β cell transdifferentiation does not contribute to preneoplastic/metaplastic ductal lesions of the pancreas by genetic lineage tracing in vivo

Abstract

Inflammatory injury to the pancreas results in regeneration of normal tissue and formation of metaplastic lesions of a ductal phenotype. These metaplastic ductal lesions (MDL) are called tubular complexes (TC), mucinous metaplasia, or pancreatic intraepithelial neoplasia. Because they are regularly found in chronic pancreatitis and pancreatic cancer, their formation is thought to represent a step in inflammation-mediated carcinogenesis. Despite these lesions' ductal character, their origin is controversial. All known pancreatic cell lineages have been suggested as the origin. In vitro studies suggest that differentiated cells in the pancreas remain highly plastic and can transdifferentiate as a mechanism of regeneration and metaplasia. In vivo studies suggest that islets, specifically β cells, may be the cell of origin. However, in vitro studies are subject to ductal cell contamination, and previous in vivo studies interpret static data rather than direct evidence. Using genetic lineage tracing in vivo, we investigate whether transdifferentiation of β cells contributes to regeneration or metaplasia in pancreatitis. RIP-CreER;Z/AP mice were used to heritably tag β cells in the adult pancreas. Injury by cerulein pancreatitis resulted in regeneration of normal tissue and metaplasia with formation of two distinct types of TC and mucinous lesions. Lineage tracing revealed that none of these MDL are of β cell origin; nor do β cells contribute to regeneration of normal acinar and ductal tissue, which indicates that the plasticity of differentiated pancreatic islet cells, suggested by earlier static and in vitro studies, plays no role in regeneration, metaplasia, and carcinogenesis in vivo.

Tissue injury can result in regeneration of normal tissue and in metaplasia. Metaplasia is often seen in response to chronic injury and can occur by selective cellular proliferation or death or by transdifferentiation. Cell lineages in the pancreas include exocrine (acinar and ductal) and endocrine cells. Inflammation of the pancreas results in the formation of metaplastic lesions of a ductal phenotype (MDL) that are consistently found in chronic pancreatitis (CP) and specimens of pancreatic cancer. Therefore, this “ductal” metaplasia is thought to represent a condition with increased risk of neoplasia. In the literature, two definitions have been used to describe MDL in the pancreas: tubular complexes (TC) and mucinous metaplasia or pancreatic intraepithelial neoplasia (PanIN) (1, 2). TC, defined as cylindrical tubes with a wide lumen lined by a monolayer of flat duct-like cells (3–5), are found in pancreatic development, regeneration, in CP, and in cancer (5–8). In some studies, TC exhibited a high cell turnover and were thought to have regenerative potential, including islet regeneration (5, 9). A progression of TC to neoplasia has been suggested in rodent models of chemical carcinogenesis (1, 10, 11) and transgenic mouse models (12). Misexpression of mucins in pancreatic cancer has shifted the focus to mucinous metaplastic lesions (MML), which are classified in the PanIN system according to their grade of atypia and risk of neoplasia (2, 13).

Despite the ductal character of TC and MML, their cellular origin remains controversial; ductal (14), acinar (12, 15), islet (16) cells, and all lineages (1) have been proposed as the cell of origin of MDL. Transdifferentiation of islet cells, specifically β cells, is proposed as a likely source of metaplasia/neoplasia in several in vitro and in vivo studies (17). In vitro, isolated pancreatic islets have been reported to transform to duct-like structures expressing ductal markers and to acquire genetic alterations typical of pancreatic cancer, such as K-ras mutations and p16 deletion (18–20). However, in vitro studies do not necessarily mirror in vivo events. Furthermore, in vitro studies are subject to cell contamination: the observation of ductal cells in cultured isolated islets/islet cells could be a result of proliferation of ductal cells contaminating the preparation. The genetic alterations observed in vitro may consequently be artifacts of long-term culture. In vivo, the location of MDL within histologic preparations of islets has been used to support their islet cell origin. Intrainsular metaplastic ductal lesions have been described in specimens of pancreatic cancer (21), in genetically engineered mice that overexpress EGF-like growth factors in β cells (22), and in a hamster model of chemical carcinogenesis (17). Clearly, the colocalization of MDL within islets observed by snapshot analysis of morphology or the momentary expression of lineage markers can be suggestive, but remain purely descriptive and do not justify conclusions about dynamic processes, including transdifferentiation and the cellular origin of ductal lesions. Furthermore, functional methodologies, in which selective chemical or genetic depletion of β cells inhibited, while stimulation of β cell neogenesis promoted, the development of metaplasia and cancer, led investigators (17, 23–26) to infer an “unequivocal role of β cells in transdifferentiation and, thus the carcinogenesis process” (17). Although these functional studies show that the presence of β cells may modify the formation of ductal lesions, they do not allow conclusions about their role as the cell of origin.

This question can be answered only by a method that allows the fate of a specific cell type and its progeny to be traced in adult tissue. With advances in transgenic technology, it has become possible to genetically label specific cell lineages and to trace their fate in vitro and in vivo. Such genetic lineage tracing is not subject to the above-mentioned weaknesses of previous studies and has recently been used to show that β cells are formed by self-duplication in the healthy adult pancreas (27). To date, there are no lineage tracing studies on the cellular origin of metaplastic lesions or regeneration of the injured pancreas.

Here, we use genetic lineage tracing in a mouse model of pancreatitis to investigate whether transdifferentiation between β cells and exocrine lineages contributes to regeneration or to the formation of MDL. RIP-CreER mice (27), in which β cells are labeled with a tamoxifen-inducible Cre/lox system, were crossed to the reporter strain Z/AP (28), in which Cre-mediated recombination results in constitutive and heritable expression of human placental alkaline phosphatase (HPAP). In bigenic RIP-CreER;Z/AP mice, tamoxifen exposure labels adult β cells and their progeny with HPAP expression. We induced acute and chronic pancreatitis in tamoxifen-injected RIP-CreER;Z/AP mice and found that the resulting MDL were not HPAP-tagged and thus not of β cell origin. Moreover, HPAP expression remained specific for β cells, and the fraction of tagged β cells remained stable in pancreatitis, indicating that transdifferentiation between β cells and exocrine lineages does not contribute to regeneration after inflammation.

Results

Tubular Formations Are a Heterogeneous Group of Lesions.

Different authors use the terms “tubular formations” and “tubular complexes” in different models of both acute and chronic pancreatitis or pancreatic cancer to designate a heterogeneous group of similar but distinct lesions.

We observed three distinct types of lesions, two of which are consistent with the definition of TC as cylindrical tubes with a wide lumen lined by a monolayer of flat, duct-like cells (3–5). The third type is consistent with mucinous metaplasia. To analyze these lesions separately, we characterized them as follows.

1. TC lined by a few large, flat cells. Lesions with a wide, empty lumen lined by a few (typically 4–5) flat wide cells (Fig. 1A) whose sparse cytoplasm exhibits staining characteristics similar to acinar cells. “Intermediate” figures can be observed between acini with slightly widened lumen and such TC, leading to the suggestion that TC are derived from acinar cells through acinar atrophy or acinar-to-ductal metaplasia (4, 8, 29).

2. TC lined by many small cells. Lesions with an empty lumen lined by a variable number of flat small cells (Fig. 1B). Because the cells are small, larger diameters result in higher cell numbers in the epithelial lining. These “duct-like” TC are seen more often with increasing acinar loss, often show a complex arrangement with branching, and can form fields composed almost entirely of TC (Fig. 1B). Similar observations may have led to the suggestion that these lesions represent proliferating ductules. Frequent expression of endocrine markers in fields of TC may suggest either that TC are derived from islets or that they may represent a pathway for islet regeneration (9).

3. MML. The epithelial cells of MML are not flat, but vary in height according to the extent of mucin expression. Their lumen is variable in diameter and can contain secreted mucins. Some lesions exhibit abundant supranuclear mucin and flat, basally located nuclei oriented perpendicularly to the basement membrane (Fig. 1C), features characteristic of early PanIN (2, 13). Their mucin is often periodic acid/Schiff's reagent (PAS)- and Alcian blue-positive. Although these lesions do not fulfill the criteria of flat cells, some authors name them TC; others use the terms metaplastic ducts, metaplastic ductal lesions, or PanIN (12, 22, 30, 31).

Fig. 1.

Pancreatitis results in formation of several types of metaplastic lesions. (A) TC with few large cells. (B) TC with many small cells. (C) MML composed of cells with abundant supranuclear mucin (arrowheads), which is frequently Alcian blue⁺ (turquoise). (Scale bar, 100 μm.)

Distinguishing among these lesions may enable us to better evaluate their cells of origin.

MDL Are Not of β Cell Origin.

Because β cells were heritably tagged before induction of pancreatitis, lesions derived from tagged β cells should express HPAP even if they have undergone complete transdifferentiation to another cell type, such as ductal cells. However, not all β cells are labeled, and the likelihood of capturing the event of a β cell-derived lesion is a function of the efficiency of the tagging system. The applied system reproducibly resulted in HPAP expression in 55% of β cells in controls. Tagging has no adverse effects on the pancreas, including β cell function (27). Assuming that it also has no effect on formation of MDL, a tagging efficiency of 50% results in a likelihood of at least 50% that a β cell-derived lesion is HPAP⁺. The fraction of tagged β cells for each animal was used to calculate the expected number of HPAP⁺ lesions. This hypothetical number was then compared with the number of tagged lesions actually observed.

Comparison of expected vs. observed tagged events suggests that none of the observed types of MDL are of β cell origin. A total of 245 TC with a few large cells were found (Fig. 2 A–C). If “large-cell TC” were derived from β cells, ≈125.5 HPAP⁺ events would be expected. In fact, only two (0.8%) large-cell TC were HPAP⁺. Similarly, 2,231 TC with small cells were identified (Fig. 2 D–F). Hypothesizing β cell origin, one would expect ≈1,115.5 to be HPAP⁺. However, only four (0.2%) “small-cell TC” were HPAP⁺; the vast majority again were not labeled (Fig. 2 D and E). Independent of the duration of injury (Fig. 2 C and F), neither type of TC was derived from β cells.

Fig. 2.

Analysis of expected vs. observed tagged events indicates that TC with a few large cells (A–C) and TC with many small cells (D–F) are not of β cell origin. Open bars represent total number of observed lesions. Gray bars indicate expected hypothesized number of HPAP⁺ events. Black bars represent the number of observed HPAP⁺ lesions. (A and D) For both types of TC, the number of observed events is far lower than the number of expected events. The likelihood that this distribution occurred by chance is extremely low (∗, binomial P < 0.000001). (B and E) HPAP staining shows strong β cell tagging (blue) within islets and complete absence of tagging in TC (arrowheads) as well as acinar cells, stromal cells, and ducts (arrows). (C and F) These results are independent of the duration of pancreatitis. AP, acute pancreatitis. 3wCP and 6wCP, 3-wk and 6-wk chronic pancreatitis, respectively. (Scale bar, 100 μm.)

MML were identified first by PAS staining (Fig. 3 A and B). Because PAS⁺ mucins can be expressed at a low level in the normal pancreas, whereas Alcian blue⁺ mucins are found only in metaplastic lesions, we confirmed metaplastic transformation by serial Alcian blue stains (Fig. 3 C and D). In HPAP/PAS double stains, 397 MML were identified. Given the tagging efficiency, one would expect ≈208.4 MML with HPAP expression. However, only two (0.5%) of these lesions were HPAP⁺ (Fig. 3 A and B). Of all PAS⁺ lesions, 223 lesions were identified in serial sections and were Alcian blue⁺ (Fig. 3 C and D). Of these, 117.1 would be expected to be HPAP⁺ if they were of β cell origin. However, only the same two tagged events were found in serial sections, indicating that MML are not of β cell origin.

Fig. 3.

Analysis of expected vs. observed tagged events indicates MML are not of β cell origin. (A and B) Lesions observed in HPAP/PAS double stains (magenta, mucin). (C and D) Lesions additionally identified in serial Alcian blue stains (turquoise, mucin). (A and C) Colored bars show the number of lesions observed. Gray bars show the expected/hypothesized number of HPAP⁺ events. Black bars represent the number of observed HPAP⁺ lesions. The number of observed events is far lower than the number of expected events. The likelihood that this distribution occurred by chance is extremely low (∗, binomial P < 0.000001). (B) HPAP/PAS stains show strong β cell tagging (blue) within islets and complete absence of labeling in MML. (C and D) The majority of MML observed in A and B also stain with Alcian blue, supporting metaplasia. (Scale bar, 100 μm.)

For each lesion, binomial testing was used to determine the likelihood that the observed distribution of events could occur by chance if the lesions were derived from β cells. Given the numbers for expected and observed tagged events, the likelihood that this distribution was observed by chance, and that the observed lesions were derived from the nontagged pool of β cells, is extremely low (binomial P < 0.000001) for all three types of lesions.

To exclude the possibility of false-negative results from epigenetic inactivation of the Z/AP reporter during inflammation and metaplastic transformation, specimens of additional animals with CP for 6 wk (6wCP) were stained for LacZ activity. HPAP⁻ cells with an intact reporter are expected to be LacZ⁺ (28). Both TC and MML were indeed LacZ⁺, demonstrating a functional reporter in these lesions (Fig. 4 A and B).

Fig. 4.

Control experiments. (A and B) Functional reporter system. MDL (arrowheads) are LacZ⁺ in X-gal stains; β cells, where Cre activity removed LacZ, are LacZ⁻ (A). (C–H) Observed tagged events may result from inflammation-mediated transgene activation. Comparison of HPAP (C) and HPAP/insulin double stains (D) reveals single events of tagged non-β cells within the exocrine pancreas. (E–G) Examples of the very rare occurrence of HPAP⁺ MDL (arrowheads). (E) TC with many small cells, (F) TC with few large cells, and (G) MML. Rare occurrences of HPAP⁺ acinar cells are also identified (arrow in F). (H) HPAP expression in an acinus (arrow and insert) and absent HPAP expression in an islet (*) in non-tamoxifen-injected animals with CP identify inflammation-mediated HPAP expression as a possible mechanism for false-positive results.

In single animals with CP, HPAP expression was identified at a very low frequency (0.0–0.4%) in morphologically normal acinar cells (Fig. 4 C and D) and in a small number of metaplastic lesions (Figs. 2 and 3 and Fig. 4 E–G). HPAP/insulin double staining revealed that these events did not reflect isolated β cells, but HPAP expression in non-insulin-producing cells (Fig. 4 C and D).

We cannot exclude the possibility that these findings represent extremely rare events of true β cell transdifferentiation. However, one other hypothesis is that they represent false-positive results because of aberrant alkaline phosphatase (ALP) activity in non-β cells. To test this possibility, we performed several control experiments. First, to determine whether these tagged cells could result from aberrant expression of endogenous ALP in response to inflammation, nontransgenic littermates were subjected to chronic inflammation. No ALP activity was identified in nontransgenic littermates (data not shown), excluding expression of endogenous ALP as a source of false-positive results. Second, to determine whether inherent leakiness of the transgenic system could result in false positives, we analyzed bigenic mice injected with tamoxifen but not subjected to pancreatitis. Absent HPAP activity in non-β cells in these healthy controls excluded false-positive HPAP expression due to leakiness (data not shown). Third, to assess whether inflammation may result in aberrant HPAP transgene activation, we analyzed bigenic mice with CP that were not tamoxifen-injected. In these animals, aberrant HPAP expression was infrequently observed in acinar cells (1–2 acini per cut) (Fig. 4H). In tamoxifen-injected animals with CP the events of HPAP⁺ TC and mucinous lesions had a low frequency (0.2–0.8%), similar to HPAP expression in acini (0.0–0.4%). Together, these data suggest that tagged lesions may indeed represent false-positive observations rather than very infrequent events of true β cell transdifferentiation.

Thus, based on direct lineage tracing in vivo, β cell transdifferentiation does not contribute to the formation of MDL observed in chemically induced pancreatitis.

Transdifferentiation Between β Cells and Exocrine Cells Does Not Contribute to Regeneration.

In previous in vitro studies, transdifferentiation of β cells has been discussed not only as the origin of MDL but also as a source of regeneration of normal exocrine and endocrine tissue (9, 17, 32–34). However, the role of transdifferentiation in vivo remains unclear. Thus, we used the same lineage-tagging strategy to determine whether transdifferentiation of β cells was important in the in vivo regenerative response after chemical pancreatitis.

Acute cerulein pancreatitis induces severe injury with loss of >10% of acinar cells and almost complete regeneration within 96 h (35). In the chronic model, repetitive injury over a period of 3–6 wk should result in replacement of the majority of exocrine tissue. This hypothesis is supported by the extensive increase in Ki-67- and BrdU-positive nuclei observed in acinar, centroacinar and ductal cells in CP vs. controls [supporting information (SI) Fig. 6 A and B]. In controls, islet cells exhibited a relatively low cell turnover, with Ki-67 expression or BrdU incorporation (72 h exposure) in ≈2% of cells (SI Fig. 6A). Surprisingly, this low proliferation rate in islets is not enhanced in CP, in contrast to the markedly increased cell turnover observed in the exocrine pancreas (SI Fig. 6B).

The low rate of cell turnover in islets and the high rate in the exocrine pancreas alone suggest that endocrine-to-exocrine transdifferentiation cannot play a major role in exocrine regeneration.

As direct evidence by lineage tracing, β cell transdifferentiation to exocrine cells occurring during CP would be detected as HPAP⁺ acinar or ductal cells. HPAP was never observed in normal pancreatic ducts or in intra-or juxta-insular ducts (SI Fig. 7), indicating that β cells did not contribute to ductal regeneration. As explained in detail above (Fig. 2), HPAP expression was observed only very rarely (0.0–0.4%) in acinar cells in CP, most likely due to inflammation-mediated HPAP expression (Fig. 4 C–H). Even if these events represent true β cell transdifferentiation, this mechanism would be insignificant and insufficient to account for exocrine regeneration.

Conversely, transdifferentiation of acinar and ductal cells or TC has been proposed as a mechanism of β cell renewal in several studies (9, 32, 33, 36, 37). A recent lineage tracing experiment reveals that β cells can be generated from acinar cells in vitro (38). In contrast, it has been shown by lineage tracing in vivo that in the normal adult pancreas and after pancreatectomy, preexisting β cells rather than other cells are the source of new β cells (27). Thus, we wanted to determine whether transdifferentiation may play a role in β cell renewal in the setting of inflammatory injury. New β cells generated during the course of pancreatitis are labeled only if they are progeny of labeled β cells but not if they are derived from a nonlabeled source, such as acinar or ductal cells or TC. Replenishment of preexisting islets with new β cells from a nonlabeled source would yield a decreased percentage of tagged β cells in islets. Maintenance of the β cell population by self-duplication would result in stable tagging.

Both frequency and distribution of labeled β cells remained stable during pancreatitis (Fig. 5). Tagging remained stable even in small clusters of endocrine cells, which are thought by some to represent newly formed endocrine tissue caught during coalescence (Fig. 5A) (39–41). Thus, islet neogenesis and replenishment of preexisting islets by transdifferentiation of an unlabeled source, including acinar and ductal cells as well as TC, was not observed (SI Discussion).

Fig. 5.

Transdifferentiation between exocrine cells and β cells does not contribute to regeneration. (A) Stable percentages of islets (black bars) and endocrine clusters (gray bars) with tagged β cells in controls and pancreatitis. (B and C) HPAP staining (blue) reveals that tagging remains specific for β cells and stable in response to inflammation. Ducts and acinar tissue remain HPAP⁻. (D) Tagging of β cells remains stable in pancreatitis. (E and F) HPAP (blue)/insulin (brown) double staining demonstrates that tagging remains specific for β cells. (Scale bars, 100 μm.)

These results indicate that transdifferentiation of β cells does not play a role in exocrine regeneration in pancreatitis and that β cell renewal in the inflamed pancreas appears to occur by self-duplication rather than by transdifferentiation.

Discussion

The pancreas is composed of a variety of cell types. Epithelial cell lineages include acinar and ductal cells as well as endocrine cells, the majority of which are β cells. Injury to the adult pancreas results in regeneration of normal epithelium and in a metaplasia of ductal phenotype. Results from various in vitro studies and in vivo models of pancreatic injury suggest that transdifferentiation among the different cell lineages may play a role in both regeneration and metaplasia. Transdifferentiation of β cells has been discussed as a mechanism of exocrine regeneration, metaplasia and cancer formation, and conversely, transdifferentiation of exocrine cells has been implicated in β cell regeneration. Recently, it has been shown by lineage tracing that in the normal adult pancreas and after pancreatectomy, new β cells are generated by self-duplication of preexisting β cells (27). However, all studies on the role of β cells in metaplasia published to date are based on interpretation of the static expression of lineage markers and lack direct evidence by lineage tracing. Our first objective, therefore, was to investigate by lineage tracing in the adult pancreas in vivo whether β cells are the origin of MDL or of normal exocrine regeneration. Our second objective was to elucidate whether in the setting of pancreatitis β cells themselves may be regenerated by transdifferentiation or, as demonstrated for the normal pancreas, by self-duplication. In human disease, the observation of the same MDL and K-ras mutations in chronic pancreatitis and in cancer has led to the assumption that ductal metaplasia occurs in inflammation and is an early step in cancer formation. The cerulein model of pancreatitis was chosen because it histologically mirrors the MDL found in human disease and results in the different MDL with sufficient frequency to draw statistically solid conclusions from lineage tracing results. The cerulein model has been frequently used to study the extent and mechanisms of regeneration after pancreatitis (7, 35, 42, 43), including a recent study to reveal the recapitulation of developmental programs in the adult regenerative response (44). Using Cre-loxP-based genetic tracing, we show that β cells contribute neither to exocrine regeneration nor to formation of metaplasia in pancreatitis. Further, we show that severe exocrine injury does not result in enhanced endocrine cell turnover, and we find no evidence of β cell renewal through transdifferentiation of other lineages in pancreatitis.

The cerulein model induces a variety of MDL that can be divided into two major groups: TC and MML. Whereas TC were originally defined as cylindrical tubes with an often dilated lumen lined by a monolayer of flattened duct-like cells (3, 4), for lack of a stringent classification comprising all metaplastic lesions, the term has also been used for other lesions, including mucinous ones. This lack of classification may explain the controversial results regarding their origin and significance. In contrast, mucinous metaplasia has been used specifically for mucinous lesions that are more stringently defined in the PanIN classification for the human pancreas (2) and in mouse models of pancreatic disease (13). However, this nomenclature does not encompass all early metaplastic changes seen in inflammation. To do justice to these lesions and their possible origin, we distinguish two distinct types of TC and MML.

All these lesions are distributed throughout the exocrine pancreas, as typically observed in human disease. Intra- and juxta-insular ducts were never identified as tagged and are not a major site of metaplasia. Previous studies reporting formation of metaplastic lesions predominantly in or beside islets used a potent carcinogen or transgenic misexpression of growth factors targeted to islets to induce these lesions (17, 22). In this study, metaplastic lesions formed in the setting of CP through repetitive inflammatory injury may more closely reflect the early events of human disease. Direct lineage tracing excludes β cells as a significant source of metaplastic lesions formed during pancreatitis. The possibility that lesions were not tagged because by chance they were all derived from nontagged β cells was excluded by binomial testing based on the assumption that a given lesion originated from a single cell. Although such clonal expansion is a common mechanism later in cancer formation, it is unlikely to play a role in the formation of early metaplastic lesions. However, if a lesion was formed by nonclonal expansion through transdifferentiation of several β cells, then the likelihood of detecting HPAP⁺ cells within these lesions would be even higher.

Whereas the significance of TC is controversial, mucinous metaplasia is thought to represent an early event in the formation of cancer from normal pancreatic epithelium (2). The MML observed in the current model express both PAS- and Alcian blue-positive mucins, mirroring the metaplastic and neoplastic mucin expression seen in human and mouse PanIN. However, progression to high-grade atypia or invasive cancer was not observed. Thus, our study shows that β cells are not the cellular origin of lesions thought to be early cancer precursors but does not demonstrate this for pancreatic cancer per se.

Pour and colleagues (23–25, 45) have shown that metaplasia/neoplasia in the pancreas appear to depend on the presence of β cells. Although our results show β cells do not directly contribute as cells of origin, they may contribute to metaplasia and carcinogenesis by indirect mechanisms such as altered signaling. Further, the present study does not formally exclude the possibility that transdifferentiation of an islet cell lineage other than β cells may contribute to metaplasia. However, the observed low proliferation in islets and the distribution of MML away from islets argue against nontagged endocrine cells as progenitors.

The second question addressed here is the role of transdifferentiation in regeneration of normal tissue. The cerulein model results in considerable exocrine damage and regeneration (35). Transdifferentiation of β cells, which would result in labeling of other lineages, was not seen in normal ductal cells and only very infrequently in acinar cells. Although we cannot exclude very infrequent events of true β cell transdifferentiation, our control experiments suggest that this rare expression of HPAP is more likely the result of inflammation-mediated transgene activation. Thus, our results reveal that transdifferentiation of β cells does not contribute to exocrine regeneration in the cerulein model.

Furthermore, the frequency and distribution of labeled β cells remained stable in the pancreatitis groups compared with baseline controls. Islet neogenesis by transdifferentiation of an unlabeled source, which would include acinar and ductal cells as well as TC, was not observed. Surprisingly, we did not see a significant increase in endocrine proliferative activity in response to pancreatitis. The frequency of proliferating cells observed in pancreatitis is not significantly different from that seen in controls; both are consistent with the low (2–3%) β cell turnover in rats (46) and a low (4–5%) BrdU incorporation of β cells in mice found at 1 wk (27). Although we did not see increased islet cell turnover in pancreatitis, extrapolation of this baseline turnover suggests that enough new β cells were generated during the chase period of up to 6 wk to detect transdifferentiation from unlabeled sources as the predominant source of new β cells. The stable percentages of tagged endocrine clusters and islets provide evidence against islet neogenesis from exocrine lineages and suggest that in the inflamed pancreas β cells are renewed by self-duplication, as in the normal pancreas (27). However, the model may not be sensitive enough to exclude a low contribution of acinar-to-β cell transdifferentiation (38) (SI Discussion).

In summary, we observed three different types of MDL, including mucinous metaplasia. In vivo lineage tracing with stringent control experiments revealed that transdifferentiation of β cells does not play a role in the formation of these metaplastic lesions in pancreatitis. Moreover, transdifferentiation of β cells does not contribute to exocrine regeneration, and we found no evidence of β cell renewal from exocrine lineages. These results suggest that the plasticity of differentiated cells, which has been suggested based on in vitro studies, does not necessarily play a role in pancreatic regeneration, metaplasia and carcinogenesis in vivo. To our knowledge, no other studies have used genetic lineage tracing to investigate transdifferentiation and the formation of metaplastic lesions in adult tissue in vivo. For the pancreas, similar studies need to be performed for acinar and ductal lineages to further elucidate the process of metaplasia and hopefully one day define the cell of origin of pancreatic cancer.

Materials and Methods

In Vivo Lineage Tracing.

RIP-CreER;Z/AP mice were genotyped by PCR as described in ref. 30. Tamoxifen (Sigma–Aldrich, St. Louis, MO) was injected subcutaneously every second day in five doses of 4 mg as described in ref. 27. Pancreatitis was induced 2.5 wk after the last tamoxifen injection.

Experimental Pancreatitis.

All experiments were approved by the Massachusetts General Hospital Subcommittee on Research Animal Care. Pancreatitis was induced by the cerulein model (35, 47, 48). Acute pancreatitis (AP) was induced by 6 hourly i.p. injections of 50 μg/kg cerulein (Sigma–Aldrich) after a fasting period of 12 h (AP, n = 6). CP was induced by three series of injections per week for 3 (3wCP, n = 4) or 6 wk (6wCP, n = 7). Bigenic animals without pancreatitis (n = 7) served as healthy controls. Further controls were bigenic animals with 6wCP but without tamoxifen exposure. Pancreata were harvested 48 (AP) or 72 (CP) h after the last injection. Specimens were fixed in 4% paraformaldehyde at 4°C for 5 h, dehydrated, and embedded in paraffin.

Histology and Immunohistochemistry.

Expression of HPAP and LacZ was identified as described (27, 28). Mucin was identified by PAS and Alcian blue (pH 2.8) staining, using standard protocols. Expression of insulin, HPAP, Ki-67, and BrdU incorporation were identified by immunohistochemistry. Antigen retrieval was performed at pH 8 for HPAP and at pH 6 for insulin, Ki-67 and BrdU (BioGenex, San Ramon, CA). Primary antibodies used were: insulin (1:100; Santa Cruz Biochemicals, Santa Cruz, CA), HPAP (1:400; Santa Cruz Biochemicals), Ki-67 (1:25; Dako, Carpinteria, CA), and BrdU (1:100; Abcam, Cambridge, MA). Biotinylated secondary antibodies were applied at a dilution of 1:1,000. Expression was visualized with diaminobenzidine (Zymed Laboratories, South San Francisco, CA). Slides were counterstained with hematoxylin or nuclear fast red.

Quantitative Analysis.

Percentage of tagged cells.

For each animal, the fraction of tagged islets and small clusters of <15 β cells was determined by analyzing the entirety of doubly or serially HPAP/insulin-stained sections through the entire pancreas. Tagging of β cells was quantified by using serial HPAP and insulin stains by counting individual cells in three randomly selected islets per animal.

Morphologic analysis: MDL.

In the cerulein model, TC are distributed throughout the pancreas. Analysis (quantification and tagging) of TC was performed by counting events in three 100× high-power field per cut in three cuts per animal. MML were observed with much lower frequency, only in CP, and were analyzed in the entirety of three HPAP/PAS double stains and serial Alcian blue stains per animal.

Statistical analysis.

Graphs represent absolute numbers (for lesions) or the mean ± SEM (for tagging). The likelihood that the differences between expected and actually observed numbers of tagged lesions were due to chance was evaluated by binomial testing.

Abbreviations

CP

chronic pancreatitis

HPAP

human placental alkaline phosphatase

MDL

metaplastic ductal lesions

MML

mucinous metaplastic lesions

PanIN

pancreatic intraepithelial neoplasia

PAS

periodic acid/Schiff reagent

TC

tubular complexes.