Journal of Undergraduate Research
Volume 9, Issue 3 - Spring 2008

Transdifferentiation of Liver Stem Cells into Insulin-Producing Cells for a Cure for Type 1 Diabetes

Wajeeh Irfan

ABSTRACT

Diabetes is a disease in which the body does not produce or properly use insulin. The transplantation of insulin-producing ß-cells from donors can effectively alleviate diabetes, but their availability is limited. In this study, we investigated the possibility of creating insulin-producing cells from adult rat liver stem (WB) cells with the capacity to respond to glucose challenges. The transdifferentiation of these liver stem cells was conduced by examining the role of transcription factor expression, particularly pancreatic duodenum homeobox protein 1 (Pdx-1) and Neurogenin3 (Ngn-3), which were shown to be essential in activation of the insulin gene, leading to insulin expression. Lentiviral vectors were used to introduce the transcription factors into WB stem cells. It was shown that Pdx-1and Ngn-3 are required to turn on the insulin genethrough an insulin reporter gene coupled with the expression of green fluorescent protein (GFP). Reverse-transcription polymerase chain reaction was conducted to observe insulin gene expression by testing for endogenous expression of various transcription factors including Pdx-1, Ngn-3, MafA, and insulin. The positive bands noted after electrophoresis for these transcription factors showed that the insulin gene was indeed being activated. This study shows that functional insulin-producing cells may be generated from liver stem cells in vitro by transdifferentiation and thus may provide a new perspective for ß-cell replacement therapy. However, as more is learned about insulin gene transcription, it will become more evident that the expression of insulin in vivo requires the integrated contributions of various factors. Due to the complex nature of these interactions, further confirmation is needed to ensure that liver stem cells can be transdifferentiated into insulin-producing cells, as shown in vitro in this study, in humans.

INTRODUCTION

Diabetes, a disease in which the body does not produce or properly use insulin, affects 20.8 million people in the United States. Insulin is a hormone that is essential in converting sugars, starches, and other foods into energy required for daily activity (1). The cause of diabetes remains largely unknown, although genetics and environmental factors, such as diet and exercise, play key roles in the presence of the disease. Although diabetes is a serious illness, people who manage their condition effectively can live long, healthy, satisfying lives. However, the deleterious effects of diabetes and related complications can eventually lead to the deterioration of quality of life and sometimes result in death.

Several types of diabetes have been identified. Type 1 diabetes is an insulin-dependent, autoimmune disorder characterized by the destruction of insulin-producing β-cells (2). Therefore, a reversal of type 1 diabetes could be achieved through replacement of functional β-cells. Historically, the transplantation of β-cells has had major challenges due to immune system rejection, as well as the scarcity of available donor cells (7). These common downfalls can be avoided by employing new methods of cell differentiation and self-renewal involving adult stem cells.

Insulin-producing cells arise from the pancreas. Recent studies have shown that liver and pancreas cell populations have a close relationship because they both originate from the same embryonic tissue layer (14). Whether these cells differentiate to liver or pancreas tissue is dictated by their locations, cell adhesion molecules, and transcription factors. Evidence suggests that human pancreatic stem cells possess the capability to differentiate into liver cells, however, the capacity of human liver stem cells to differentiate into pancreatic cells remains largely unknown.

In this study, we are primarily interested in the plasticity of adult liver stem cells, and their capability to transdifferentiate into insulin-producing pancreas cells, with the capacity to synthesize insulin in response to glucose challenges. We examined the transdifferentiation of these liver stem cells by measuring the potency of combined transcription factor expression, particularly pancreatic duodenum homeobox protein 1 (Pdx-1) and Neurogenin3 (Ngn-3), on the production of insulin.

Transcription factors are proteins that regulate transcription by regulating the binding of RNA polymerase, which initiates transcription of a gene. By binding either upstream or downstream to a promoter region, transcription factors serve to repress or enhance gene expression by blocking or assisting the binding of RNA polymerase (6). Translation, or production, of a protein follows transcription. In this study, activation of the insulin gene, which leads to the translation of the insulin protein hormone, will be monitored.

Insulin functions as the main regulator of blood glucose levels. Produced exclusively by ß-cells located in the endocrine islets of the pancreas, the release of insulin is controlled by fluctuations in blood glucose concentrations (10). The ß-cell-specific expression of the insulin gene is mainly controlled at the transcriptional level through distinct elements located in the promoter region that interact with Pdx-1 and Ngn-3.

Pdx-1 is a transcription factor that is essential for pancreatic development and mature ß-cell function (8). It has been shown to convert hepatocytes into both exocrine and endocrine pancreatic cells in mice, but it fails to selectively convert hepatocytes exclusively into insulin producing cells (9). The expression of this protein is essential for the pancreas to mature, and is involved in regulating the expression of numerous genes responsible for insulin-producing β-cells to function effectively. Ngn-3 is the key transcription factor in orchestrating the determination of cell-type fates towards pancreatic endocrine cells (12). The location of Pdx-1 and Ngn-3 in the pancreatic development pathway can be noted from Scheme 1. This pathway shows the progression of the progenitor cell to endocrine cells such as β-cells. While manipulating these transcription factors we proposed to study the WB stem cell line to measure its capability to transdifferentiate into insulin-producing cells.

Phase 1. Transcription factor control of pancreas development

WB cells are adult rat liver stem cells that are capable of expressing precursors for insulin production. These cells are simple epithelial cells that express a phenotypic repertory that overlaps, but remains different from, that of both bile duct epithelial cells and hepatocytes (15). Due to their complex phenotype, WB cells may be embryonic or undifferentiated versions of either bile duct epithelial cells or hepatocytes. One of the main reasons liver stem cells were selected for this study, in addition to the fact that liver cells arise from the same embryonic germ layer (the endoderm) as pancreatic cells, was because both liver and pancreas cells have glucose-sensing systems. These two advantages that liver stem cells possess made them the best choice for transdifferentiation into cells with pancreatic cell behavior.

Lentiviral (LV) vectors were employed to introduce the transcription factor genes into the nucleus of the WB cells. They were also used to express green fluorescent protein (GFP) driven by an insulin-1 promoter. The lentivirus functions by entering the nucleus of the WB cells, integrating into the genome, and constitutively expressing protein (4). Using a lentiviral vector has numerous advantages over other methods of delivering transcription factors into the cells. Lentiviral vectors do not elicit an immune response in vivo, can be pseudo-typed with ease to infect specific tissues and cell lines with high efficiency, and can infect both dividing and non-dividing cells, unlike other retroviruses. Furthermore, lentiviral vectors retain stable and lasting heritable expression (5).

Transdifferentiated WB cells can express pancreatic cell differentiation-related transcripts, such as RNA precursors for insulin production, which are detectable by laboratory techniques such as reverse-transcription polymerase chain reactions (RT-PCR). Visualization of the protein following RT-PCR can determine insulin content and release and the cells can be observed to see if they effectively differentiated into insulin-producing entities.

It is hypothesized that the liver stem cells due to their plasticity will transdifferentiate along new lineages when transduced with a combination of Pdx-1 and Ngn-3 into insulin-producing cells. Previous studies have shown that Pdx-1 alone is able to activate both pancreatic endocrine and exocrine genes, but its expression may have deleterious side effects (14). Therefore, in this study, Pdx-1 and Ngn-3 were used in combination to avoid the side effects from expression of Pdx-1 alone. The purpose of this study is to create a source of insulin-producing cell surrogates from liver stem cells via lentiviral mediated expression of transcription factors Pdx-1 and Ngn-3. It is expected that these transcription factors will regulate the gene expression for the release of insulin. Results from this analysis may provide evidence that cells from other organs, besides the pancreas, can be transdifferentiated to produce insulin in a similar fashion as pancreatic ß-cells. A novel source of insulin-producing cells can aid in curing type 1 diabetes.

MATERIALS AND METHODS

Vector Construction

Lentiviral constructs were used to introduce the transcription factors into the WB stem cells. These constructs contained mouse Pdx-1 or Ngn-3 genes and were constructed by inserting the mouse Pdx-1 coding sequence or Ngn-3 sequence into the Ptyf vector while under the control of the elongation factor-1a (EF1α) promoter. A Pdx-1-VP16 expression vector was made by combining the coding sequence for the 80-residue VP16 activation domain to the C-terminus of mouse Pdx-1.          

The plasmid pTYF-EFPdxVP16 was used to insert transcription factors into the liver stem cells. The promoter, CMV-TATA-TAR, is the region of DNA located upstream (towards 5' region) of the gene that is required for its transcription (13). It selectively allows and represses the gene it controls. The EF1α Enhancer is the region of DNA that binds with proteins and in turn regulates the binding of RNA polymerase, which increases the activity of the promoter (11). The Pdx-1-VP-16 region serves as a selectable marker, and during cloning only Escherichia coli bacterial colonies possessing specific ampicillin resistance genes were able to grow. The numerous factors labeled on the outside of the plasmid map (Phase 2), such as HinD III 5434, are palindromic sequences recognized by specific restriction enzymes as sites for cleavage.

Phase 2. Plasmid map for pTYF-EFPdxVP16

In order to examine activity of various genes related to insulin-regulation, we constructed lentiviral vectors conjugating GFP or red fluorescent protein (RFP) to the promoters of these various genes. For example, the insulin promoter was conjugated to GFP so that expression of GFP acted as a reporter to demonstrate insulin promoter activation. The same method was used to examine the activity of other transcription factors involved in pancreatic development besides Pdx-1 and Ngn-3, such as Pax-4.

Plasmid isolation

E. coli bacteria cells were transformed with the constructed plasmids. The colonies were selected for on ampicillin antibiotic plates. For the large scale isolation of the plasmid, the cells were incubated in LB broth with the ampicillin in a 37°C bacterial shaker. Following a 12 hour incubation period, the plasmid was isolated using the Maxi prep kit (Qiagen) from the bacterial cells.

Virus preparation

Lentivirus was produced and the titer was determined by the laboratory of Dr. Lung-Ji Chang in Department of Molecular Genetics and Microbiology at the University of Florida College of Medicine in Gainesville, FL (3).

Transduction of vector into WB cells

Rat liver stem (WB) cells were grown in a six-well plate in RPMI medium, supplemented with 10% FCS in a 5% CO2 incubator. At 70% confluence, these cells were then transduced with an overactive form of Pdx-1 (PV), Ngn-3, or both genes, at a multiplicity of infection (MOI) of 20. Polybrene (1:2000 dilution) was used to enhance the transduction efficiency. Transduction efficiency was determined by transducing the WB cells with LV- EF1α -GFP for 72 hours. The cells were examined for reporter gene expression (GFP) using fluorescence microscopy to determine promoter activity.

RNA isolation / cDNA preparation

RNA was isolated from the pancreas of mice using Trizol reagent and manufacturer’s protocols. cDNA was then synthesized by RT-PCR using Superscript III reverse transcriptase enzyme (Invitrogen, CA) with random hexamer primers. Table 1 shows the amounts of solutions for the RT conducted. After making this 11 µl mixture, the solution was incubated at 70° C and added to a second mixture consisting of PCR solutions indicated in Table 2. This 20 ml solution was then incubated at 42° C for 1 hour.

 

Polymerase chain reaction

Total RNA was prepared from the WB cells using Trizol reagent. Transcriptional gene expression related to the pancreas was determined by RT-PCR. All primers used were purchased from Life Technologies. Thirty cycles were used for all PCRs performed by a PCR machine, and Table 3 shows the PCR cycle conditions. PCR machines, which are also known as thermal cycles, rapidly change temperatures for PCR reactions, allowing PCR to cycle between primer annealing, DNA amplification, and strand melting cycles. PCR products were separated by electrophoresis using 2.0% agarose gels.

RESULTS

Transduction efficiency of lentiviral vectors

In order to verify if the lentivirus vector had successfully integrated into the WB cell genome, the transduction efficiency was examined. Transduction was assessed using the constitutive expression of GFP by the lentivirus. For this, the WB cells were transduced with LV-pEF1α-GFP. Following transduction, the cells were observed for GFP expression using fluorescence microscopy. The GFP fluorescence was visible under the microscope and transduction efficiency was measured to be more than 75%. Figure 1b is representative of this efficiency at x100 original magnification. Figure 1a and 1c are phase pictures highlighting the number of viable cells in the culture. These pictures accompany all result pictures from this study. Figure 1d shows that no fluorescence was observed before cells were transduced with LV-pEF1α-GFP.

Figure 1. Transduction efficiency of lentiviral vectors in liver stem cells. A: Phase picture of viable WB cells. (Original magnification: x100) B: WB cells transduced with LV- pEF1α-GFP. C: Phase picture of control WB cells D: No fluorescence observed before cells were transduced with LV- pEF1α-GFP.

Pdx-1-VP16 (PV) is an overactive form of Pdx-1. PV and Ngn-3 alone can turn on the NeuroD promoter, but not the insulin reporter gene. The NueroD promoter controls the transcriptional factor NeuroD, found later in the differentiation pathway of pancreatic endocrine cells than Pdx-1 or Ngn-3. Following transduction of WB cells with LV-PV or LV-Ngn-3 with GFP, the NeuroD promoter was activated as shown in Figure 2.

Figure 2. WB cells showing activation of NeuroD promoter. A: Phase picture of viable WB cells. (Original magnification: x400) B: WB cells tranduced with LV-pNueroD-GFP on day 3 showing activation of NeuroD promoter by PV. C: Phase picture of viable WB cells (Original magnification: x400) D: WB cells tranduced with LV-pNueroD-GFP on day 3 showing activation of NeuroD promoter by Ngn-3.

Confirmation of transcription factor protein expression

Immunocytochemical staining was conducted using anti-Pdx-1 and anti-Ngn-3 antibodies to confirm transcription factor protein expression. Figure 3 shows protein expression following transduction of WB cells with LV-PV, LV-Ngn-3, or a combination of LV-PV and LV-Ngn-3. Figure 3a and 3b show negative controls illustrating that transduction of these transcription factors did indeed lead to this expression.

Figure 3. Confirmation of transcription factor protein expression by immunocytochemistry. A:  WB cell isotype negative control (GFP). (Original magnification: x400) B: WB cells transduced with LV-PV and LV-Ngn-3 negative control showing signal background C: WB cells transduced with LV-Ngn-3 using anti-Ngn-3 antibodies D: WB cells transduced with LV-PV and LV-Ngn-3 using anti-Ngn-3 antibodies E:WB cells transduced with LV-PV using anti-Pdx-1 antibodies F: WB cells transduced with LV-PV and LV-Ngn-3 using anti-Pdx-1 antibodies. 

PV and Ngn-3 are required to turn on the insulin gene

To observe the effects of PV and Ngn-3 on insulin gene activity, the cells were transduced with LV-PV, LV-Ngn-3, or a combination of LV-PV and LV-Ngn-3 with reporter constructs insulin-GFP and pPax-4-RFP. Then the cells were observed for the reporter gene expression every 24 hours under the fluorescence microscope. On day 4, both green and red florescence was observed, which indicate that the insulin and Pax-4 promoters were activated, as shown under x100 magnification in Figure 4e and Figure 4f respectively. The Pax-4 promoter controls the transcriptional factor Pax-4, which is found later in the differentiation pathway of pancreatic endocrine cells than Pdx-1 or Ngn-3. Figure 4b shows WB cells treated with LV-PV alone. This figure shows no insulin reporter gene expression, thus alone it is unable to activate expression.

Figure 4. Activation of insulin and Pax-4 promoters with combination of PV and Ngn-3. A: Phase picture of viable WB cells. (Original magnification: x100) B: WB cells transduced with LV-pInsulin-GFP on day 4 showing that LV-PV alone does not activate insulinpromoter C: WB cells transduced with LV-pPax-4-RFP on day 4 showing that LV-PV alone does not activate Pax-4 promoter. D: Phase picture of viable WB cells. (Original magnification: x100) E: WB cells transduced with LV-pInsulin-GFP on day 4 showing that combination of PV and Ngn-3 activate insulin promoter. F: WB cells transduced with LV-pPax-4-RFP on day 4 showing that combination of PV and Ngn-3 activate Pax-4 promoter.

Three-dimensional cell clustering promotes cell differentiation

To report the activity of the insulin promoter and Pax-4 gene, WB cells were first transduced with LV-RIP-GFP and LV-Pax-4-RFP.  Then the cells were transduced with LV-PV and LV-Ngn-3 while being cultured under high glucose conditions. Figure 5 shows the formation of three-dimensional cell clusters by the WB cells. Clustering of insulin-producing cells is important because islet-like structures are essential for eliciting physiological responses to fluctuations in blood glucose levels. This figure also shows that these liver stem cells expressed both insulin and Pax-4 genes.

Figure 5. Liver stem cells formed 3-D cell clusters and expressed both insulin (green) and Pax-4 (red) genes. A:  WB cell cluster control. (Original magnification: x400) B: WB cells transduced with LV-RIP-GFP C: WB cells transduced with LV-Pax-4-RFP D: WB cells transduced with both LV-PV and LV-Ngn-3.

Transcriptional profile of the genetically modified WB cells

To confirm that the insulin gene was activated, endogenous expression of various transcription factors was tested by PCR using different primer sets. For instance, Ngn-3 primers were used when examining the role of Ngn-3 in activating the insulin promoter. The results are shown in Figure 6 and 7. The positive bands corresponding to PV, Ngn-3, or a combination of PV and Ngn-3 cDNA, show that the insulin promoter is indeed being activated by the transcription factors Pdx-1, Ngn-3, and MafA. cDNA from insulinomas, which are pancreatic endocrine tumor genes, was used to show that the PCR was functioning properly and served as a positive control.

Figure 6. RT-PCR data showing endogenous expression of insulin, MafA, and PV.

 

Figure 7. RT-PCR data showing endogenous expression of Ngn-3

DISCUSSION

The purpose of this study was to transdifferentiate WB liver stem cells into insulin-producing cell surrogates via lentiviral mediated expression of transcription factors Pdx-1 and Ngn-3. It was hypothesized that the liver stem cells would transdifferentiate along new lineages, due to their plasticity, into insulin-producing cells and that these transcription factors would regulate gene expression for the release of insulin, thereby achieving physiologically blood glucose conditions and a cure for type 1 diabetes.

In order to verify if the lentiviral vector had successfully integrated into the WB cell genome, the transduction efficiency was examined and shown to be more than 75%. It was also shown that LV-PV, an over-active form of Pdx-1, and LV-Ngn-3 alone could activate the NeuroD promoter gene, but not the insulin promoter gene.  Through the fluorescence of GFP, it was shown that LV-PV and LV-Ngn-3 are required in combination to activate the insulin gene. WB cells treated with just LV-PV showed no insulin promoter activation, thus LV-PV alone was unable to activate GFP expression.

Multiple sequences along the promoter region of the insulin gene contribute to its activity by serving as binding sites for sequence-specific DNA-binding proteins found in the nucleus of the ß-cell. In order to confirm that the insulin promoter gene was activated, the endogenous expression of various transcription factors such as Pdx-1, Ngn-3, MafA, and insulin, was tested by RT-PCR. Positive bands using PV, Ngn-3, or a combination of PV and Ngn-3 cDNA showed that the insulin promoter gene was indeed activated by these transcription factors. 

Insulin and MafA, the upstream transcription factor to insulin, were also tested because they are end markers in the transdifferentiation pathway of liver stem cells into insulin-producing cells. Pdx-1 and Ngn-3, the two transcription factors that this study focused on, are both beginning transcription factors that up regulate intermediate transcription factors, such as Nkx2.2 and Nkx6.1. These intermediates bind to the insulin promoter and lead to insulin gene transcription. Checking the expression of these intermediate transcription factors can provide further evidence that Pdx-1 and Ngn-3 transdifferentiate liver stem cells into insulin-producing cells.

These studies confirmed that liver stem cells could transdifferentiate along new lineages into insulin-producing surrogates in vitro. Future studies should examine transplantation of these insulin-producing cells in vivo to determine their ability to restore physiological blood glucose levels in humans. However, as more is learned about the molecular mechanisms underlying insulin gene transcription and regulation, it will become more evident that the expression of insulin in vivo requires the integrated contributions of various factors. Due to the complex nature of these interactions, further confirmation is needed to ensure that liver stem cells can be transdifferentiated into insulin-producing cells, as shown in vitro in this study, in humans. Thus, it cannot be assumed that a complete understanding of insulin gene transcription and regulation has been attained.

---

ACKNOWLEDGEMENTS

This study was conducted in the laboratory of Dr. Lijun Yang in the Department of Pathology, Immunology, and Laboratory Medicine in the University of Florida College of Medicine. I am extremely appreciative for the supervision, training, and continuous support provided by Dr. Yang and the members of her laboratory, including Vijay Koya, Dr. Dong-Qi Tang, Dr. Shun Lu, Dr. Shi-Wu Li, and Cathy Sun.

---

REFERENCES

1. Atkinson MA, Eisenbarth. "Type 1 diabetes: New Perspectives on Disease Pathogenesis and Treatment." Lancet 358:221–229, 2001.
2. Cao, L.Z., Tang, D.Q., Li, S.W., Horb, M.A. and Yang, L.J. "External factors are necessary for Pdx-1 transfected hepatic cells to transdifferentiate into functional pancreatic endocrine insulin-producing cells." Diabetes 53(12):3168-3178, 2004.
3. Chang LJ, Zaiss AK. "Lentiviral vectors. Preparation and use." Methods Mol Med 2002;69:303–318.
4. Chen, X, Tang, D.Q., Xie, C.Q., Thompson, W.E., Chang, L.J, Chen, Y.E, & Yang, L.J. "Genetic Engineering of Human Embryonic Stem Cells with Lentiviral Vectors." Stem Cell and Development Aug; 14:367-377, 2005.
5. Chang LJ and E Gay. "The molecular genetics of lentiviral vectors - current and future perspectives." Curr Gene Ther 1:237–251, 2001.
6. Ferber S, Halkin A, Cohen H, Ber I, Einav Y, Goldberg I, Barshack I, Seijffers R, Kopolovic J, Kaiser N, Karasik A. "Pancreatic and duodenal homeobox gene 1 induces expression of insulin genes in liver and ameliorates streptozotocin-induced hyperglycemia." Nat Med 6:568 –572, 2000.
7. Habener JF. "The role of pancreatic duodenum homeobox protein-1 in the development of diabetes mellitus." Drug News Perspect 2002;15:491–497.
8. Hui H, Perfetti R. "Pancreas duodenum homeobox-1 regulates pancreas development during embryogenesis and islet cell function in adulthood." Eur J Endocrinol 2002;146:129–141.
9. Li Y, Zhang R. "Generation of insulin producing cells from PDX-1 gene modified human mesenchymal stem cells." J Cell Physiol Apr, 211 (1): 36-44, 2007.
10. Noguchi H, Xu G, Matsumoto S, Kaneto H, Kobayashi N, Bonner-Weir S,
    Hayashi S. "Induction of pancreatic stem/progenitor cells into insulin-producing cells by adenoviral-mediated gene transfer technology." Cell Transplant, 15 (10): 929-38, 2006.
11. Tang DQ, Lu S, Sun YP, Rodriguez E, Chou W, Yang C, Cao LZ, Chang L.J, & Yang, L.J. "Reprogramming liver-stem WB cells into functional insulin-producing cells by persistent expression of Pdx1- and Pdx1-VP16 mediated by lentiviral vectors." Laboratory Investigation  Jan;86(1):83-93 , 2006
12. Wang AY, Ehrhardt A, Xu H, Kay MA." Adenovirus transduction is required for the correction of diabetes using PDX-1 or Ngn-3 in the liver." Mol Ther Feb 15 (2): 255-63, 2007.
13. Xiong, C, Xie C.Q., Zhang, L, Zhang, J., Xu, K., Fu, M., Thompson, WE, Yang, L.J., & Chen, YE. "Derivation of adipocytes from human embryonic stem cells." Stem Cell and Development Dec;14(6):671-675, 2005.
14. Yang, Lijun and Shiwu Li. "In Vitro Trans-differentiation of Adult Hepatic Stem Cells into Pancreatic Endocrine Hormone-Producing Cells." PNAS 99:12, 2002.
15. Yang L, Li S, Hatch H, Ahrens K, Cornelius JG, Petersen BE, Peck AB. "In vitro trans-differentiation of adult hepatic stem cells into pancreatic endocrine hormone-producing cells." Proc Natl Acad Sci USA, 99:8078–8083, 2002.

---

--top--

Back to the Journal of Undergraduate Research