acknowledgments
We thank Annett Dannemann for expert technical assistance. We thank the Institute of Pathology at the Technical University of Munich, where whole pancreatic cross-section scanning was performed.
Funding
This work was supported by grants of the Deutsche Forschungsgemeinschaft to GVF (Emmy Noether Program, FI1719/2-1) and IE (ES 285/6-I). Work in MH's laboratory was supported by funding from the National Institutes of Health (NIH CA172045).
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://dx.org/10.1016/j.pan.2017.04.014.
Footnotes
Conflict of interest
The authors declare no conflict of interest.
References
- 1.Ying H, Dey P, Yao W, Kimmelman AC, Draetta GF, Maitra A, et al. Genetics and biology of pancreatic ductal adenocarcinoma. Genes Dev. 2016;30(4):355–85. doi: 10.1101/gad.275776.115. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Basturk O, Hong SM, Wood LD, Adsay NV, Albores-Saavedra J, Biankin AV, et al. A revised classification system and recommendations from the baltimore consensus meeting for neoplastic precursor lesions in the pancreas. Am J Surg Pathol. 2015;39(12):1730–41. doi: 10.1097/PAS.0000000000000533. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Aichler M, Seiler C, Tost M, Siveke J, Mazur PK, Da Silva-Buttkus P, et al. Origin of pancreatic ductal adenocarcinoma from atypical flat lesions: a comparative study in transgenic mice and human tissues. J Pathol. 2012;226(5):723–34. doi: 10.1002/path.3017. [DOI] [PubMed] [Google Scholar]
- 4.Kopp JL, von Figura G, Mayes E, Liu FF, Dubois CL, Morris JPT, et al. Identification of Sox9-dependent acinar-to-ductal reprogramming as the principal mechanism for initiation of pancreatic ductal adenocarcinoma. Cancer Cell. 2012;22(6):737–50. doi: 10.1016/j.ccr.2012.10.025. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.von Figura G, Fukuda A, Roy N, Liku ME, Morris JP, IV, Kim GE, et al. The chromatin regulator Brg1 suppresses formation of intraductal papillary mucinous neoplasm and pancreatic ductal adenocarcinoma. Nat Cell Biol. 2014;16(3):255–67. doi: 10.1038/ncb2916. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Heiser PW, Cano DA, Landsman L, Kim GE, Kench JG, Klimstra DS, et al. Stabilization of beta-catenin induces pancreas tumor formation. Gastroenterology. 2008;135(4):1288–300. doi: 10.1053/j.gastro.2008.06.089. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Pan FC, Bankaitis ED, Boyer D, Xu X, Van de Casteele M, Magnuson MA, et al. Spatiotemporal patterns of multipotentiality in Ptf1a-expressing cells during pancreas organogenesis and injury-induced facultative restoration. Development. 2013;140(4):751–64. doi: 10.1242/dev.090159. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Tuveson DA, Shaw AT, Willis NA, Silver DP, Jackson EL, Chang S, et al. Endogenous oncogenic K-ras(G12D) stimulates proliferation and widespread neoplastic and developmental defects. Cancer Cell. 2004;5(4):375–87. doi: 10.1016/s1535-6108(04)00085-6. [DOI] [PubMed] [Google Scholar]
- 9.Srinivas S, Watanabe T, Lin CS, William CM, Tanabe Y, Jessell TM, et al. Cre reporter strains produced by targeted insertion of EYFP and ECFP into the ROSA26 locus. BMC Dev Biol. 2001;1:4. doi: 10.1186/1471-213X-1-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.von Figura G, Morris JPT, Wright CV, Hebrok M. Nr5a2 maintains acinar cell differentiation and constrains oncogenic Kras-mediated pancreatic neoplastic initiation. Gut. 2014;63(4):656–64. doi: 10.1136/gutjnl-2012-304287. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Bailey JM, Hendley AM, Lafaro KJ, Pruski MA, Jones NC, Alsina J, et al. p53 mutations cooperate with oncogenic Kras to promote adenocarcinoma from pancreatic ductal cells. Oncogene. 2016;35(32):4282–8. doi: 10.1038/onc.2015.441. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Figure 1 Specific recombination in adult acinar cells. (A, B) YFP staining in Ptf1aCreER; LSL-KrasG12D; R26RYFP mice reveals specific expression of YFP in acinar (arrows) but not ductal (arrow heads) or centroacinar (*) cells. (C, D) Immunofluorescence costaining for (C) amylase (red) or (D) DBA lectin (red), YFP (green), DAPI (blue) in Ptf1aCreER; LSL-KrasG12D; R26RYFP mice reveals specific expression of YFP in acinar (arrows) but not ductal (arrow heads) or centroacinar (*) cells. (E) Quantification of YFP and amylase or DBA lectin double positive area in relation to total amylase or DBA lectin positive area. (A-D) Scale bars represent 100 μM; mice were analyzed at 6 weeks after tamoxifen administration.