| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online May 31, 2005 as doi:10.1096/fj.04-3024fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,1
* Institute of Bioengineering, University Miguel Hernandez, San Juan, Alicante, Spain; and
Department of Surgery, National University of Singapore, Singapore
1Correspondence: Instituto de Bioingenieria, Universidad Miguel Hernandez, 03550-San Juan, Alicante, Spain. E-mail: bernat.soria{at}umh.es
SPECIFIC AIMS
Embryonic stem cells (ESCs) can differentiate in vitro into a variety of cell lineages, including insulin-producing cells. Pancreatic ß-cells derived from the foregut endoderm, as well as certain embryonic neurons derived from the neuroectoderm, display the ability to express insulin genes. In this study, we asked whether ESCs bioengineered in vitro to express insulin may derive from the ectoderm.
PRINCIPAL FINDINGS
1. Differentiation and characterization of ectodermal committed R1 cells
The first publication from our laboratory demonstrated that R1 mouse ESCs can differentiate in vitro via embryoid bodies (EBs) to insulin-containing cells. The insulin-positive cells were purified using an antibiotic selection system. However, this protocol had a high degree of heterogeneity, presenting very low amounts of insulin in 99% of the selected clones. It has been reported that insulin-producing cells derived from ectoderm also contain very low amounts of the hormone and display specific regulatory mechanisms differing from insulin-positive cells derived from the endoderm. Therefore, we considered the possibility that the main part of the insulin-positive clones selected using transgenes (including the insulin promoter) and EBs could derive from mixtures of ectoderm/endoderm in which the first pathway predominates. However, this question is difficult to address in protocols working via EBs because it is technically more complicated than the isolation of pure ecto- or endodermal-derived cell populations.
This problem can be overcome in cell cultures committed to more restricted differentiation pathways, such as ectoderm. This can be achieved by keeping ESCs in monolayer culture in the absence of leukemia inhibitory factor (LIF), as described in previous reports. Figure 1
shows the pattern of gene expression of clonal R1 undifferentiated ESCs (selected by resistance to hygromycin), EBs derived from these cells, and R1 monolayers in the absence of LIF. R1-ESCs presented a prominent expression of Oct3/4 gene. The mRNAs for glial fibrillary acidic protein (GFAP), neurofilament-200 (N-200) (markers of ectoderm) (Fig. 1A
), myelin binding protein (MBP), and tyrosine hydroxylase (TH) (neuronal markers) (Fig. 1B
) were barely detected in R1 undifferentiated cells. Glutamine synthetase (GS) (neuronal marker) and Otx2 (marker of neuronal precursors) mRNAs were expressed in R1 ESCs (Fig. 1B
). On the contrary, mRNAs for brachyury, mesoderm marker and 
-fetoprotein (AFP), FoxA2, FoxA3, GATA4, and GATA5, endoderm markers were not detected (Fig. 1A
).
|
When R1 ESCs cells were cultured in EBs for 20 days, a discrete down-regulation of Oct3/4 (Fig. 1A
) and SSEA-1 was observed, which was confirmed as well by immunocytochemistry (data not shown). This was accompanied by an increased and consistent expression of GFAP, N-200 (EBs from 1–2 days), brachyury (EBs from 5–7 days), and endoderm markers (EBs from 7–15 days) (Fig. 1A
). Finally, R1 cells cultured in monolayer in the absence of LIF for the same period (20 days) displayed only expression of ectoderm markers, such as GFAP and N-200 (Fig. 1A
). To study further the R1-ectoderm committed cells, we analyzed other genes present in pancreatic ß-cells but also present in ectoderm, such as Pdx-1, Isl-1, Pax6, and Ngn3. The expression of all these genes, with the exception of Ngn3, was increased in R1 cells cultured under these particular monolayer conditions (Fig. 1B
). Cells forced to grow in monolayer in the absence of LIF increased or maintained the expression level for Otx-2, TH, GS, and MBP, and induced AChE (Fig. 1B
).
To confirm the absence of endoderm-derived cells in the monolayer conditions, we performed another set of experiments using R1 cells transfected with AFP promoter directing the expression of enhanced green fluorescent protein (EGFP). Undifferentiated R1 ESCs did not express EGFP, but EB growth induced fluorescence, detected by day 8 in a polar region of the EB (not shown). This localization persisted for 30 days. However, a parallel cell culture maintained in monolayer in the absence of LIF did not show expression of EGFP for 30 days.
2. Induction of the insulin gene in ectodermal committed R1 cells
Insulin-producing cells derived from R1 ESCs have been isolated using a gating system, allowing a cell selection based on the resistance to the marker gene expression profile analyzed by RT-PCR and proliferation events did not change significantly in R1 neomycin-resistant cells (Fig. 2
A-0) compared with the clonal control cells, except for morphology (Fig. 2B
-0) and a lower expression of Oct3/4 and SSEA-1. The Ca2+ response to different extracellular factors was typical of ectodermal cells.
|
No expression of the insulin gene was detected under these particular culture conditions, indicating that additional extracellular factors may be necessary to induce this gene. Therefore, we analyzed the possibility of inducing insulin gene expression by adding different media to the neomycin-selected subpopulation. Astrocyte-conditioned medium, ITSF medium, and pancreatic buds conditioned medium maintained a similar pattern of gene expression for Pdx1 and Isl-1 and Pax6 down-regulation. Insulin and nestin expression was not detected, nor were endodermal markers, i.e., AFP (Fig. 2A
). Variable changes in cell morphology were observed.
Incubation of monolayers with N2B27 medium supplemented with nicotinamide displayed modest coinduction of insulin and nestin genes (Fig. 2A
). Cell morphology showed round cells with elongations forming attached colonies (Fig. 2B
). Finally, cells incubated in the presence of INS-1 conditioned medium displayed significant insulin gene expression as well as Pdx-1, Isl-1, and the neuroectodermal markers nestin (Fig. 2A
), N-200, GFAP, AChE, MBP, and GS. The amount of insulin detected by radioimmunoassay in these cells was 0.5 ng/mg protein. Only insulin II expression was detected under these particular culture conditions (not shown) with no expression of endoderm markers (i.e., AFP). Analysis of cell morphology showed cytoplasmic elongations that resemble neuron-like structures (Fig. 2B
).
CONCLUSIONS AND SIGNIFICANCE
Our findings show that insulin gene (insulin II) is capable of being expressed in committed cells that express only ectodermal markers, particularly nestin, and in which the early endodermal markers were not detected. This suggests the possibility of an ectodermal differentiation pathway in bioengineering protocols used to obtain insulin-producing cells.
Insulin-positive tissues derived from ectoderm and endoderm are not exactly identical, displaying key differences in the mechanisms involved in regulating the gene itself, the amount of hormone synthesized and secreted, and the physiological role of the end-products.
We propose two alternative strategies for future research (Fig. 3
). One of these considers that the ectodermal insulin-producing cells are easily obtained in vitro and could be manipulated to obtain adequate amounts of mature hormone and improved secretion in response to nutrients. The possibility of bioengineering nestin-positive cells in order to obtain ectodermal populations capable of mimicking mature ß-cell function deserves further exploration.
|
Alternatively, the possibility to obtain endoderm-rich cell cultures from which ß-cell derives has to be strongly considered.
FOOTNOTES
To read the full text of this article, go to http://fasebj.org/cgi/doi/10.1096/fj.04-3024fje;
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
