The role of the vasculature niche on insulin-producing cells generated by transdifferentiation of adult human liver cells
Insulin-dependent diabetes is a multifactorial disorder that could be theoretically cured by functional pancreatic islets and insulin-producing cell (IPC) implantation. Regenerative medicine approaches include the potential for growing tissues and organs in the laboratory and transplanting them when the body cannot heal itself. However, several obstacles remain to be overcome in order to bring regenerative medicine approach for diabetes closer to its clinical implementation; the cells generated in vitro are typically of heterogenic and immature nature and the site of implantation should be readily vascularized for the implanted cells to survive in vivo. The present study addresses these two limitations by analyzing the effect of co-implanting IPCs with vasculature promoting cells in an accessible site such as subcutaneous. Secondly, it analyzes the effects of reconstituting the in vivo environment in vitro on the maturation and function of insulin-producing cells.
IPCs that are generated by the transdifferentiation of human liver cells are exposed to the paracrine effects of endothelial colony-forming cells (ECFCs) and human bone marrow mesenchymal stem cells (MSCs), which are the “building blocks” of the blood vessels. The role of the vasculature on IPC function is analyzed upon subcutaneous implantation in vivo in immune-deficient rodents. The paracrine effects of vasculature on IPC maturation are analyzed in culture.
Co-implantation of MSCs and ECFCs with IPCs led to doubling the survival rates and a threefold increase in insulin production, in vivo. ECFC and MSC co-culture as well as conditioned media of co-cultures resulted in a significant increased expression of pancreatic-specific genes and an increase in glucose-regulated insulin secretion, compared with IPCs alone. Mechanistically, we demonstrate that ECFC and MSC co-culture increases the expression of CTGF and ACTIVINβα, which play a key role in pancreatic differentiation.
Vasculature is an important player in generating regenerative medicine approaches for diabetes. Vasculature displays a paracrine effect on the maturation of insulin-producing cells and their survival upon implantation. The reconstitution of the in vivo niche is expected to promote the liver-to-pancreas transdifferentiation and bringing this cell therapy approach closer to its clinical implementation.
KeywordsVasculature Insulin-producing cells Transdifferentiation Bone marrow–derived mesenchymal stem cells Endothelial colony-forming cells Pancreatic transcription factors
Connective tissue growth factor
Endothelial colony-forming cells
Mesenchymal stem cells
Pancreatic transcription factors
Islet transplantation is a promising procedure for the restoration of glycemic control in type 1 diabetes patients [1, 2]. However, this approach is limited by the shortage of donor-derived organs, the low rates of islet allograft survival, and the resulting high number of islets that are required per procedure. Among other cell replacement therapy approaches for diabetics is the option to reprogram adult extra-endocrine pancreatic tissues into functional glucose-regulated insulin-producing cells (IPCs). We demonstrated that adenoviral-mediated ectopic expression of pancreatic transcription factors (pTFs) induces transdifferentiation of adult liver cells into IPCs, both in vivo in mice and in vitro in human tissue [3, 4, 5, 6, 7, 8, 9, 10, 11]. The transdifferentiation protocol resulted in the generation of a heterogeneous population of cells in which the capacity for insulin production was restricted to a subpopulation of the human liver cells in the culture [10, 12]. Moreover, the survival of the implanted cells in vivo is low, either under the kidney capsule or subcutaneously [8, 13].
Pancreatic islets are extremely vascularized structures that have high levels of oxygen consumption. Although they form only 1–2% of the pancreatic mass, pancreatic islets receive 5–10% of the pancreatic blood with capillary density, which is ten times more than the surrounding exocrine pancreas . Moreover, their proximity to the blood vessels plays an important role in the provision of non-nutritional signals to them, creating a vascular niche in which cross-talk between the endocrine β cells and endothelial cells is necessary to ensure proper pancreatic differentiation and β cell development and function [15, 16, 17, 18, 19, 20]. While endocrine pancreatic cells express vascular endothelial growth factor A (VEGF-A), which is a crucial factor in maintaining the microvasculature of the islet [21, 22, 23, 24], the vasculature produces several paracrine factors that modulate gene expression, proliferation, and the survival of β cells. Among these are thrombospondin-1 , transforming growth factor beta (TGFβ) , endothelin-1 , hepatocyte growth factor (HGF) , and connective tissue growth factor (CTGF) [29, 30].
Numerous reports indicate the key role that the endothelium has on pancreatic β cell differentiation and survival and the beneficial effect of the endothelial cells on the function of implanted islets [31, 32, 33, 34, 35]. In addition to the endothelial cells, intra-islet microvessels consist of one or more layers of perivascular cells [36, 37]. Studies have shown that the inclusion of human bone marrow mesenchymal stem cells (MSCs), which are able to differentiate into pericytes or smooth muscle-like cells when directly contacting endothelial cells, leads to increased vascularization  and survival of implanted pancreatic islets [39, 40].
Herein, we aim to analyze the effect of human bone marrow–derived MSCs and endothelial colony-forming cells (ECFCs) on the survival of IPCs, their function in vivo implantation, and the potential paracrine role of the MSCs and ECFCs on IPC maturation and function in vitro.
Human liver tissues and cultures
Liver tissues were used with the approval of the Sheba Medical Center Committee on Clinical Investigations (the institutional review board). All liver specimen donors or guardians on behalf of minors provided written informed consent for the collection of all samples and subsequent analyses. No donor organs were obtained from executed prisoners or other institutionalized persons. Tissue processing, isolation, and maintenance of cell cultures were described previously [10, 12].
ECFC and MSC cultures
Human bone marrow–derived mesenchymal stem cells (MSCs) were isolated and propagated as previously described . Cord-blood endothelial colony-forming cells (ECFCs) are a gift from Prof. Joyce Bischoff (Vascular Biology Program, Children’s Hospital Boston, and Harvard Medical School). Cells were isolated and phenotypically characterized as described [41, 42].
Induction of TD in vitro
Cells were plated in a low-glucose DMEM media (1 g/L glucose) supplemented with 2 mM l-glutamine, 10% serum, antibiotics, 10 mM nicotinamide, 20 ng/ml EGF, and 5 nM exendin-4. The cells were infected with 1000MOI Ad-CMV-PDX1 and 250MOI Ad-CMV-NEUROD1. After 48 h, the cells were harvested, counted, infected with 50MOI of Ad-CMV-MAFA, and re-plated as above. The infected cells were incubated for an additional 72 h before harvest.
Animal procedures were approved by the Institutional Animal Care and Use Committee of Sheba Medical Center.
SCID-beige mice (8–9 weeks old, 18–19 g) were housed in an air-conditioned environment, under a 12-h light/dark cycle, and handled according to institutional animal welfare regulations. Generation of vascularized implants by co-implantation of endothelial cells and mesenchymal stem cells was previously described [38, 43, 44]. Shortly, IPCs (106 cells) were mixed with ECFCs (106 cells) and MSCs (106 cells) in 200 μl of Matrigel (BD Biosciences, San Jose, CA) and injected subcutaneously into the mice. IPCs (106 cells) without ECFCs and MSCs served as a control.
Glucose tolerance test
Mice fasted for 6 h were injected intraperitoneally with glucose in saline at 2 mg/g body weight. Serum was collected for human C-peptide level analyses. The ultrasensitive Human C-Peptide ELISA kit (Mercodia, Uppsala) with 3% cross-reactivity to proinsulin but no cross-reactivity to mouse c-peptide was used according to the manufacturer’s instructions.
Histology and staining
The implants were removed, fixed in 4% formaldehyde, embedded in paraffin, and stained as described in , using anti-human HLA-A (abcam ab52922, 1:100), anti-insulin (Sigma, clone K36 AC10, 1:75), and anti-human CD31 (Dako Cytomation, Clone JC70A, M0823, 1:50 dilution) antibodies. Analysis of HLA-positive cells was done by counting cells in random five × 40 fields from each implant. Histological analysis of microvessel density (MVD) was performed as described previously ; microvessel density was assessed by counting CD31 positive red blood cell–filled lumens. Values reported for each experimental condition correspond to the average MVD value ± standard error of the mean (SEM) obtained from all the individual implants.
Immunofluorescence: antibodies used in this study included guinea pig anti-insulin (1:400; Dako) and mouse anti-glucagon (1:800; Abcam). Fluorescent images were taken on a Nikon C1 confocal microscope at an original magnification × 40 as described .
Cell culturing in the Transwell® system
IPCs were plated on the bottom of 12-well plates (105 cells/well). After the cell adherence to the surface (3–4 h), the Transwell® inserts pre-coated with fibronectin were added to each well. The inserts were plated with ECFCs (105 cells/insert) or MSCs (105 cells/insert) or ECFCs /MSCs (1:1 ratio, total 105 cells/insert) or with IPCs (105 cells/insert) as control. After 72 h of co-culturing, the inserts were removed and the IPCs in the bottom of the wells were analyzed.
ECFCs and MSCs were plated alone or at 1:1 ratio at 7.5 × 104/cm2 for 48 h in EBM2 media supplemented with 2 mM l-glutamine, 20% serum, and antibiotics. The conditioned media was collected, aliquoted, and frozen at − 20 °C. Heated conditioned media: the conditioned media was heated to 56 °C for 20 min in a water bath before supplemented to the cells.
Isolation of ECFCs and MSCs from the co-cultures
ECFCs and MSCs were plated in six-well plates alone or at 1:1 ration at 2 × 104/cm2. Human dermal fibroblasts, at the same density, were plated as a control. Five days later, cells were trypsinized and immune-separated using CD31 magnetic beads (Dynabeads®, Cat # 111.55D, Invitrogen) according to the manufacturer’s instructions.
Insulin secretion in vitro
RNA was prepared from cells using RNAeasy kit (Qiagen) and reverse-transcribed. Relative expression of indicated genes was measured by StepOne Real-Time PCR System using Fast SYBR Green Master Mix (Applied Biosystems). Relative quantitative analyses were performed according to the comparative CT method by using the arithmetic formula 2−(ΔΔCt). The cDNA levels were normalized to β-ACTIN and GAPDH as the reference as described previously [10, 12]. Human pancreas total RNA (ThermoFisher) was used as a positive control.
Statistical analyses were performed using two-sample Student’s t test assuming equal variances.
Formation of the de novo blood vessels promotes the survival and function of IPCs in vivo
To analyze the effect of de novo vascularization on the maturation and function of IPCs, we co-implanted them with human bone marrow–derived MSCs and human cord-blood ECFCs in severe combined immunodeficiency (SCID)-beige mice. IPCs were generated by transdifferentiation of adult human liver cells that were induced by transcription factors, as was previously described . MSCs and ECFCs were isolated and characterized [41, 42] (see also Additional file 1: Figure S1).
MSCs and ECFCs could affect the insulin production of the implanted IPCs by providing better oxygen and nutrient supplies and therefore promoting the survival of the cells. In addition, they could provide a preferred niche that supplies the implanted cells with growth factors needed for their maturation. To address the question of the beneficial effect of the blood vessels on the IPCs functionality, we established a controlled in vitro experimental system.
ECFC/MSC co-culture promoted the pancreatic β cell-like maturation of IPCs in vitro
Either ECFCs or MSCs slightly increased the pancreatic gene expression of the IPCs (Fig. 3b). However, only ECFC and MSC co-culture in the upper compartment resulted in a significant increase in glucose-regulated insulin secretion and a robust increase in the activation of pancreatic-specific genes, such as GCG, SST, and PAX4 expression (Fig. 3b, c). Thus, our data suggest that the combination of ECFCs and MSCs displays a specific synergistic paracrine effect on the maturation of IPCs and β cell-like function.
Co-culturing ECFCs and MSCs promotes the expression of growth factors
This initial insight suggests that co-culturing ECFCs with MSCs results in significant changes in the expression of the growth factors, cytokines, and other molecules that upon protein secretion could account for the paracrine effects of the conditioned media that are at low levels or absent in each individual type of cell when cultured alone.
The conditioned media of the ECFCs and MSCs promotes the pancreatic differentiation
The reprogramming of liver cells toward IPCs is a sequential and temporally controlled process that resembles pancreatic differentiation . While PDX1 and NEUROD1 activate the transdifferentiation process required for the initiation of the reprogramming process (days 1–3), MAFA promotes IPC maturation and is therefore required at later stages (days 3–6). We sought to analyze whether the “vasculature-derived conditioned media” is mainly efficient at the initial steps of pancreatic differentiation (days 1–3) or during IPC maturation (days 3–6). We either added the conditioned media from the co-cultured ECFCs and MSCs at day 1 together with PDX1 and NEUROD1 or on day 3 together with MAFA, or we added the conditioned media at both time points.
The greatest effect of the conditioned media on the activation of the pancreatic phenotype occurred when it was supplemented on the first day with the ectopic expression of PDX1 and NEUROD1 (Fig. 5c–e); insulin gene expression and PAX4 gene expression were significantly higher when the conditioned media was added on the first day. Glucose-induced insulin secretion was similarly elevated regardless of the stage of conditioned media supplementation (Fig. 5e).
These data suggest that the transdifferentiation process may benefit from the paracrine effects of ECFCs and MSCs during the initial stages of the process, although some promoting effects may also occur during the entire process (Fig. 5c–e).
The present study demonstrates the beneficial role the vasculature plays in promoting pancreatic β cell-like differentiation and the functioning of reprogrammed human liver cells. The mixture of ECFCs and MSCs improved the survival and insulin secretion of the transdifferentiated IPCs in vivo. Moreover, using an isolated in vitro model, we showed that ECFCs and MSCs display a bilateral effect toward each other, since only their co-culture displayed a synergistic, paracrine effect on the expression of pancreatic-specific genes and on the glucose-regulated insulin secretion of the IPCs. This paracrine effect was potentially mediated by the secreted heat-sensitive proteins and other compounds. Mechanistically, we demonstrated that the proximity of the ECFCs and MSCs to each other in the co-culture was obligatory for the increase of the expression of factors such as CTGF and ACTIVIN-A, known to play key roles in pancreatic differentiation.
Our approach included the co-implantation of MSCs and ECFCs with IPCs. A combination of several types of endothelial cells and mural cells has been previously used for islet transplantation, such as HUVEC and MSCs  and HUVEC and fibroblasts . Several advantages are associated with using ECFC and bone marrow–derived MSCs. First, both cell types can be isolated from the bone marrow of an adult donor. This enables the autologous transplantation of all three types of cells without immune rejection. Second, both cell types are highly proliferative cells in vitro, and their combination has been shown to form long-lasting, perfused microvascular networks that alleviate ischemia in models of hind limb ischemia and myocardial infarction . Lastly, these two cell types were shown to have reciprocal interactions in our study (Fig. 4). Previously, it was demonstrated that ECFCs modulated the regenerative potential of the MSCs by paracrine secretion of angiocrine factors, whereas the MSCs modulated the ECFCs’ immunogenicity .
The identity of the endothelial cells and mural cells may also have affected the combination of the secreted cytokines and growth factors found in the conditioned media and, as a result, affected the maturation of the IPCs. Our in vitro assays clearly indicated that co-culturing ECFCs with MSCs resulted in increasing the expression of the analyzed CTGF and ACTIVIN-A genes because the cells influenced each other. The specific paracrine role of CTGF and ACTIVIN-A in promoting the transdifferentiation process should be further investigated by direct supplementation of CTGF and ACTIVIN-A proteins and by blocking these factors’ activity using neutralizing antibodies. The combination of the secreted cytokines and growth factors from co-culturing ECFCs with MSCs promoted the transdifferentiation process, mainly at the initial step of reprogramming.
Different combinations of secreted cytokines and growth factors are expected to affect the reprogramming process in different ways: some may promote the maturation of IPCs, while others may inhibit the process. The secreted cytokines and growth factors in the conditioned media could promote the reprogramming process by affecting the epigenome or by activating specific pancreatic gene expression or induction of post-transcriptional modifications. Our data suggest that the promoting effect of the conditioned media was potentially attributable to the post-transcriptional stages of insulin production, since it had no effect on ectopic insulin promoter activity (see Additional file 2: Figure S2).
Taken together, our data suggest the importance of the vasculature niche on the therapeutic outcome of reprogrammed IPCs in the treatment of diabetes. The vasculature niche not only promotes implant survival in vivo, it also provides the signals required by the reprogrammed cells to differentiate toward functional, pancreatic insulin-producing cells. Indeed, the vasculature niche is expected to play a crucial role in many regenerative medicine approaches for degenerative diseases in general.
We thank Dr. Nira Bloom (Division of hematology, Sheba Medical Center) for her help in the isolation and characterization of MSCs. We thank Prof Yuval Dor and Tehila Dahan (Department of Developmental Biology and Cancer Research, The Hebrew University, Jerusalem, Israel) for their assistance in the immunofluorescent staining. We wish to thank members of the lab for fruitful discussions.
This study was financially supported by the research grant “Autologous cell replacement therapy for diabetic patients by liver cells transdifferentiation” (Dia-cure), P_37_794, POC-A. 1- A.1.1.4-E-2015 (2016 - 2020)/ SMIS 106897, and by Orgenesis LTD. SG is supported by Talpiot Medical Leadership Program of Chaim Sheba Medical Center, Israel.
Availability of data and materials
All data generated or analyzed during this study are included in this published article [and its supplementary information files].
IM, SF, and SG designed the experiments. FZ, GN, EM, GL, and OZ performed the experiments. IM, SF, SG, FZ, GN, EM, GL, OZ, MG, EM, SD, IP, and AB analyzed the data. IM, SF, and SG drafted the manuscript. All authors read and approved the final manuscript.
Ethics approval and consent to participate
Liver tissues were used with the approval of the Sheba Medical Center Committee on Clinical Investigations (the institutional review board). All liver specimen donors or guardians on behalf of minors provided written informed consent for the collection of all samples and subsequent analyses. No donor organs were obtained from executed prisoners or other institutionalized persons.
Animal procedures were approved by the Institutional Animal Care and Use Committee of Sheba Medical Center.
Consent for publication
SF is an employee and consultant at Orgenesis. IML is an employee at Orgenesis. The other authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
- 6.Meivar-Levy I, Sapir T, Berneman D, Weissbach T, Polak-Charcon S, Ravassard P, et al. Human liver cells expressing albumin and mesenchymal characteristics give rise to insulin-producing cells. J Transplant [Internet]. 2011;2011:252387.Google Scholar
- 8.Sapir T, Shternhall K, Meivar-Levy I, Blumenfeld T, Cohen H, Skutelsky E, et al. From the cover: cell-replacement therapy for diabetes: generating functional insulin-producing tissue from adult human liver cells. Proc Natl Acad Sci U S A. 2005;102(22):7964–9.PubMedPubMedCentralCrossRefGoogle Scholar
- 12.Cohen H, Barash H, Meivar-Levy I, Molakandov K, Ben-Shimon M, Gurevich M, et al. The Wnt/beta-catenin pathway determines the predisposition and efficiency of liver-to-pancreas reprogramming. Hepatology (Baltimore, Md). 2018; Epub 2018/02/03.Google Scholar
- 36.Grainger SJ, Carrion B, Ceccarelli J, Putnam AJ. Stromal cell identity influences the in vivo functionality of engineered capillary networks formed by co-delivery of endothelial cells and stromal cells. Tissue Eng Part A. 2013;19(9–10):1209–22 Epub 2012/12/12.PubMedPubMedCentralCrossRefGoogle Scholar
- 39.Buitinga M, Janeczek Portalska K, Cornelissen DJ, Plass J, Hanegraaf M, Carlotti F, et al. Coculturing human islets with proangiogenic support cells to improve islet revascularization at the subcutaneous transplantation site. Tissue Eng Part A. 2016;22(3–4):375–85 Epub 2016/02/13.PubMedPubMedCentralCrossRefGoogle Scholar
- 43.Lin RZ, Moreno-Luna R, Li D, Jaminet SC, Greene AK, Melero-Martin JM. Human endothelial colony-forming cells serve as trophic mediators for mesenchymal stem cell engraftment via paracrine signaling. Proc Natl Acad Sci U S A. 2014;111(28):10137–42 Epub 2014/07/02.PubMedPubMedCentralCrossRefGoogle Scholar
- 44.Shafiee A, Patel J, Lee JS, Hutmacher DW, Fisk NM, Khosrotehrani K. Mesenchymal stem/stromal cells enhance engraftment, vasculogenic and pro-angiogenic activities of endothelial colony forming cells in immunocompetent hosts. Sci Rep. 2017;7(1):13558 Epub 2017/10/21.PubMedPubMedCentralCrossRefGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.