Limits...
FGF4 and retinoic acid direct differentiation of hESCs into PDX1-expressing foregut endoderm in a time- and concentration-dependent manner.

Johannesson M, Ståhlberg A, Ameri J, Sand FW, Norrman K, Semb H - PLoS ONE (2009)

Bottom Line: Finally, further characterization of the PDX1(+) cells suggests that they represent foregut endoderm not yet committed to pancreatic, posterior stomach, or duodenal endoderm.In conclusion, we show that RA and FGF4 jointly direct differentiation of PDX1(+) foregut endoderm in a robust and efficient manner.Part of RA's activity is mediated by FGF signaling.

View Article: PubMed Central - PubMed

Affiliation: Lund Center for Stem Cell Biology and Cell Therapy, Lund University, Lund, Sweden.

ABSTRACT

Background: Retinoic acid (RA) and fibroblast growth factor 4 (FGF4) signaling control endoderm patterning and pancreas induction/expansion. Based on these findings, RA and FGFs, excluding FGF4, have frequently been used in differentiation protocols to direct differentiation of hESCs into endodermal and pancreatic cell types. In vivo, these signaling pathways act in a temporal and concentration-dependent manner. However, in vitro, the underlying basis for the time of addition of growth and differentiation factors (GDFs), including RA and FGFs, as well as the concentration is lacking. Thus, in order to develop robust and reliable differentiation protocols of ESCs into mature pancreatic cell types, including insulin-producing beta cells, it will be important to mechanistically understand each specification step. This includes differentiation of mesendoderm/definitive endoderm into foregut endoderm--the origin of pancreatic endoderm.

Methodology/principal findings: Here, we provide data on the individual and combinatorial role of RA and FGF4 in directing differentiation of ActivinA (AA)-induced hESCs into PDX1-expressing cells. FGF4's ability to affect endoderm patterning and specification in vitro has so far not been tested. By testing out the optimal concentration and timing of addition of FGF4 and RA, we present a robust differentiation protocol that on average generates 32% PDX1(+) cells. Furthermore, we show that RA is required for converting AA-induced hESCs into PDX1(+) cells, and that part of the underlying mechanism involves FGF receptor signaling. Finally, further characterization of the PDX1(+) cells suggests that they represent foregut endoderm not yet committed to pancreatic, posterior stomach, or duodenal endoderm.

Conclusion/significance: In conclusion, we show that RA and FGF4 jointly direct differentiation of PDX1(+) foregut endoderm in a robust and efficient manner. RA signaling mediated by the early induction of RARbeta through AA/Wnt3a is required for PDX1 expression. Part of RA's activity is mediated by FGF signaling.

Show MeSH

Related in: MedlinePlus

Analysis of expression and role of FGF and RA receptors during the FGF4/RA differentiation protocol.(A) Relative mRNA expression of PDX1, fibroblast growth factor receptors 1 and 2 (FGFR1, FGFR2), and retinoic acid receptor beta (RARβ) on different days (D1–D16) during the induction. RA, retinoic acid; F4, fibroblast growth factor 4. Cells were always pretreated with Activin-A (AA)-induction for three days. (B) Pharmacological inhibition of FGF (SU)- and RA (AGN)-receptor signaling on day nine of the FGF4/RA differentiation protocol. SU, SU5402 at 5 ng/mL; AGN, AGN193109 at 1 µM; 0 = no addition of any substance; AA, activin A; AA0, AA the first three days followed by no addition of FGF4/RA. Data is based on three separate experiments with cell line Hues-3 (subclone 52).
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC2651644&req=5

pone-0004794-g005: Analysis of expression and role of FGF and RA receptors during the FGF4/RA differentiation protocol.(A) Relative mRNA expression of PDX1, fibroblast growth factor receptors 1 and 2 (FGFR1, FGFR2), and retinoic acid receptor beta (RARβ) on different days (D1–D16) during the induction. RA, retinoic acid; F4, fibroblast growth factor 4. Cells were always pretreated with Activin-A (AA)-induction for three days. (B) Pharmacological inhibition of FGF (SU)- and RA (AGN)-receptor signaling on day nine of the FGF4/RA differentiation protocol. SU, SU5402 at 5 ng/mL; AGN, AGN193109 at 1 µM; 0 = no addition of any substance; AA, activin A; AA0, AA the first three days followed by no addition of FGF4/RA. Data is based on three separate experiments with cell line Hues-3 (subclone 52).

Mentions: The pivotal role of RA and FGF4 in endoderm and pancreas development led us to investigate their role in directing differentiation of putative DE, obtained through the frequently used three-day Activin A/Wnt3a induction protocol [33] (AA-induction) (Fig. S1 and S2), into PDX1+ posterior foregut endoderm. So far, FGF4 has not been tested for its activity in patterning ESC-derived gut endoderm. In the absence of RA, FGF4 was unable to induce PDX1 expression (data not shown). Since it was previously shown that RA promotes differentiation into PDX1+ cells when added four days after the AA-induction [7], we tested whether FGF4 synergized with RA in directing DE into PDX1+ cells. Indeed, PDX1 expression measured on day twelve increased when FGF4 was added directly after AA-induction and before the RA-treatment (Fig. 1B). Notably, FGF4 exhibited its effect on PDX1 expression in a concentration-dependent manner. Importantly, endogenous expression of FGF4 is only detected in undifferentiated cells and not at later time-points (Fig. S3A). To further optimize the protocol, the timing of RA addition was considered. In fact, the timing of RA addition has in most previous efforts been rather arbitrary, based on the fact that it should be added rather soon after the DE-induction. Logically, the timing of RA addition should be based on RARβ expression, which so far has not been determined. Therefore, we examined the timing of RARβ expression after AA-induction. Interestingly, RARβ was upregulated directly after the AA-induction on day four, and subsequently downregulated (day eight) in the absence of any exogenous growth and differentiation factor (GDF) (Fig. 1C and 5A; compare AA D4 with AA D8). Based on these findings we then tested various combinations of FGF4 and RA to achieve optimal induction of PDX1 expression during a twelve-day period (Fig. 1D). Moreover, PDX1 expression increased at day 12 when RA was added at day four compared to at day eight (Fig. S3B). Further optimisation of the protocol revealed that the highest PDX1-expression level was obtained when RA was kept throughout the whole protocol, i.e. for 13 days after the activin induction. Subtraction of RA at earlier time points (from day 10, 12 or 14) diminished the relative expression of PDX1 (Fig. S3C). Yet further prolonged treatment with RA and FGF4 still increases PDX1-expression, but at this point cells could start to deteriorate, probably due to high confluence. Notably, the marked increase in cell number, but lack of effect on relative PDX1 expression, upon addition of FGF4 (compare condition 3 with 2 in Fig. 1D) suggests that FGF4 primarily affect cell survival. Moreover, the cell viability assay AlamarBlue indicated that FGF4 promotes cell viability by reducing cytotoxic effects possibly exhibited by RA (Fig. S3D). Based on this observation we show that continuous treatment with RA and FGF4 (1.1 ng/ml) after the AA-induction resulted in efficient induction of PDX1 mRNA expression (∼25-fold increase in relative PDX1 mRNA expression on day 13; Fig. 1D). Immunofluorescence analysis was used to confirm that the observed increase in PDX1 mRNA expression was paralleled by a significant increase at the protein level (Fig. 1E). It should be noted that in cells that did not receive treatment with RA and FGF4, no PDX1-protein was detected. Efforts were made to passage the cells to new plates at this stage, but under currently used experimental conditions the cells failed to survive this treatment. The effect of RA and FGF4 was also evident by changes in cell morphology. Treatment with RA and FGF4 resulted in smaller cells that often were assembled in small cell clusters (Fig. 2).


FGF4 and retinoic acid direct differentiation of hESCs into PDX1-expressing foregut endoderm in a time- and concentration-dependent manner.

Johannesson M, Ståhlberg A, Ameri J, Sand FW, Norrman K, Semb H - PLoS ONE (2009)

Analysis of expression and role of FGF and RA receptors during the FGF4/RA differentiation protocol.(A) Relative mRNA expression of PDX1, fibroblast growth factor receptors 1 and 2 (FGFR1, FGFR2), and retinoic acid receptor beta (RARβ) on different days (D1–D16) during the induction. RA, retinoic acid; F4, fibroblast growth factor 4. Cells were always pretreated with Activin-A (AA)-induction for three days. (B) Pharmacological inhibition of FGF (SU)- and RA (AGN)-receptor signaling on day nine of the FGF4/RA differentiation protocol. SU, SU5402 at 5 ng/mL; AGN, AGN193109 at 1 µM; 0 = no addition of any substance; AA, activin A; AA0, AA the first three days followed by no addition of FGF4/RA. Data is based on three separate experiments with cell line Hues-3 (subclone 52).
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC2651644&req=5

pone-0004794-g005: Analysis of expression and role of FGF and RA receptors during the FGF4/RA differentiation protocol.(A) Relative mRNA expression of PDX1, fibroblast growth factor receptors 1 and 2 (FGFR1, FGFR2), and retinoic acid receptor beta (RARβ) on different days (D1–D16) during the induction. RA, retinoic acid; F4, fibroblast growth factor 4. Cells were always pretreated with Activin-A (AA)-induction for three days. (B) Pharmacological inhibition of FGF (SU)- and RA (AGN)-receptor signaling on day nine of the FGF4/RA differentiation protocol. SU, SU5402 at 5 ng/mL; AGN, AGN193109 at 1 µM; 0 = no addition of any substance; AA, activin A; AA0, AA the first three days followed by no addition of FGF4/RA. Data is based on three separate experiments with cell line Hues-3 (subclone 52).
Mentions: The pivotal role of RA and FGF4 in endoderm and pancreas development led us to investigate their role in directing differentiation of putative DE, obtained through the frequently used three-day Activin A/Wnt3a induction protocol [33] (AA-induction) (Fig. S1 and S2), into PDX1+ posterior foregut endoderm. So far, FGF4 has not been tested for its activity in patterning ESC-derived gut endoderm. In the absence of RA, FGF4 was unable to induce PDX1 expression (data not shown). Since it was previously shown that RA promotes differentiation into PDX1+ cells when added four days after the AA-induction [7], we tested whether FGF4 synergized with RA in directing DE into PDX1+ cells. Indeed, PDX1 expression measured on day twelve increased when FGF4 was added directly after AA-induction and before the RA-treatment (Fig. 1B). Notably, FGF4 exhibited its effect on PDX1 expression in a concentration-dependent manner. Importantly, endogenous expression of FGF4 is only detected in undifferentiated cells and not at later time-points (Fig. S3A). To further optimize the protocol, the timing of RA addition was considered. In fact, the timing of RA addition has in most previous efforts been rather arbitrary, based on the fact that it should be added rather soon after the DE-induction. Logically, the timing of RA addition should be based on RARβ expression, which so far has not been determined. Therefore, we examined the timing of RARβ expression after AA-induction. Interestingly, RARβ was upregulated directly after the AA-induction on day four, and subsequently downregulated (day eight) in the absence of any exogenous growth and differentiation factor (GDF) (Fig. 1C and 5A; compare AA D4 with AA D8). Based on these findings we then tested various combinations of FGF4 and RA to achieve optimal induction of PDX1 expression during a twelve-day period (Fig. 1D). Moreover, PDX1 expression increased at day 12 when RA was added at day four compared to at day eight (Fig. S3B). Further optimisation of the protocol revealed that the highest PDX1-expression level was obtained when RA was kept throughout the whole protocol, i.e. for 13 days after the activin induction. Subtraction of RA at earlier time points (from day 10, 12 or 14) diminished the relative expression of PDX1 (Fig. S3C). Yet further prolonged treatment with RA and FGF4 still increases PDX1-expression, but at this point cells could start to deteriorate, probably due to high confluence. Notably, the marked increase in cell number, but lack of effect on relative PDX1 expression, upon addition of FGF4 (compare condition 3 with 2 in Fig. 1D) suggests that FGF4 primarily affect cell survival. Moreover, the cell viability assay AlamarBlue indicated that FGF4 promotes cell viability by reducing cytotoxic effects possibly exhibited by RA (Fig. S3D). Based on this observation we show that continuous treatment with RA and FGF4 (1.1 ng/ml) after the AA-induction resulted in efficient induction of PDX1 mRNA expression (∼25-fold increase in relative PDX1 mRNA expression on day 13; Fig. 1D). Immunofluorescence analysis was used to confirm that the observed increase in PDX1 mRNA expression was paralleled by a significant increase at the protein level (Fig. 1E). It should be noted that in cells that did not receive treatment with RA and FGF4, no PDX1-protein was detected. Efforts were made to passage the cells to new plates at this stage, but under currently used experimental conditions the cells failed to survive this treatment. The effect of RA and FGF4 was also evident by changes in cell morphology. Treatment with RA and FGF4 resulted in smaller cells that often were assembled in small cell clusters (Fig. 2).

Bottom Line: Finally, further characterization of the PDX1(+) cells suggests that they represent foregut endoderm not yet committed to pancreatic, posterior stomach, or duodenal endoderm.In conclusion, we show that RA and FGF4 jointly direct differentiation of PDX1(+) foregut endoderm in a robust and efficient manner.Part of RA's activity is mediated by FGF signaling.

View Article: PubMed Central - PubMed

Affiliation: Lund Center for Stem Cell Biology and Cell Therapy, Lund University, Lund, Sweden.

ABSTRACT

Background: Retinoic acid (RA) and fibroblast growth factor 4 (FGF4) signaling control endoderm patterning and pancreas induction/expansion. Based on these findings, RA and FGFs, excluding FGF4, have frequently been used in differentiation protocols to direct differentiation of hESCs into endodermal and pancreatic cell types. In vivo, these signaling pathways act in a temporal and concentration-dependent manner. However, in vitro, the underlying basis for the time of addition of growth and differentiation factors (GDFs), including RA and FGFs, as well as the concentration is lacking. Thus, in order to develop robust and reliable differentiation protocols of ESCs into mature pancreatic cell types, including insulin-producing beta cells, it will be important to mechanistically understand each specification step. This includes differentiation of mesendoderm/definitive endoderm into foregut endoderm--the origin of pancreatic endoderm.

Methodology/principal findings: Here, we provide data on the individual and combinatorial role of RA and FGF4 in directing differentiation of ActivinA (AA)-induced hESCs into PDX1-expressing cells. FGF4's ability to affect endoderm patterning and specification in vitro has so far not been tested. By testing out the optimal concentration and timing of addition of FGF4 and RA, we present a robust differentiation protocol that on average generates 32% PDX1(+) cells. Furthermore, we show that RA is required for converting AA-induced hESCs into PDX1(+) cells, and that part of the underlying mechanism involves FGF receptor signaling. Finally, further characterization of the PDX1(+) cells suggests that they represent foregut endoderm not yet committed to pancreatic, posterior stomach, or duodenal endoderm.

Conclusion/significance: In conclusion, we show that RA and FGF4 jointly direct differentiation of PDX1(+) foregut endoderm in a robust and efficient manner. RA signaling mediated by the early induction of RARbeta through AA/Wnt3a is required for PDX1 expression. Part of RA's activity is mediated by FGF signaling.

Show MeSH
Related in: MedlinePlus