Experimental Evidence Supporting the Lack of Primary Stem Cells in Adult Pancreatic Tissue

Experimental Evidence Supporting  the Lack of Primary Stem Cells in Adult  Pancreatic Tissue

JiaQing Gong 1 FuZhou Tian JianDong Ren GuoDe Luo   Department of General Surgery, The People’s Liberation Army General Hospital of Chengdu Command,  Chengdu , China

Key Words ：Pancreatic stem cells  Nestin  5-Bromo-2  -deoxyuracil  nucleotide  Pancreatic duodenal homeobox-1  

Abstract 

 Purpose: To investigate the origin and localization of pancreatic stem cells in adult pancreatic tissues and to determine the primary mechanism underlying the participation  of these cells in repairing pancreatic injuries

Methods: Sprague-Dawley rats were divided into experimental and  control groups. The experimental group was given intraperitoneal injections of cerulein to induce acute pancreatitis. At  6 h, 1, 2, 3, 5 and 7 days, 5 rats from the experimental group  and 2 rats from the control group were sacrificed; all sacrificed animals were intraperitoneally injected with 5-bromo- 2  -deoxyuracil nucleotides (BrdU) 6 and 3 h prior to sacrifice.  The pathological changes of pancreatic tissue were observed. The stem cell marker nestin and the cell proliferation  marker BrdU were detected with immunohistochemistry.  Pancreatic duodenal homeobox-1 (PDX-1) was determined  by real-time PCR. Results: (1) The pathological changes of  acute pancreatitis can be divided into three phases: the edema and apoptosis phase, the hemorrhagic necrosis phase,  and the reconstruction phase. (2) Nestin-positive cells main-

ly appeared in the interlobular vascular lumen after cerulein  injection, and they peaked at day 3 when the positive cells  spread all over the pancreatic tissues. (3) BrdU-positive cells  began to appear in the area surrounding the interlobular region, and the number of positive cells peaked on day 7. (4)  The expression of PDX-1 mRNA initially increased, then decreased and gradually got close to a normal level. Conclusion: Primary pancreatic stem cells may not exist in the adult  pancreatic tissues. The so-called pancreatic stem cells may  actually originate from bone marrow stem cells. When pancreatic tissue is injured, bone marrow stem cells may participate in the repair.

Copyright © 2010 S. Karger AG, Basel and IAP

Introduction

Stem cell research has become a very attractive field  in recent years. The presence of stem cells has been confirmed in multiple organs [1–3] . Many investigators  have reported the existence of stem cells in pancreatic tissue [4, 5] , which could shed light on the treatment of  severe acute pancreatitis (SAP), one of the most difficult  medical problems worldwide. SAP is generally believed  to be an excessive systemic inflammatory response syndrome and leads to distant organ damage and multiple  organ dysfunction syndrome, even mortality in this  condition [6, 7] . How can deterioration of SAP be prevented? If the stem cell plays a leading role for the tissue  injury repair in the early period of SAP, the problem may  be solved. However, are there primary stem cells in adult  pancreatic tissue? Dor et al. [8] found that the nascent  cells in mice from a partial pancreatic resection originated from the self-replication of preexisting  cells in  pancreatic islets, rather than from pancreatic stem cells.  In 2005, Levine and Mercola [9] published a paper in the  New England Journal of Medicine to support the findings of Dor et al. [8] . Meanwhile, in recent years, some  researchers have found that bone marrow stem cells can  differentiate into pancreatic cells under certain conditions [10, 11] . In a summary of the literature, we found  that the major differences in pancreatic stem cell research are (1) whether pancreatic stem cells exist in adult  pancreatic tissue, (2) the source and location of the pancreatic stem cells, and (3) the principal mechanisms for  involvement of pancreatic stem cells in the repair of  pancreatic tissues.

An important and commonly used marker for identification of pancreatic stem cells is nestin. It is still controversial whether this protein is the most specific marker.  Previous research showed that nestin-positive cells have  a strong proliferation and regeneration capacity. Nestin  was first widely recognized as a specific marker for neural  stem cells [12] . It is highly expressed in neural epithelial  cells during embryonic development, but not in the adult  central and peripheral nervous system [13, 14] . In addition, the expression of nestin has also been shown in rat  bone marrow [15] and human embryonic stem cells [16,  17] . Recently, investigators found that nestin protein is  also expressed in human pancreatic islet cells and embryonic pancreatic tissues [18, 19] . Therefore, many researchers believe that pancreatic nestin-positive cells are also  pancreatic stem cells [20, 21] . Pancreatic duodenal homeobox-1 (PDX-1) is a transcription factor in islet  cells,   cells and the endocrine cells scattered in the duodenum.  PDX-1 activation can promote the expression of insulin,  somatostatin and other important genes in  cells. So  PDX-1 is essential for the formation of endocrine and  exocrine pancreatic glands during embryonic development [22, 23] . Therefore, PDX-1 is also used as an important marker in pancreatic stem cell research [24, 25] . In this study, we established a rat model of acute pancreatitis, located pancreatic stem cells using nestin as a marker,  and monitored cell proliferation as well as regeneration  during pancreatic repair with 5-bromo-2-deoxyuridine  (BrdU) labeling [26] . At same time, we determined the  expression level of PDX-1, an important regulator in stem  cells, to determine the origin and localization of pancreatic stem cells as well as to develop a better understanding  of the mechanism underlying stem cells’ participation in  pancreatic tissue repair.

Materials and Methods

Materials

Anti-nestin polyclonal antibody, BrdU labeling reagent and  rabbit anti-rat BrdU monoclonal antibody were purchased from  Sigma. A ready-to-use immunohistochemistry kit (SABC method) was purchased from Wuhan Boster Biological Technology.  The DAB reagent came from Beijing Zhongshan Biotechnology.  Other reagents included TRIzol (Invitrogen), Ribolock inhibitor  (MBI), agarose (Bio-Rad), RevertAid TM M-MuLV Reverse Transcriptase (MBI), SYBR Premix Ex Taq II (Takara), and PCR Mixture (SinoBio). Equipment used in this study included a centrifuge  with refrigeration (Thermo Scientific), a UV spectrophotometer  (Amersham), a Gene cycler (Bio-Rad), a nucleic acid electrophoresis system (Beijing Baygene), and an imaging analysis system  (UVP). 

Animal Groups and Animal Model Preparation

Forty-two 8-week-old Sprague-Dawley (SD) rats weighing  120 8 20 g were provided by the Animal Experimental Center,  HuaXi Medical University. The rats were raised at 18–28 ° C in a  40–70% humidity environment. The animals were randomly divided into an experimental group (30 rats) and a control group  (12 rats). The rats were fasted for 12 h with access to water before  model preparation and fasted for 24 h with access to water after  the disease model was induced. After the 24-hour fasting period,  the rats were given free access to water and standard rat chow.  The rats in the experimental group were given intraperitoneal  injection of cerulein at a dose of 50  g/kg b.w. once an hour for  a total of four times. At the same time points, the rats in the control group were given 0.5 ml normal saline. Five rats from the  experimental group and 2 from the control group were sacrificed  by cervical dislocation at 6 h, 1, 2, 3, 5 and 7 days after model  preparation. The animals that were sacrificed were intraperitoneally injected with BrdU at a dose of 100 mg/kg b.w. 6 and 3 h  before sacrifice.

Pathological Morphology

both groups to observe histological changes. Serial paraffin sections were prepared at 5  m for conventional HE staining. Pathological changes in pancreatic tissues from both groups were observed under light microscopy.

Immunohistochemistry

Nestin and BrdU were detected using the SABC staining method. Slides were treated with polylysine to prevent the sections from  falling off. Paraffin sections were cut at 5  m, dewaxed in xylene,  rehydrated in graded alcohol to water, and washed in 0.01 M PBS  three times for 2 min each time. Subsequently, the sections were  incubated in 3 M H 2 O 2 solution for 10 min to inactivate peroxidase  and microwaved in citric acid buffer at 100 ° C for 10 min for antigen retrieval. After naturally cooling the sections for 25 min, the  sections were washed in 0.01 M PBS three times for 2 min each time  and blocked using normal goat serum solution at room temperature for 10 min. After removing excess liquid, the primary antibody, rabbit anti-rat IgG, was added to the sections without a wash  and incubated at 4 ° C overnight. Then the sections were washed in  0.01 M PBS three times for 2 min each. The secondary antibody,  biotin-conjugated goat anti-rabbit IgG, was added to the sections  and incubated at 37 ° C for 20 min, followed by three washes with  0.01 M PBS for 2 min each time. SABC reagent was added to the  slides and incubated at 37 ° C for 20 min. Next, the slides were  rinsed in 0.01 M PBS three times for 2 min each time and developed  with DAB using a DAB staining kit. One drop each of reagents A,  B, and C was added to 1 ml of distilled water and mixed well. The  mixture was added to the sections. The sections were restained  with hematoxylin for 50 s and then dehydrated, cleared, and  mounted. PBS buffer was used as a negative control for the primary antibody. Cells with brown particles specifically distributed in  the cytoplasm and the nucleus were considered to be positively  stained. The positive indices of the sections were analyzed using  the Motic Image Advanced 3.2 microscopic image analysis system.

Real-Time PCR  

The primers were designed using Primer 5.0 and synthesized  by Sangon Biotech (Shanghai). The PDX-1 upstream primer was 5  -GCTAATGGTGGACCGCAAC-3  , and the downstream  primer was 5  -GCAGTGAGCACTGAAGCGA-3  . The length of  the amplified fragment was 290 bp. The upstream primer for the  internal reference of  -actin was 5  -TGACGTGGACATCCGCAAAG-3  , and the downstream primer was 5  -CTGGAAGGTGGACAGCGAGG-3 . The length of the amplification fragment was 210 bp. Total RNA was extracted using the one-step  TRIzol method. The reverse transcription reaction mixture contained 10  l of 5 ! buffer, 5  l of 10 mmol/l dNTP, 1  l of RNase  inhibitor, 3  l of 100 mg/l Oligo(dT), 1  l of total RNA (2–4  g),  2  l of 200 U/  l M-MuLV, and RNase-free water for a total volume of 50  l. The reaction was incubated at 42 ° C for 6 min and  95 ° C for 5 min and then was rapidly cooled to 10 ° C in an ice bath  and stored at –40 ° C until use. The real-time PCR reaction contained 12.5  l of 2 ! ExTaq and 0.8 l each of 10 M upstream  and downstream primer, 2  l of cDNA, and RNase free water for  a total volume of 25  l. The reaction condition was optimized to  for the following steps: (1) pre-denature at 95 ° C for 3 min; (2) denature at 94 ° C for 20 s, anneal at 56 ° C for 30 s, and extend at 72 ° C  for 30 s for 40 cycles; (3) denature at 95 ° C for 1 min; (4) anneal at  55 ° C for 20 s for 80 cycles.

Analysis of Data

The data were expressed as mean 8 SD. Differences between  the groups were evaluated with ANOVA, followed by post-hoc  testing with Fisher’s least significant difference method. p ! 0.05  was considered statistically significant.

Results

The pathophysiological progression of acute pancreatitis can be divided into three phases ( fig. 1 ). First, there  is the edema and apoptosis phase (6 h to 1 day), where the  specimens generally display a high degree of edema without obvious hemorrhage. Under a light microscope, acinar cells showed apparent edema with a lot of vacuolelike structures dispersed in the pancreatic acini and islets. Concentrated dark-stained clump-like materials  were visible in the vacuole-like structures. The peak  amount of vacuole-like structures occurred between 6 h  and 1 day. After day 1, the amount of vacuole-like structures gradually decreased. The interlobular space significantly increased, which was likely caused by inflammatory edema. A small number of inflammatory cells were  found to be clustered in the interlobular region and the  cells gradually spread between the acini. According to the  literature and our observations, a large number of acinar  cells undergo active apoptosis at the early stage of inflammation due to the self-protective mechanism of pancreatic tissues against self-digestion. The concentrated  clump-like material we observed in the vacuolar structures may be apoptotic cells [27, 28] . In the second phase,  the hemorrhagic necrosis phase (2–3 days), the edema of  the specimens was significantly diminished compared to  the edema observed in the previous phase with dispersed  hemorrhagic spots. Interlobular vascular damage was detected with a microscope along with an effusion of red  blood cells from the vascular lumen. A large number of  red blood cells accumulated in the interlobular region.  Some of the pancreatic lobules exhibited patchy necrosis.  Acinar-like cells were rarely observed in the necrotic regions, but large amounts of inflammatory cells were  found in these areas and macrophages were the most

Fig. 1. Pathophysiological progression of acute pancreatitis: at 6 h  following cerulein injection, acinar cells showed apparent edema  with a lot of vacuole-like structures dispersed in the pancreatic  acini and islets, and the peak amount of vacuole-like structures  occurred in 1 day ( a , b ). At 2–3 days, edema of the specimens was  significantly diminished, and interlobular vascular damage was  detected along with an effusion of red blood cells from the vascular lumen, and some red blood cells accumulated in the interlobular region ( c , d ). At 5–7 days, pancreatic edema subsided, and the  reduction in the number of red blood cells accumulated between  the acini and in the interlobular region was observed, and a clear  boundary detected between the necrotic area and the normal acinar cells ( e , f ).

Fig. 2. Distribution characteristics of nestin-positive cells: at 6 h,  nestin-positive cells were mainly found in the lumen of interlobular vessels in clusters ( a , b ). At 1 day, the nestin-positive cells were  primarily distributed in the interlobular region and in the vascular cavity surrounding the tissues. Some of these cells had an outside distribution along the vascular lumen in two parallel lines ( c ,  d ). At 2 days, nestin-positive cells spread to the glandular lobules  and the pancreatic islets, and the majority of the positive cells were  in the interlobular region ( e , f ). At 3 days, large numbers of nestinpositive cells were found in the pancreatic lobules, and the positive  cells in the pancreatic islets increased significantly ( g , h ). At 5  days, the number of nestin-positive cells were significantly reduced ( i ). At 7 days, the positive cells were rarely found in the  glandular lobules and the pancreatic islets ( j ).

common cell type. A small number of inflammatory cells  were found among the acinar cells within the normal  pancreatic lobules. In the third phase, the reconstruction  phase (5–7 days), pancreatic edema subsided and there  were patches of dark red. The reduction in the number of  red blood cells accumulated between the acini and in the  interlobular region was observed under a microscope. A  clear boundary was detected between the necrotic area  and the normal acinar cells. Large quantities of inflammatory cells were found in the necrotic region. Among  these cells, a large number of macrophages were detected  with inflammatory cells scattered around them. The interacinar space returned to normal and vacuole-like  structures were rarely observed.

Distribution Characteristics of Nestin-Positive Cells   ( fig. 2 )

The nestin protein was expressed at a low level in normal pancreatic tissues with a diffused distribution in acinar cells and islet cells. Six hours after the disease model was prepared, the nestin-positive cells were mainly  found in the lumen of interlobular vessels in clusters. A  small number of positive cells were found around the  vascular cavity. Nestin-positive cells were rarely detected  in glandular lobules and pancreatic islets. One day after  the cerulein was injected, the nestin-positive cells were  still primarily distributed in the interlobular region and  in the vascular cavity surrounding the tissues. Some of  these cells had an outside distribution along the vascular  lumen in two parallel lines. On day 2, after acute pancreatitis was induced, nestin-positive cells spread to the  glandular lobules and the pancreatic islets with a large  number of nestin-positive cells dispersed among the acinar cells. Nestin-positive cells were also detected within  the pancreatic islets and the majority of the positive cells  were in the interlobular region. On day 3, large numbers  of nestin-positive cells were found in the pancreatic lobules, and the positive cells in the pancreatic islets increased significantly. On day 5, the number of nestinpositive cells were significantly reduced within the pancreatic lobules and the islets. Few nestin-positive cells  were observed in the interlobular region and within the  lumen of the interlobular vessels. On day 7, the nestinpositive cells continued to decrease, and the positive cells  were rarely found in the glandular lobules and the pancreatic islets.  Distribution Characteristics of BrdU-Positive Cells   ( fig. 3 )

Fig. 3. Distribution characteristics of BrdU-positive cells: at 6 h,  BrdU-positive cells were rarely found ( a ). At 1 day, a small quantity of BrdU-positive cells were observed in the acinar tissues ( b ).  At 2–3 days, more BrdU-positive cells were distributed around the  interlobular region and within the inflamed tissues ( c , d ). At 5  days, a large number of BrdU-positive cells were observed in the  pancreatic lobules and the islets ( e , f ). At 7 days, the BrdU-positive  cells connected in patches in the inflammatory cell elimination  areas ( g , h ).

BrdU-positive cells were rarely found in normal tissues.  Six hours to 1 day after cerulein injection, a few BrdUpositive cells were observed in the acinar tissues. More BrdU-positive cells were distributed in the region with accumulated inflammatory cells, which may be related to the  proliferation of macrophages. From day 2 to day 3, a small  amount of BrdU-positive cells were distributed around the  interlobular region, and the number of BrdU-positive cells  within the inflamed tissues continued to increase. On day  5, a large number of BrdU-positive cells were observed in  the pancreatic lobules and the islets. The inflammatory  cells in the inflamed tissues were not evenly distributed.  Large amounts of BrdU-positive cells were distributed in  the ‘elimination areas’ with few inflammatory cells. On  day 7, a few BrdU-positive cells were observed in normal  pancreatic tissues. Some of the inflammatory cell elimination areas were locally connected in patches. Many BrdUpositive cells were observed in these patches.

Expression of PDX-1 mRNA in Pancreatic Tissues   ( fig. 4 )

The PDX-1 mRNA expression levels of the control  group and the pancreatitis model group at different time  points are shown in figure 5 . The electrophoresis results for the PCR products are shown in figure 4 . Overall, PDX- 1 mRNA was positively expressed in the normal pancreatic tissues, but at a low level. After pancreatitis was induced, the expression level of the PDX-1 gene gradually  increased and reached its peak on day 3. The expression  gradually decreased afterwards and got close to a normal  level on day 7.

Fig. 4. After pancreatitis was induced, the  expression level of the PDX-1 gene gradually increased and reached its peak on day  3. The expression gradually decreased afterwards and got close to a normal level on  day 7.

Fig. 5. Expression levels of the PDX-1 mRNA: the expression level of the PDX-1 gene reached its peak on day 3. *  p ! 0.05 vs. controls, #   p ! 0.01 vs. controls.

Discussion

In recent years, determining whether of stem cells are  present in pancreatic tissues has been controversial. Also,  the localization and the role of stem cells in pancreatic  tissue repair, if they exist in pancreatic tissue, remains  unknown. Based on the present study, we believe that primary stem cells likely do not exist in adult pancreatic tissues, and bone marrow stem cells may be the original  source of the stem cells identified in adult pancreatic tissues by other researchers [4, 5] . The following evidence  supports our conclusions. First, the nestin protein is considered to be an important surface marker of pancreatic  stem cells. In our current study, we found very few nestinpositive cells in normal pancreatic tissues. However, in  the pancreatitis model group, an increased number of  positive cells were detected in the pancreatic interlobular  region at 6 h following cerulein injection. These positive  cells were mainly found in the interlobular vascular lumen. Some of these cells crossed the vascular wall and  lined the blood vessels, which indicated that these nestinpositive cells likely come from the bone marrow hematopoietic system. It has been showed in other studies that  bone marrow stem cells positively express nestin [15–17] .  Second, within 1–3 days following the cerulein injection,  the number of nestin-positive cells gradually increased in  the interlobular regions and their surrounding tissues.  The increase may have resulted from bone marrow stem  cells that penetrated the interlobular blood vessels and  spread into the surrounding tissues of the interlobular  region such as the pancreatic acini and the islets. Third,  5 days after the cerulein injection, the number of nestinpositive cells gradually decreased and the majority of  these cells were distributed around the interlobular vessels. Some of the cells were also observed along blood vessels, which may be related to the elimination of the stimulus for pancreatic injuries and the initiation of pancreatic tissue repair. Fourth, in a study on cell proliferation  following pancreatic injuries, BrdU-positive cells were  initially found in the acinar tissue around the pancreatic  interlobular region. This may be related to the involvement of stem cells in the early repair of damaged tissues.  Fifth, real-time PCR analysis of the pancreatic homeobox  gene PDX-1 (pancreatic/duodenal homeobox-1) showed  that the expression level of PDX-1 gradually increased following cerulein injection and peak expression occurred  on day 3. Afterwards, the expression decreased then increased slightly and maintained a steady level of expression. The changes in PDX-1 expression were consistent  with the trend in the number of nestin-positive cells,  which indicates that the number of nestin-positive cells  may be regulated by the PDX-1 gene. The nestin-positive  cells identified in the present study are presumably pancreatic stem cells.

To be accurate, these nestin-positive cells are not pancreatic stem cells, but pancreatic progenitor cells with the  potential to differentiate into pancreatic acinar cells and  islet cells. Zulewski et al. [29] found that the nestin-positive cells isolated from pancreatic islets can differentiate  into pancreatic endocrine and exocrine cells. Although  no direct evidence was obtained for the differentiation of  nestin-positive cells into pancreatic cells, we surmise that  nestin-positive cells are involved in the repair of damaged  pancreatic tissues. The evidence is shown below. First,  nestin-positive cells, i.e. pancreatic stem cells, are among the pancreatic cells that produce an initial response to   tissue damage within pancreatic tissues. Only a few dispersed cells exhibited stem cell characteristics within  normal pancreatic tissues. Following tissue damage, the  stem cells immediately spread from the interlobular regions to the pancreatic tissues, which may be related to  the involvement of stem cells in tissue repair following  marked apoptosis of acinar and islet cells in the early  stage of pancreatitis. Second, a low apoptosis level of acinar and islet cells was also observed in normal pancreatic tissues [30] . Therefore, stem cell-based repair may be  present in normal pancreatic tissue as well. Third, as an  indicator of cell proliferation, BrdU-positive cells were  first found in the peripheral area of the interlobular regions and spread through the pancreatic tissues over  time, which is consistent with the location and diffusion  of nestin-positive cells.

If our conclusion that stem cells may not exist in adult  pancreatic tissues is tenable, then the pancreatic stem  cells discovered by many researchers may actually be the  bone marrow stem cells that entered the pancreatic tissues right after pancreatic tissue damage and became  progenitor cells with pluripotent differentiation potential. Therefore, we propose that bone marrow stem cells  can become progenitor cells with the potential to differentiate into acinar and islet cells as long as they are provided with a microenvironment similar to pancreatic tissues. Further studies on the relationship between bone  marrow stem cells and pancreatic cells will help develop  new treatments for SAP, a problem with worldwide significance.

References

1 Brignier AC, Gewirtz AM: Embryonic and  adult stem cell therapy. J Allergy Clin Immunol 2010; 125(suppl 2):S336–S344.  

2 Hyun I: The bioethics of stem cell research  and therapy. J Clin Invest 2010; 120: 71–75.   

3 Hviid Nielsen T: What happened to the stem  cells? J Med Ethics 2008; 34: 852–857.  

4 Madsen OD: Pancreas phylogeny and ontogeny in relation to a ‘pancreatic stem cell’. C R  Biol 2007; 330: 534–537.  

5 Yang C, Wang JM, Du CY, et al: Expression  of stem cell markers CK-19 and PDX-1  mRNA in pancreatic islet samples of different purity from rats. Hepatobiliary Pancreat  Dis Int 2007; 6: 544–548.   

6 Bhatia M, Wong FL, Cao Y, et al: Pathophysiology of acute pancreatitis. Pancreatology  2005; 5: 132–144.  

7 Gardner TB, Vege SS, Chari ST, et al: Faster  rate of initial fluid resuscitation in severe  acute pancreatitis diminishes in-hospital  mortality. Pancreatology 2009; 9: 770–776.   

8 Dor Y, Brown J, Martinez OI, et al: Adult  pancreatic  cells are formed by self-duplication rather than stem-cell differentiation.  Nature 2004; 429: 41–46.  

9 Levine F, Mercola M: No pancreatic endocrine stem cells? N Engl J Med 2005; 351:   1024–1026.   

10 Shi Y, Hou L, Tang F, et al: Inducing embryonic stem cells to differentiate into pancreatic  cells by a novel three-step approach  with activin A and all- trans -retinoic acid.  Stem Cells 2005; 23: 656–662.   

11 Oh SH, Muzzonigro TM, Bae SH, et al: Adult  bone marrow-derived cells trans -differentiating into insulin-producing cells for the  treatment of type I diabetes. Lab Invest 2004;   84: 607–617.  

12 Messam CA, Hou J, Berman JW, et al: Analysis of the temporal expression of nestin in  human fetal brain derived neuronal and glial progenitor cells. Brain Res Dev Brain Res  2002; 134: 87–92.  

13 Lendahl U, Zimmerman LB, McKay RD:  CNS stem cells express a new class of intermediate filament protein. Cell 1990; 60: 585– 595.   

14 Dahlstrand J, Zimmerman LB, McKay RD, et  al: Characterization of the human nestin gene  reveals a close evolutionary relationship to  neurofilaments. J Cell Sci 1992; 103: 589–597.

15 Wislet-Gendebien S, Hans G, Leprince P, et  al: Plasticity of cultured mesenchymal stem  cells: switch from nestin-positive to excitable  neuron-like phenotype. Stem Cells 2005; 23:   392–402.   

16 Abraham EJ, Leech CA, Lin JC, et al: Insulinotropic hormone glucagon-like peptide-1  differentiation of human pancreatic islet-derived progenitor cells into insulin-producing  cells. Endocrinology 2002; 143: 3152–3161.   

17 Huang H, Tang X: Phenotypic determination and characterization of nestin-positive  precursors derived from human fetal pancreas. Lab Invest 2003; 83: 539–547.  

18 Hunziker E, Stein M: Nestin-expressing cells  in the pancreatic islets of Langerhans. Biochem Biophys Res Commun 2000; 271: 116– 119.   

19 Humphrey RK, Bucay N, Beattie GM, et al:  Characterization and isolation of promoterdefined nestin-positive cells from the human  fetal pancreas. Diabetes 2003; 52: 2519–2525.   

20 Street CN, Lakey JR, Seeberger K, et al: Heterogenous expression of nestin in human  pancreatic tissue precludes its use as an islet  precursor marker. J Endocrinol 2004; 180:   213–225.   

21 Ishiwata T, Kudo M, Onda M, et al: Defined  localization of nestin-expressing cells in L - arginine-induced acute pancreatitis. Pancreas 2006; 32: 360–368.  

22 Dutta S, Gannon M, Peers B, et al: PDX: PBX  complexes are required for normal proliferation of pancreatic cells during development.  Proc Natl Acad Sci USA 2001; 98: 1065–1070.  

23 Jonsson J, Carlsson L, Edlund T, et al: Insulin  promoter-factor 1 is required for pancreas  development in mice. Nature 1994; 371: 606– 609.  

24 Taguchi M, Otsuki M: Co-localization of  nestin and PDX-1 in small evaginations of  the main pancreatic duct in adult rats. J Mol  Histol 2004; 35: 785–789.   

25 Gagliardino JJ, Del Zotto H, Massa L, et al:  Pancreatic duodenal homeobox-1 and islet  neogenesis-associated protein: a possible  combined marker of activateable pancreatic  cell precursors. J Endocrinol 2003; 177: 249– 259.   

26 Liu T, Wang CY, Gou SM, et al: PDX-1 expression and proliferation of duct epithelial  cells after partial pancreatectomy in rats.  Hepatobiliary Pancreat Dis Int 2007; 6: 424– 429.  

27 Bhatia M: Apoptosis versus necrosis in acute  pancreatitis. Am J Physiol 2004; 286:G189– G196.   

28 Campo GM, Avenoso A, Campo S, et al:  Chondroitin-4-sulphate reduced oxidative  injury in caerulein-induced pancreatitis in  mice: the involvement of NF-  B translocation and apoptosis activation. Exp Biol Med  2008; 233: 741–752.  

29 Zulewski H, Abraham EJ, Gerlach MJ, et al:  Multipotential nest in positive stem cells isolated from adult pancreatic islets differentiate ex vivo into pancreatic endocrine, exocrine, and hepatic phenotypes. Diabetes  2001; 50: 521–533.   

30 Trulsson L, Sandström P, Sundqvist T, et al:  The Influence of a load of L -arginine on serum amino acids and pancreatic apoptosis/ proliferation and ATP levels in the rat. Pancreas 2004; 29: 113–120.