Abstract
Persistent acinar to ductal metaplasia (ADM) is a recently recognized precursor of pancreatic ductal adenocarcinoma (PDAC). Here we show that the ADM area of human pancreas tissue adjacent to PDAC expresses significantly higher levels of regenerating protein 3A (REG3A). Exogenous REG3A and its mouse homolog REG3B induce ADM in the 3D culture of primary human and murine acinar cells, respectively. Both Reg3b transgenic mice and REG3B-treated mice with caerulein-induced pancreatitis develop and sustain ADM. Two out of five Reg3b transgenic mice with caerulein-induced pancreatitis show progression from ADM to pancreatic intraepithelial neoplasia (PanIN). Both in vitro and in vivo ADM models demonstrate activation of the RAS-RAF-MEK-ERK signaling pathway. Exostosin-like glycosyltransferase 3 (EXTL3) functions as the receptor for REG3B and mediates the activation of downstream signaling proteins. Our data indicates that REG3A/REG3B promotes persistent ADM through binding to EXTL3 and activating the RAS-RAF-MEK-ERK signaling pathway. Targeting REG3A/REG3B, its receptor EXTL3, or other downstream molecules could interrupt the ADM process and prevent early PDAC carcinogenesis.
Similar content being viewed by others
Introduction
Pancreatic ductal adenocarcinoma (PDAC) remains one of the most lethal human cancers with a 5-year survival rate <5%. The poor prognosis of PDAC has been attributed to multiple factors including late diagnosis, the lack of sensitive and specific biomarkers to detect PDAC, and the lack of effective measures to prevent its development and interrupt its progression1. Three precursor lesions of PDAC have been identified: intraductal papillary mucinous neoplasm, pancreatic intraepithelial neoplasia (PanIN), and mucinous cystic neoplasm2. PanIN is by far the most well-characterized precursor lesion with a histological step-wise progression from low-grade to high-grade PanINs and then to invasive adenocarcinoma, which correspond molecularly to a series of genetic events including the activation of the Kras oncogene and the inactivation of the tumor suppressor genes Cdk2na, Tp53, and transcriptional factor Smad4/Dpc43.
It has been well documented that PanINs can arise from pancreatic acini undergoing acinar to ductal metaplasia (ADM)4,5. The normal pancreatic acinus is composed of wedge-shaped cells containing zymogen granules arranged around a small lumen. Acinar differentiation contributes to pancreatic tissue homeostasis and the function of the exocrine pancreas. Histologically, ADM is characterized by acinar cells losing zymogen granules and adopting the appearance of duct-like structures. Loss of acinar markers such as Mist1, Amylase, Carboxypeptidase, Elastase, and gain of ductal markers such as Cytokeratin 19 (Ck19), Mucin 1, Duodenal homeobox 1 (Pdx1), and Sry-related high-mobility group box 9 (Sox9) are hallmarks of ADM6,7,8,9. In patients with pancreatitis or other forms of pancreatic injury, ADM is transient and reversible10,11. However, persistent ADM may progress to PanIN, and finally to an invasive tumor9,12,13,14. It is yet unclear what molecular mechanism drives the formation and malignant progression of ADM.
The family of regenerating (REG) proteins is a group of C-type lectin-like proteins discovered in patients with pancreatitis and during pancreatic islet regeneration after injury. Five REG family members including REG1A, REG1B, REG3A, REG3G, and REG4 have been identified in humans15. REG3A (homolog of mouse REG3B), also known as pancreatitis-associated protein (PAP) due to its upregulation in response to inflammatory stimulants, is generally expressed at low level in normal pancreas16,17. REG3A has been described as a proliferating factor following liver and skin injuries as well as a driver of pancreatic cancer cell growth through JAK2/STAT3 signaling pathways in response to interleukin-616,18. Other members of the REG proteins family, i.e., REG1A and REG4, have been shown to be overexpressed in PDAC patients and to accelerate cell proliferation and tumor growth in vitro17,19,20. We previously reported that ADM can be induced by adding recombinant REG3B in the 3D culture of acinar cells in the context of wild type Kras21. In the current study, we demonstrated the involvement of REG3A/REG3B in the ADM process using several models including: human pancreas tissue from patients with PDAC, in vitro cultures of primary human and murine pancreatic acinar cells, an in vivo transgenic mouse model with caerulein-induced pancreatitis, and recombinant REG3B-treated wild-type mice with caerulein-induced pancreatitis. We also identified a candidate receptor for REG3B on the acinar cell membrane, and unveiled the downstream signaling pathway that leads to ADM formation and maintenance.
Results
REG3A is overexpressed in the ADM area adjacent to PDAC in human pancreas
To determine the expression pattern of REG3A, we performed immunohistochemical staining on human pancreatic tissue sections that contained areas of transition from normal acini to ADM and to PDAC (Fig. 1a with enlarged view of ADM in Fig. 1b). There was a higher expression of REG3A in the ADM area than the normal, PanIN and PDAC areas (Fig. 1c with enlarged view in Fig. 1d). We then performed REG3A/CK19 double staining on pancreatic tissue sections of eight patients (Supplementary Fig. 1a–p). Immature ADM (indicated by black arrow in Supplementary Fig. 1q) showed no or very weak expression of CK19. As the ADM matures and develops a clearer duct-like morphology (indicated by asterisk in Supplementary Fig. 1q), there is a gradual increase in CK19 expression. Colocalization of CK19 and REG3A was well displayed in the enlarged view of mature ADM (Supplementary Fig. 1q). Quantification analysis showed that 99% of the ADM area stained positively for REG3A, whereas normal acini had minimal (4%) REG3A positivity (Supplementary Fig. 1r). REG3A expression was reduced in PanIN and PDAC, compared with ADM. We then measured the level of Reg3a mRNA in micro-dissected areas of normal pancreatic tissue with matched ADM zone by RT-qPCR. The level of Reg3a mRNA in ADM area was nearly 2-fold higher than in normal pancreatic tissue (Supplementary Fig. 2a). This data suggests that REG3A protein and mRNA are highly expressed in human ADM.
REG3B is elevated in caerulein-induced pancreatitis and maintains prolonged ADM in mice
To determine whether REG3B is involved in the ADM process in vivo, we challenged wild type C57BL/6J mice (WT) with caerulein to induce pancreatitis (Fig. 2a), which is a reliable and widely used mouse model for ADM induction22. At day 3, caerulein-treated mice had a larger pancreas-to-body weight ratio than the control group (Fig. 2b, c). Histologically, the exocrine compartment of caerulein-treated mice displayed areas of ADM, whereas the control group (PBS-injected mice) did not (Fig. 2d). Quantification analysis showed a statistically significant increase in ADM events in the caerulein-treated WT mice compared to PBS-treated mice (Fig. 2e). Western blot analysis showed an increase of REG3B protein with a concomitant reduction of AMYLASE in caerulein-treated WT mice (Fig. 2f and Supplementary Fig. 4a). Gain of ductal marker Ck19 mRNA and loss of acinar marker Mist1 mRNA in caerulein-treated mice was demonstrated by RT-qPCR (Fig. 2g). Immunohistochemically, REG3B was highly enriched in ADM areas, particularly in naïve ADM (Fig. 2h). Therefore, ADM can be induced 1 day after the final caerulein injection, and correlates with REG3B overexpression in the pancreas tissue. However, at day 10, no ADM phenotype was observed in WT mice or PBS control groups (Fig. 3b, WT PBS and WT Cae groups). This suggests that the caerulein-induced ADM in WT mice detected at day 3 (Fig. 2e) was transient, and the pancreas almost completely recovers after 1 week.
To verify whether Reg3b promotes and maintains ADM in vivo, we used a gain-of-function Reg3b transgenic (TG) mouse model23. Western blot analysis confirmed a 2.7-fold higher REG3B expression level in the pancreas of TG mice than that of WT mice (Supplementary Fig. 2b). We subjected TG and WT mice to caerulein-induced pancreatitis, along with a REG3B group (WT injected with caerulein and recombinant REG3B protein) and two control groups (WT and TG injected with PBS) (Fig. 3a). In contrast to the disappearance of ADM at day 10 in caerulein-treated WT mice (Fig. 3b), both the WT Cae injected with recombinant REG3B and the TG Cae mice showed persistent ADM one week after the final caerulein injection (Fig. 3b, WT Cae+REG3B and TG Cae groups). Furthermore, focal PanIN was observed in two out of five REG3B TG mice injected with caerulein. Reduction in the mRNA expression of Mist1 and increase in Ck19 were also found in both WT Cae+REG3B and TG Cae groups compared to the control groups (WT PBS, WT Cae, and TG PBS) 1 week after the final caerulein injection (Fig. 3c). This was further validated by immunofluorescence staining of CK19 and AMYLASE. WT Cae+REG3B and TG Cae mice showed a higher expression of CK19 and a lower expression of AMYLASE in the ADM area than the control groups (WT PBS, WT Cae, and TG PBS) (Fig. 3d). To further explore the capacity of ectopic delivery of REG3B on maintaining ADM, we analyzed the histomorphology of WT Cae+REG3B and WT Cae groups at 2, 4, and 10 weeks after the final caerulein injection. Strikingly, scattered ADM lesions were observed at 2, 4, and 10 weeks after the final injection of caerulein (Fig. 3e and quantification data in Supplementary Fig. 2c). This data clearly demonstrates that REG3B treatment was able to maintain ADM and even promote ADM to PanIN in some TG Cae mice.
Exogenous REG3A/REG3B induces ADM in human and mouse primary acinar cells in vitro
To directly visualize the effect of REG3B on ADM formation and facilitate molecular pathway analysis, we established an in vitro 3D culture model of ADM for both human and murine acinar cells using TGFɑ-induced ADM as a positive control24. Recombinant REG3A or REG3B protein induced ADM in human and mouse primary acinar cells, respectively, as evidenced by spherical-like ductal morphology comparable with the TGFɑ-treated group (Fig. 4a). Quantification analysis revealed a 3-fold to 4-fold increase in ADM events in REG3A-treated or REG3B-treated primary acinar cells compared with untreated acini (Fig. 4b). Furthermore, increased expression of CK19 and decreased expression of AMYLASE in REG3A-treated or REG3B-treated primary acinar cells was revealed by immunofluorescence staining (Fig. 4c). Consistently, in primary murine acinar cells, we found a higher level of mRNA expression of ductal-specific markers Ck19 and Nestin, and a lower level of mRNA expression of acinar-specific markers Mist1, Ptf1a, and Cpa by RT-qPCR analysis (Fig. 4d). This finding was also observed in human primary acinar cells (Fig. 4d). Therefore, both REG3A and REG3B were able to induce ADM in the 3D culture model.
REG3B induces ADM through the RAS-RAF-MEK-ERK signaling pathway
KRAS hyperactivity, mitogen-activated protein kinase (MAPK) signaling, and inflammatory signaling pathways have all been described to different extents as contributing factors on ADM25,26,27. To elucidate the molecular mechanism of REG3B-mediated ADM, we first detected the activation of relevant signaling pathway proteins by western blot in both in vivo and in vitro models. We then validated our findings using Reg3b gene knockdown and specific protein inhibitors. Firstly, in in vivo experiments, we found an increase in active RAS (RAS-GTP), KRAS, p-MEK, and p-ERK in the pancreatic tissue of WT Cae+REG3B mice, TG PBS mice and TG Cae mice in comparison with WT PBS mice (Fig. 5a with quantification data in Supplementary Fig. 2b). This finding was confirmed in vitro in the 266-6 cell line treated with REG3B and in human primary acinar cells treated with REG3A, demonstrated by an increase in the expression of active RAS (RAS-GTP), p-BRAF, p-MEK, and p-ERK (active RAS was not tested in human primary acinar cells) (Fig. 5b with quantification data in Supplementary Figs. 2d, e, 4c and 5a). Secondly, siRNA knockdown of the Reg3b gene dramatically reduced the expression of phosphorylated BRAF, MEK, ERK, and total KRAS in the 266-6 acinar cell line (Fig. 5c with quantification data in Supplementary Figs. 2f and Supplementary Fig. 5b). Thirdly, both LY300912, a RAF inhibitor, and Trametinib, a MEK inhibitor, inhibited BRAF, MEK, and ERK phosphorylation in a dose dependent manner and attenuated the effects of REG3B on acinar cell lines (Fig. 5d, e with quantification data in Supplementary Figs. 2g–h and 5c, d). Finally, LY3009120 or Trametinib treatment was able to completely block ADM induced by REG3B in the 3D culture model, as shown in bright field images of the morphology (Fig. 5f, g) and validated by the loss of ductal-specific CK19 and the increase of acinar-specific AMYLASE protein expression in confocal immunofluorescence images (Fig. 5h). Baseline information of AMYLASE and CK19 staining in untreated mouse primary acinar cells was given (Supplementary Fig. 2i). Collectively, our results demonstrate that REG3B promotes ADM through the RAS-RAF-MEK-ERK signaling pathway.
EXTL3 functions as a receptor for REG3B in ADM
To find the entry point for REG3B to activate the RAS-RAF-MEK-ERK signaling pathway, we sought to identify its potential binding protein on the acinar cell membrane. It has been shown that upon REG1 or REG3A stimulation, EXTL3 activated the PI3K pathway in keratinocytes and B cells18,28,29. Therefore, we first counterstained REG3B and EXTL3 by dual immunofluorescence in 2D culture of REG3B-treated acinar cell lines and in 3D culture of REG3A-induced ADM in primary human acinar cells in vitro. Both the REG3B-treated acinar cell lines and the REG3A-induced ADM in primary human acinar cells showed strong co-localization of REG3B and EXTL3 (Fig. 6a). The interaction between REG3B and EXTL3 in acinar cells was further demonstrated by the detection of EXTL3 by co-immunoprecipitation with REG3B antibody in mouse primary acinar cells and AR42J acinar cell line. In 266-6 and mouse primary acinar cells, we were able to pull down REG3B with EXTL3 antibody as well (Fig. 6b and Supplementary Fig. 6a). Silencing the EXTL3 receptor with a neutralizing antibody completely abolished REG3B-induced ADM, as demonstrated by the loss of ductal morphology, and a very low expression of CK19 and high expression of AMYLASE in the 3D culture model (Fig. 6c with quantification data in Supplementary Fig. 2j). REG3B-induced ADM was impaired by blocking EXTL3 with a neutralizing antibody (Supplementary Fig. 3a). Taken together, these findings suggested that EXTL3 was not only a receptor for REG3B in acinar cells, but also facilitated ADM formation induced by REG3B. To confirm that EXTL3 is involved in regulating downstream RAS signaling molecules, we examined the effect of silencing Extl3 by siRNA or EXTL3 neutralizing antibody on REG3B-induced RAS-RAF-MEK-ERK signaling pathway in the 266-6 acinar cell line (one out of three constructs of si-EXTL3 worked). We observed a slight reduction in the expression of total KRAS and phosphorylated BRAF, MEK, and ERK proteins after silencing Extl3 in the 266-6 acinar cells (Fig. 6d with quantification data in Supplementary Figs. 3b and 6b). In the context of REG3B-treated 266-6 acinar cells, REG3B-induced phosphorylated MEK and ERK and total KRAS expression was successfully inhibited upon silencing Extl3 expression by siRNA (Fig. 6e with quantification data in Supplementary Fig. 3c and Supplementary Fig. 6c). EXTL3 neutralizing antibody treatment also successfully inhibited REG3B-induced phosphorylation of MEK and ERK in 266-6 and AR42J cell lines and mouse and human primary acinar cells, as well as total KRAS expression in mouse and human primary acini (Fig. 6f with quantification data in Supplementary Fig. 3e–g and Supplementary Fig. 7). Therefore, the ADM process was attributed to the interaction between REG3B and EXTL3 on the acinar cell membrane and subsequent activation of the downstream RAS-RAF-MEK-ERK signaling pathway.
JAK2/STAT3 is another well studied signaling pathway involved in PDAC tumorigenesis and REG3A-related research. Loncle et al.30 found that REG3B served as a downstream molecule of interleukin-17 during ADM, and further promoted pancreatic cancer cell growth through the GP130/JAK2/STAT3 pathway. Thus, we also explored the impact of REG3B on JAK2/STAT3 signaling in pancreatic acinar cells with TGFα treated cells as positive control. TGFα activated JAK/STAT3 signaling pathway after 72 h in both AR42J and 266-6 cell lines. Remarkably, exogenous REG3B alone failed to activate p-JAK or p-STAT3 after 30 min and 72 h of treatment, in either 266-6 or AR42J cell lines (Fig. 7 and Supplementary Fig. 8). However, REG3B combined with EXTL3 neutralizing antibody enhanced p-JAK2 and p-STAT3 expression (Fig. 7c–e). These data suggest that REG3B binding to EXTL3 may preferentially activate MAPK signaling, rather than JAK2/STAT3.
Discussion
Initially discovered as an inflammation-induced regenerative protein in pancreas and other organs31,32, REG3A/REG3B attracted new attention due to its involvement in PDAC development and progression16,30,33,34,35,36. In Reg3B-deficient mice, Loncle et al.30 showed that PanIN lesions in the oncogenic Kras-driven PDAC model were attenuated and Gironella et al.33 showed that pancreatic tumor growth was impaired. In the 3D culture of mouse acinar cells, we have previously reported our observation that REG3A promoted ADM formation with concurrent activation of the mitogen-activated protein kinase pathway21. This study is designed to further delineate the role of REG3A/REG3B in the formation and maintenance of ADM and its molecular mechanism.
We were unable to detect any REG3A protein or mRNA in normal human pancreatic acini. However, both REG3A protein and mRNA were markedly elevated in ADM tissue adjacent to PDAC. REG3A protein and mRNA were present in the PDAC zone, but the level of expression was lower than that of the ADM zone. This data indicates that REG3A is primarily involved in the ADM process, i.e. the early stage of PDAC tumorigenesis.
We were able to induce ADM formation in both normal human and mouse primary acinar cells by adding recombinant REG3A or REG3B protein, respectively, into the 3D culture media in vitro. We observed a transient and reversible ADM phenotype in the wild type mice with caerulein-induced pancreatitis in the absence of exogenous REG3B. In contrast, Reg3b TG Cae or WT Cae+REG3B mice sustained ADM for more than 1 week after the final caerulein injection. Furthermore, PanIN were observed in some TG Cae mice. ADM persisted in WT Cae+ REG3B mice for up to 10 weeks. The combination of both in vitro 3D culture and the dynamic in vivo data convincingly proved that REG3B is involved in the formation and maintenance of persistent ADM. The fact that no ADM was observed in REG3B-treated WT mice without caerulein treatment and Reg3b TG mice without caerulein treatment indicates that acinar injury, as a result of inflammation induced by caerulein or trauma inflicted during the isolation of the acinar cells from the pancreas, is the pre-requisite for ADM initiation. Endogenous or exogenous REG3B can only augment the extent and promote persistent ADM after the initial injury occurs.
Mutation of the Kras oncogene has been considered to be the critical driver of PDAC tumorigenesis37. Most PDAC tumorigenesis studies were conducted in cancer cells and mouse models harboring Kras mutations38,39,40. The fact that we were able to induce ADM in both in vitro and in vivo models without the need to introduce Kras mutation supports the notion, that Kras mutation is not an absolutely required molecular event in the ADM process41,42. On the other hand, the fact that no invasive PDAC developed in any of our Reg3b TG Cae or WT Cae+REG3B mice suggest that additional molecular drivers including mutant Kras will be required for the development of full-scale invasive PDAC43.
The RAS-RAF-MEK-ERK signaling pathway is involved in the initiation and malignant progression of various types of cancers44,45. Lineage-tracing studies and other reports demonstrated the requirement of RAS-RAF-MEK-ERK signaling in Kras-induced ADM46,47. Although it has been shown that REG3A, along with other Reg family members, could mediate components of RAS-RAF-MEK-ERK signaling pathway in pancreatic cancer cells48, pancreatic β cells49 and gastric cancer cells50, the detailed mechanism by which REG3A/REG3B regulates pancreatic acinar cells undergoing ADM was poorly described. In vivo, we observed the expression of downstream proteins of the RAS-RAF-MEK-ERK signaling pathway by western blot, and found an increase of active RAS (RAS-GTP) protein in the pancreatic tissue of the WT Cae+REG3B and TG Cae mice. In vitro, REG3A treatment of primary human acinar cells, REG3B treatment of primary murine acinar cells as well as the acinar cell line 266-6, all consistently stimulated the expression of active RAS, KRAS, p-BRAF, p-MEK, and p-ERK. Suppressing the Reg3b gene by siRNA dramatically reduced the expression of phosphorylated downstream proteins of the RAS-RAF-MEK-ERK signaling pathway. Treatment with the RAF inhibitor LY3009120 or the MEK1/2 inhibitor Trametinib successfully blocked REG3B-activated p-MEK or p-ERK. In conjunction with decreased protein expression, both inhibitors were able to reverse REG3B-induced ADM in the 3D culture of mouse primary acinar cells at the concentrations required to block REG3B-stimulated activity of these kinases. These approaches from different angles have clearly proven that REG3A/REG3B promotes ADM through the RAS-RAF-MEK-ERK signaling pathway. However, our data by no means indicates that RAS-RAF-MEK-ERK signaling is the exclusive pathway. In fact, Liu et al.51 reported that REG3B could also activate the GP130-JAK2-STAT3 signaling pathway to accelerate pancreatic cancer cell growth. Future, more extensive studies may unveil other signaling pathways that are also involved in the REG3B-driven ADM process.
For REG3B/REG3A to exert its function, it requires an entry point on the acinar cells, i.e., a receptor to facilitate the interaction with downstream molecules within the cell. Since EXTL3 is more highly expressed on the cell membrane of pancreatic acini than on other cell types and has been implicated as the putative receptor for REG3A in epithelial keratinocytes and for REG1 in neurites18,52,53, we evaluated whether EXTL3 could interact with REG3B. Dual immunofluorescence staining demonstrated co-localization of REG3B and EXTL3 in REG3B-treated acinar cell lines and in the in vitro 3D culture model of human primary acinar cells. Co-immunoprecipitation experiments revealed an interaction between REG3B and EXTL3. Silencing EXTL3 with a neutralizing antibody in human and mouse primary acinar cells resulted in a reduction of ADM events and inhibition of BRAF, MEK, and ERK phosphorylation in human and mouse primary acinar cells and 266-6 and AR42J cells. This was also shown in 266-6 cells by Extl3 siRNA treatment. Since other proteins, such as REG1A can also bind to EXTL3 and activate the RAS-RAF-MEK-ERK signaling pathway, silencing EXTL3 may also reduce its downstream signaling that was activated by other proteins. We also screened several other previously reported potential receptors for REG3B such as GP13030 and EGFR51, but did not find any positive co-immunoprecipitation with REG3B in acinar cell lines (Supplementary Fig. 3d). The fact that we were unable to validate the result of others could be due to multiple factors, such as differences in the exact experimental conditions. In addition, the targeting receptor for REG3A may vary among different cell types and diseases. In our opinion, negative CO-IP results cannot completely exclude the possibility of other proteins being a potential receptor for REG3B.
It seems that the candidate receptor EXTL3 was not involved in JAK2/STAT3 activation after exogenous REG3B stimulation. In vitro ADM formation induced by REG3B is primarily dependent on downstream RAS-RAF-MEK-ERK signaling through EXTL3. JAK2/STAT3 was only activated when the primary receptor EXTL3 was blocked by a neutralizing antibody. Our data provides evidence that REG3B interacts with EXTL3 on the acinar cell membrane and this interaction promotes ADM by activating the downstream RAS-RAF-MEK-ERK signaling pathway.
Previously, we reported high expression of REG3B in mouse islet cells and its protective role against streptozotocin-induced diabetes mellitus23. In this study, we found high protein and mRNA expression of REG3B in both human and mouse pancreatic acinar cells that underwent ADM. In both settings, REG3A/REG3B functions like an “autocrine” molecule, where the protein binds to the receptor on the cell membrane and affects the cell that produces it. The identification of the EXTL3 receptor on the acinar cell membrane provides further supporting evidence for the “autocrine” mechanism of REG3A/REG3B.
In summary, our results clearly demonstrate that, in the absence of Kras mutation, REG3A/REG3B can induce and maintain ADM by interacting with EXTL3 and activating the downstream RAS-RAF-MEK-ERK signaling pathway (Fig. 8). In patients with pancreatitis or other types of pancreas injury, it might be worthwhile to measure the REG3A protein level in the pancreatic fluid through endoscopic retrograde cholangiopancreatography (ERCP), or in pancreatic biopsy tissue, or in patient’s serum in order to predict the risk of early PDAC tumorigenesis. Targeting REG3A/REG3B, its receptor EXTL3, or other downstream molecules involved in this dynamic process could potentially interrupt the ADM process and prevent early PDAC carcinogenesis.
Methods
Human samples, cell lines, and mouse models
This study was approved by the Institutional Research Ethics Board of the McGill University Health Center (MP-37-2018-4399, MP-37-2018-3171). Normal human pancreatic tissues (n = 6) were obtained from the Department of Surgery, McGill University Health Center. Formalin fixed paraffin embedded (FFPE) human PDAC tissue blocks were obtained from the archived tissue at the Department of Pathology, McGill University Health Center. Informed consents were signed by all patients involved in the study.
The pancreatic acinar cell lines AR42J (CRL-1492™) and 266-6 (CRL-2151™) were purchased from ATCC and cultured in a 37 °C and 5% CO2 incubator. AR42J and 266-6 cells were maintained in RPMI1640 medium (Invitrogen) and Dulbecco’s Modified Eagle Medium (DMEM, Wisent), respectively. Both were supplemented with 20% fetal bovine serum, 100 µg/mL penicillin, and 100 µg/mL streptomycin.
Animal protocols were approved by the McGill University Animal Care Committee (AUP 2012-7052). C57BL/6J mice were purchased from the Jackson Laboratory. RIP-I/Reg3B transgenic mice were established and validated by genotyping, as previously described by our group23. Adult male C57BL/6J mice (6–12 weeks old) were deprived of food 12 h prior to the in vivo experiments, but given free access to water. Acute pancreatitis was induced by intraperitoneal (i.p.) injection of caerulein (Sigma, C9026) dissolved in PBS at a dose of 50 μg/kg at hourly intervals eight times daily, for two consecutive days54. Some mice were sacrificed one day post-caerulein injection (day 3), others were followed by one i.p. injection of recombinant REG3B protein (100 μg/kg) dissolved in PBS daily for seven consecutive days and sacrificed 1 week, 2 weeks, 4 weeks, or 10 weeks later. Recombinant human REG3A and mouse REG3B protein were purchased from R&D Systems (5940-RG and 5110-RG). PBS served as a negative control for injection.
Primary acinar cell isolation and three-dimensional culture
Primary pancreatic acinar cells were isolated from the pancreas of six patients (Supplementary Table 1). Briefly, the pancreas was dissected, washed twice with ice-cold HBSS, minced into 1–5 mm pieces and digested with collagenase P (37 °C with a shaker)46. The digestion was terminated by adding an equal volume of ice-cold HBSS with 5% FBS. The digested pancreatic tissue was washed and pipetted through a 500 µm mesh and then a 105 µm mesh. The supernatant was added dropwise to 20 mL HBSS containing 30% FBS. The acinar cells were then pelleted (1000 rpm/min for 2 min at 4 °C) and suspended in 10 mL RPMI1640 complete medium (1% FBS, 0.1 mg/mL trypsin inhibitor and 1 µg/mL dexamethasone), referred to as “3D culture media”. The cell suspension was gently mixed with an equal volume of matrigel (Corning) and plated in a matrigel pre-coated 8-well chamber (300 µL per well). After solidification, 200 µL of 3D culture media, with or without REG3B (100 nM) or TGFα (50 ng/mL) was added. Media was replaced every other day. On day 5, the amount of ductal structures was counted in all wells from fifteen 20× fields on the EVOS microscope (Thermo Fisher Scientific) and bright field images were recorded.
Histology, histochemistry, immunohistochemistry, and immunofluorescence
Formalin fixed tissue was processed, embedded in paraffin, and cut into 5 µm sections. Hematoxylin and Eosin (H&E) (Thermo Fisher Scientific, 7221, 7111) staining was performed according to the clinical laboratory standard. Immunohistochemistry (IHC) was performed using primary antibodies against REG3A (1:150, R&D Systems, MAB5965), Amylase (1:100, Cell signaling technology) and CK19 (1:100, DSHB). Immunofluorescence staining was performed using primary antibodies against CK19 (1:100, DSHB, TROMAII), Amylase (1:100, Cell signaling technology, 3796S), REG3B (1:100, R&D Systems, AF5110 or AF1996) and EXTL3 (1:70, Biotech, NBP1-31645) according to the manufacturers’ instructions.
Western blotting
Proteins were extracted, fractionated by SDS-PAGE, and transferred to nitrocellulose membrane (Bio-Rad). Blots were blocked, incubated with primary antibodies (1:1000) at 4 °C overnight, then washed and incubated with secondary antibodies (HRP-conjugated goat anti-mouse, anti-rabbit, and rabbit anti-goat IgG were used, depending on the primary antibodies used). The membranes were briefly incubated with ECL detection reagent (Bio-Rad) to visualize the proteins and images were then captured by ImageQuant Las4000 (GE Healthcare Bio-Science). The primary antibodies were: p-BRAF (T401,Abcam, ab68215), BRAF (Cell Signaling Technology, 14814S), p-MEKk1/2 (Cell Signaling Technology, 9154S), MEK1/2 (Cell Signaling Technology, 4694S), p-ERK1/2 (Cell Signaling Technology,T202/Y204, 4370S), ERK1/2 (Cell Signaling Technology, 5013S), GAPDH (Invitrogen, MA5-15738), REG3B (R&D Systems, AF5110, AF1996), EXTL3 (Biotech, NBP1-31645), Amylase (Cell signaling technology, 3796S), p-JAK2 (Cell Signaling technology, 3776), JAK2 (Cell Signaling technology, 3230), p-STAT3 (Cell Signaling technology, 9145), STAT3 (R&D Systems, MAB1799) and KRAS (Biotech, NBP2-45536). Active RAS was measured using the Active Ras Detection Kit (Cell Signaling Technology #8821) according to the manufacturers’ instructions. Full, uncropped blot images are included in Supplementary Figs. 4–8.
Co-immunoprecipitation (Co-IP) assays
Total protein lysates of 266-6 and AR42J cell lines and primary acini were obtained from 100 mm dishes at 90% confluence. Normal serum was added to the lysate and kept on ice for 1 h. Two milligram of proteins were pre-cleared by incubation with 50 µL of Protein-G agarose beads (Invitrogen) for 1 h at 4 °C with gentle agitation. An aliquot of supernatant was kept as the input lysate prior to immunoprecipitation. Protein lysates were incubated overnight at 4 °C on a rocking wheel, with anti-REG3B or anti-EXTL3 antibody (20 μg/mL). Then, 50 µL of beads were added to the samples, and the mixture was rotated for 2 h at 4 °C. After three washing steps in lysis buffer to reduce non-specific binding to the beads, the lysate-bead/antibody conjugate mixture (used as after-clearing lysate) was eluted with 50 µL 2× Laemmli sample buffer without DTT for 10 min at 50 °C (elution 1). Fifty microliter of 2× Laemmli sample buffer with DTT (100 mM) was added to the pelleted beads and boiled for 5 min (elution 2). Input lysate, after-clearing lysate, elution 1 and elution 2 were analyzed by western blotting for EXTL3 or REG3B.
RNA isolation and real-time RT-qPCR
Following the manufacturer’s instructions, total RNA was extracted from mouse pancreas or 266-6 cell line with the RNeasy Plus Mini Kit (Qiagen). Single-stranded cDNA was then synthesized from 1 μg of extracted mRNA using a cDNA synthesis kit (SuperScript IV, Thermo Fisher Scientific). DNA from formalin-fixed paraffin embedded (FFPE) tissues was extracted with the RecoverAll Total Nucleic Acid Isolation Kit (Ambion). Real-time PCR was performed on an Applied Biosystems 7500 Sequence Detection system with SYBR Green Master Mix (Applied Biosystems) containing 1/10 of total volume cDNA and 10 nmol of each pair of primers (sequences shown in Supplementary Table. S2). The cycling conditions were: 50 °C for 2 min, 95 °C for 2 min for initial denaturation; 95 °C for 15 s, 57 °C for 15 s, and 72 °C for 1 min for 40 cycles followed by the melting curve stage. The relative expression of target gene was evaluated based on the threshold cycle (Ct), and calculated as 2−ΔΔCT. All genes were normalized to the internal control YWHAZ (Tyrosine 3-Monooxygenase/Tryptophan 5-Monooxygenase Activation Protein Zeta) or RPL13A (Ribosomal Protein L13a). All samples were tested in at least three independent experiments.
RNA interference and transient transfection
Small interfering RNA (siRNA) oligonucleotides targeting Extl3 and Reg3b (Supplementary Table. S3) were designed and synthesized by Origene. 266-6 and AR42J cell lines were transfected with siRNA (20 nM) or scrambled siRNA according to the manufacturer’s instructions. Cells were grown in a 6-well plate and transfected with siRNA at 200 pmol diluted in RNAimax and Optimem medium (Invitrogen). Knockdown efficiency of Reg3b or Extl3 was confirmed by western blotting.
Statistics and reproducibility
Each experiment was performed at least three independent times with at least duplicate samples. Data were expressed as mean ± SD. A one-way ANOVA or Student’s t-test was used between two groups to compare the difference by SPSS 21.0, and formatted with GraphPad Prism version 6.0 software. Tukey multiple-comparison posttest was performed for statistical analysis of multiple groups. A two-tailed P value of <0.05 was considered as statistically significant. *P < 0.05; **P < 0.01; ***P < 0.001. ns no significance.
Reporting summary
Further information on research design is available in the Nature Research Reporting Summary linked to this article.
Data availability
The datasets generated and/or analyzed during the current study can be found in Supplementary Data. 1 or are available from the corresponding author upon request.
References
Ghaneh, P., Costello, E. & Neoptolemos, J. P. Biology and management of pancreatic cancer. Gut 56, 1134–1152 (2007).
Distler, M., Aust, D., Weitz, J., Pilarsky, C. & Grutzmann, R. Precursor lesions for sporadic pancreatic cancer: PanIN, IPMN, and MCN. BioMed. Res. Int. 2014, 474905 (2014).
Feldmann, G., Beaty, R., Hruban, R. H. & Maitra, A. Molecular genetics of pancreatic intraepithelial neoplasia. J. Hepato Biliary Pancreat. Surg. 14, 224–232 (2007).
Grippo, P. J. & Sandgren, E. P. Acinar-to-ductal metaplasia accompanies c-myc-induced exocrine pancreatic cancer progression in transgenic rodents. Int. J. Cancer 131, 1243–1248 (2012).
Guerra, C. et al. Chronic pancreatitis is essential for induction of pancreatic ductal adenocarcinoma by K-Ras oncogenes in adult mice. Cancer Cell 11, 291–302 (2007).
Prevot, P. P. et al. Role of the ductal transcription factors HNF6 and Sox9 in pancreatic acinar-to-ductal metaplasia. Gut 61, 1723–1732 (2012).
Chuvin, N. et al. Acinar-to-ductal metaplasia induced by transforming growth factor beta facilitates KRAS(G12D)-driven pancreatic tumorigenesis. Cell. Mol. Gastroenterol. Hepatol. 4, 263–282 (2017).
Wong, C. H., Li, Y. J. & Chen, Y. C. Therapeutic potential of targeting acinar cell reprogramming in pancreatic cancer. World J. Gastroenterol. 22, 7046–7057 (2016).
Storz, P. Acinar cell plasticity and development of pancreatic ductal adenocarcinoma. Nat. Rev. Gastroenterol. Hepatol. 14, 296–304 (2017).
Jensen, J. N. et al. Recapitulation of elements of embryonic development in adult mouse pancreatic regeneration. Gastroenterology 128, 728–741 (2005).
Carriere, C., Young, A. L., Gunn, J. R., Longnecker, D. S. & Korc, M. Acute pancreatitis markedly accelerates pancreatic cancer progression in mice expressing oncogenic Kras. Biochem. Biophys. Res. Commun. 382, 561–565 (2009).
Maitra, A., Fukushima, N., Takaori, K. & Hruban, R. H. Precursors to invasive pancreatic cancer. Adv. Anat. Pathol. 12, 81–91 (2005).
Zhu, L., Shi, G., Schmidt, C. M., Hruban, R. H. & Konieczny, S. F. Acinar cells contribute to the molecular heterogeneity of pancreatic intraepithelial neoplasia. Am. J. Pathol. 171, 263–273 (2007).
Morris, J. P. T., Cano, D. A., Sekine, S., Wang, S. C. & Hebrok, M. Beta-catenin blocks Kras-dependent reprogramming of acini into pancreatic cancer precursor lesions in mice. J. Clin. Investig. 120, 508–520 (2010).
Hartupee, J. C., Zhang, H., Bonaldo, M. F., Soares, M. B. & Dieckgraefe, B. K. Isolation and characterization of a cDNA encoding a novel member of the human regenerating protein family: Reg IV. Biochim. Biophys. Acta 1518, 287–293 (2001).
Liu, X. et al. REG3A accelerates pancreatic cancer cell growth under IL-6-associated inflammatory condition: involvement of a REG3A-JAK2/STAT3 positive feedback loop. Cancer Lett. 362, 45–60 (2015).
Zhou, L. et al. Upregulation of REG Ialpha accelerates tumor progression in pancreatic cancer with diabetes. Int. J. Cancer 127, 1795–1803 (2010).
Lai, Y. et al. The antimicrobial protein REG3A regulates keratinocyte proliferation and differentiation after skin injury. Immunity 37, 74–84 (2012).
Takayama, R. et al. Serum tumor antigen REG4 as a diagnostic biomarker in pancreatic ductal adenocarcinoma. J. Gastroenterol. 45, 52–59 (2010).
He, X. J. et al. REG4 contributes to the invasiveness of pancreatic cancer by upregulating MMP-7 and MMP-9. Cancer Sci. 103, 2082–2091 (2012).
Li, Q. et al. Reg proteins promote acinar-to-ductal metaplasia and act as novel diagnostic and prognostic markers in pancreatic ductal adenocarcinoma. Oncotarget. 7, 77838–77853 (2016).
Lerch, M. M. & Gorelick, F. S. Models of acute and chronic pancreatitis. Gastroenterology 144, 1180–1193 (2013).
Xiong, X. et al. Pancreatic islet-specific overexpression of Reg3beta protein induced the expression of pro-islet genes and protected the mice against streptozotocin-induced diabetes mellitus. Am. J. Physiol. Endocrinol. Metab. 300, E669–E680 (2011).
Means, A. L. et al. Pancreatic epithelial plasticity mediated by acinar cell transdifferentiation and generation of nestin-positive intermediates. Development 132, 3767 (2005).
Halbrook, C. J. et al. Mitogen-activated protein kinase kinase activity maintains acinar-to-ductal metaplasia and is required for organ regeneration in Pancreatitis. Cell. Mol. Gastroenterol. Hepatol. 3, 99–118 (2017).
Liou, G.-Y. et al. Macrophage-secreted cytokines drive pancreatic acinar-to-ductal metaplasia through NF-κB and MMPs. J. Cell Biol. 202, 563–577 (2013).
Logsdon, C. D. & Ji, B. Ras activity in acinar cells links chronic pancreatitis and pancreatic cancer. Clin. Gastroenterol. Hepatol. 7, S40–S43 (2009).
Parikh, A., Stephan, A.-F. & Tzanakakis, E. S. Regenerating proteins and their expression, regulation and signaling. Biomol. Concepts 3, 57–70 (2012).
Kobayashi, S. et al. Identification of a receptor for reg (regenerating gene) protein, a pancreatic beta-cell regeneration factor. J. Biol. Chem. 275, 10723–10726 (2000).
Loncle, C. et al. IL17 functions through the novel REG3beta-JAK2-STAT3 inflammatory pathway to promote the transition from chronic pancreatitis to pancreatic cancer. Cancer Res. 75, 4852–4862 (2015).
Graf, R. et al. Coordinate regulation of secretory stress proteins (PSP/reg, PAP I, PAP II, and PAP III) in the rat exocrine pancreas during experimental acute pancreatitis. J. Surg. Res. 105, 136–144 (2002).
Lorchner, H. et al. Reg3beta is associated with cardiac inflammation and provides prognostic information in patients with acute coronary syndrome. Int. J. Cardiol. 258, 7–13 (2018).
Gironella, M. et al. Reg3beta deficiency impairs pancreatic tumor growth by skewing macrophage polarization. Cancer Res. 73, 5682–5694 (2013).
Folch-Puy, E. REG3β contributes to the immunosuppressive microenvironment of pancreatic cancer. Oncoimmunology 2, e26404 (2013).
Wang, J. et al. SOCS3 methylation in synergy with Reg3A overexpression promotes cell growth in pancreatic cancer. J. Mol. Med. 92, 1257–1269 (2014).
Nigri, J. et al. PAP/REG3A favors perineural invasion in pancreatic adenocarcinoma and serves as a prognostic marker. Cell. Mol. Life Sci. 74, 4231–4243 (2017).
Waters, A. M. & Der, C. J. KRAS: the critical driver and therapeutic target for pancreatic cancer. Cold Spring Harbor Perspect. Med. https://doi.org/10.1101/cshperspect.a031435 (2018).
von Figura, G., Morris, J. P. T., Wright, C. V. & Hebrok, M. Nr5a2 maintains acinar cell differentiation and constrains oncogenic Kras-mediated pancreatic neoplastic initiation. Gut 63, 656–664 (2014).
Corcoran, R. B. et al. STAT3 plays a critical role in KRAS-induced pancreatic tumorigenesis. Cancer Res. 71, 5020–5029 (2011).
Gruber, R. et al. YAP1 and TAZ control pancreatic cancer initiation in mice by direct up-regulation of JAK-STAT3 signaling. Gastroenterology 151, 526–539 (2016).
Liou, G. Y. et al. Mutant KRas-induced mitochondrial oxidative stress in acinar cells upregulates EGFR signaling to drive formation of pancreatic precancerous lesions. Cell Rep. 14, 2325–2336 (2016).
Criscimanna, A., Coudriet, G. M., Gittes, G. K., Piganelli, J. D. & Esni, F. Activated macrophages create lineage-specific microenvironments for pancreatic acinar- and beta-cell regeneration in mice. Gastroenterology 147, 1106.e1111–1118.e1111 (2014).
Kong, B. et al. Dynamic landscape of pancreatic carcinogenesis reveals early molecular networks of malignancy. Gut 67, 146–156 (2018).
Burotto, M., Chiou, V. L., Lee, J.-M. & Kohn, E. C. The MAPK pathway across different malignancies: a new perspective. Cancer 120, 3446–3456 (2014).
Dhillon, A. S., Hagan, S., Rath, O. & Kolch, W. MAP kinase signalling pathways in cancer. Oncogene 26, 3279–3290 (2007).
Shi, G. et al. Maintenance of acinar cell organization is critical to preventing Kras-induced acinar-ductal metaplasia. Oncogene 32, 1950–1958 (2013).
Collins, M. A., Yan, W., Sebolt-Leopold, J. S. & Pasca di Magliano, M. MAPK signaling is required for dedifferentiation of acinar cells and development of pancreatic intraepithelial neoplasia in mice. Gastroenterology 146, 822–834.e827 (2014).
Xu, Q. et al. Microarray-based gene expression profiling reveals genes and pathways involved in the oncogenic function of REG3A on pancreatic cancer cells. Gene 578, 263–273 (2016).
Ding, Y. et al. Reg3α overexpression protects pancreatic β cells from cytokine-induced damage and improves islet transplant outcome. Mol. Med. 20, 548–558 (2014).
Kadowaki, Y. et al. Reg protein is overexpressed in gastric cancer cells, where it activates a signal transduction pathway that converges on ERK1/2 to stimulate growth. FEBS Lett. 530, 59–64 (2002).
Liu, X. et al. REG3A accelerates pancreatic cancer cell growth under IL-6-associated inflammatory condition: Involvement of a REG3A–JAK2/STAT3 positive feedback loop. Cancer Lett. 362, 45–60 (2015).
Acquatella-Tran Van Ba, I. et al. Regenerating islet-derived 1α (Reg-1α) protein is new neuronal secreted factor that stimulates neurite outgrowth via exostosin Tumor-like 3 (EXTL3) receptor. J. Biol. Chem. 287, 4726–4739 (2012).
Wu, Y. et al. Hyperglycaemia inhibits REG3A expression to exacerbate TLR3-mediated skin inflammation in diabetes. Nat. Commun. 7, 13393 (2016).
Ding, L. et al. Glycogen synthase kinase-3beta ablation limits pancreatitis-induced acinar-to-ductal metaplasia. J. Pathol. 243, 65–77 (2017).
Acknowledgements
We thank Fazila Chouiali (Histopathology Platform, Research Institute of the McGill University Health Center), Shi Bo Feng and Min Fu (Molecular Imaging Platform, Research Institute of the McGill University Health Center) for their technical support. This work was supported by Canadian Institute of Health Research (J.L.L., Z.H.G.; PJ9-173593 and PJT-175208), and China Scholarship Council (HRZ; 201606280269).
Author information
Authors and Affiliations
Contributions
Conceptualization, Z.H.G., A.L.G.C., H.R.Z., and Q.L.; Methodology, H.R.Z., N.K., J.M.P., and A.L.G.C.; Software, H.R.Z., J.M.P., and A.L.G.C.; Validation, A.L.G.C., J.M.P., and H.R.Z.; Formal analysis, A.L.G.C., J.M.P., and H.R.Z.; Investigation, H.R.Z., J.M.P., N.K., and Q.L.; Resources, G.Z. and Y.W.; Original draft preparation, H.R.Z.; Review and editing, Z.H.G., A.L.G.C., J.M.P., and H.R.Z.; Supervision, Z.H.G., B.Y.S., J.L.L., and A.G.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Zhang, H., Corredor, A.L.G., Messina-Pacheco, J. et al. REG3A/REG3B promotes acinar to ductal metaplasia through binding to EXTL3 and activating the RAS-RAF-MEK-ERK signaling pathway. Commun Biol 4, 688 (2021). https://doi.org/10.1038/s42003-021-02193-z
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s42003-021-02193-z
- Springer Nature Limited
This article is cited by
-
Reg4 deficiency aggravates pancreatitis by increasing mitochondrial cell death and fibrosis
Cell Death & Disease (2024)
-
BRCA mutational status shapes the stromal microenvironment of pancreatic cancer linking clusterin expression in cancer associated fibroblasts with HSF1 signaling
Nature Communications (2022)
-
Spatially restricted drivers and transitional cell populations cooperate with the microenvironment in untreated and chemo-resistant pancreatic cancer
Nature Genetics (2022)