Methotrexate and Valproic Acid Affect Early Neurogenesis of Human Amniotic Fluid Stem Cells from Myelomeningocele

Abstract

Myelomeningocele (MMC) is a severe type of neural tube defect (NTD), in which the backbone and spinal canal do not close completely during early embryonic development. This condition results in serious morbidity and increased mortality after birth. Folic acid significantly reduces, and conversely, folate antagonist methotrexate (MTX) and valproic acid (VPA) increase the occurrence of NTDs, including MMC. How these pharmacological agents exactly influence the early neurulation process is still largely unclear. Here, we characterized human amniotic fluid-derived stem cells (AFSCs) from prenatally diagnosed MMC and observed an effect of MTX and VPA administration on the early neural differentiation process. We found that MMC-derived AFSCs highly expressed early neural and radial glial genes that were negatively affected by MTX and VPA exposure. In conclusion, we setup a human cell model of MMC to study early neurogenesis and for drug screening purposes. We also proposed the detection of early neural gene expression in AFSCs as an additional MMC diagnostic tool.

Figure 1

Morphological and fluorescence activated cell-sorting (FACS) analysis of MMC-AFSCs and healthy AFSCs. (a) Microscopy images from MMC-AFSCs and healthy AFSCs before neural induction process. Scale bar = 100 μm. (b) Representative flow cytometry plots and histograms from MMC-AFSCs and healthy AFSCs for the expression of mesenchymal markers CD44, CD73, CD90, CD117, and CD105. (c) Cell surface expression of mesenchymal markers CD44, CD73, CD90, CD117, and CD105 determined in MMC-AFSC and healthy AFSC populations by flow cytometry analysis, shown as percentage. N = 4 and values are indicated as mean ± SD. NS, not significant.

Figure 2

Gene expression analysis of MMC-AFSCs and healthy AFSCs. (a) Expression profile of pluripotency genes OCT4, NANOG, early neural gene PAX6, mesoendodermal genes PDGFRβ and GATA4, myogenic transcription factor PAX7, osteochondrogenic genes SOX9 and RUNX2, and COL2A1 in MMC-AFSCs and healthy AFSCs by RT-qPCR analysis. Data were represented as relative fold expression, normalized for the housekeeping gene GAPDH; NS, not significant. N = 4 and values are indicated as mean ± SD. (b) Expression profile of early neural and radial glial genes SOX2, NESTIN, BLBP, and PAX3 in MMC-AFSCs and healthy AFSCs by RT-qPCR analysis. Data were represented as relative fold expression, normalized for the housekeeping gene GAPDH; N = 4 and values are indicated as mean ± SD. ^∗∗P < 0.01 Healthy versus MMC (NESTIN and PAX3); ^∗∗∗∗P < 0.0001 Healthy versus MMC (SOX2 and BLBP).

Figure 3

Early neural differentiation potential of MMC-AFSCs and healthy AFSCs. (a, b) IF analysis of MMC-AFSCs and healthy AFSCs at day 15 of neural induction, exposed or not with MTX and VPA, showing positive cells for early neural markers SOX2 (green) and NESTIN (red). Representative images of healthy AFSCs (clone number 1) and MMC-AFSCs (clone number 5) were shown for all healthy and MMC lines. Nuclei were counterstained with Hoechst (blue). N = 4, scale bar = 100 μm. (c) Enumeration of SOX2⁺ cells in neural-derived MMC-AFSCs and healthy AFSCs at day 15 of neural induction, exposed or not with MTX and VPA, shown as percentage. N = 4 and values are indicated as mean ± SD. ^∗∗∗∗P < 0.0001 Healthy versus Healthy + MTX, Healthy versus Healthy + VPA, Healthy + MTX versus Healthy + VPA, Healthy versus MMC, MMC versus MMC + MTX, MMC versus MMC + VPA, Healthy + MTX versus MMC + MTX, and Healthy + VPA versus MMC+VPA. (d) Enumeration of NESTIN⁺ cells in neural-derived MMC-AFSCs and healthy AFSCs at day 15 of neural induction, exposed or not with MTX and VPA, shown as percentage. N = 4 and values are indicated as mean ± SD. NS, not significant. ^∗P < 0.05 Healthy + VPA versus MMC + VPA; ^∗∗P < 0.01 Healthy versus Healthy + MTX; ^∗∗∗P < 0.001 Healthy + MTX versus MMC+MTX, Healthy + MTX versus Healthy + VPA, and MMC + MTX versus MMC + VPA; and ^∗∗∗∗P < 0.0001 Healthy versus Healthy + VPA, MMC versus MMC + MTX, and MMC versus MMC + VPA.

Figure 4

Proliferative and immature neural differentiation potential of MMC-AFSCs and healthy AFSCs. (a, b) IF analysis of MMC-AFSCs and healthy AFSCs at day 15 of neural induction, exposed or not with MTX and VPA, showing positive cells for Ki67 (green) and the immature neural differentiation marker βIII-tubulin (red). Representative images of healthy AFSCs (clone number 1) and MMC-AFSCs (clone number 5) were shown for all healthy and MMC lines. Nuclei were counterstained with Hoechst (blue). N = 4, scale bar = 100 μm. (c) Enumeration of βIII-tubulin⁺ cells in neural-derived MMC-AFSCs and healthy AFSCs at day 15 of neural induction, exposed or not with MTX and VA, shown as percentage. N = 4 and values are indicated as mean ± SD. NS, not significant. ^∗P < 0.05 Healthy + VPA versus MMC + VPA; ^∗∗∗∗P < 0.0001 Healthy versus MMC, MMC versus MMC + MTX, and MMC versus MMC + VPA. (d) Enumeration of Ki67⁺ cells in neural-derived MMC-AFSCs and healthy AFSCs at day 15 of neural induction, exposed or not with MTX and VPA, shown as percentage. N = 4 and values are indicated as mean ± SD. NS, not significant. ^∗∗∗∗P < 0.0001 Healthy versus MMC, Healthy versus Healthy + MTX, Healthy versus Healthy + VPA, MMC versus MMC + MTX, and MMC versus MMC + VPA.

Figure 5

Early radial glial differentiation of MMC-AFSCs and healthy AFSCs. (a, b) IF analysis of MMC-AFSCs and healthy AFSCs and at day 15 of neural induction, exposed or not with MTX and VPA, showing positive cells for the early radial glial marker BLBP (green) and the neural crest stem cell migration marker PAX3 (red). Representative images of healthy AFSCs (clone number 1) and MMC-AFSCS (clone number 5) were shown for all healthy and MMC lines. Nuclei were counterstained with Hoechst (blue). N = 4, scale bar = 100 μm. (c) Enumeration of BLBP⁺ cells in neural-derived MMC-AFSCs and healthy AFSCs at day 15 of neural induction, exposed or not with MTX and VPA, shown as percentage. N = 4 and values are indicated as mean ± SD. NS, not significant. ^∗∗∗∗P < 0.0001 Healthy versus Healthy + MTX, Healthy versus Healthy + VPA, Healthy versus MMC, MMC versus MMC + MTX, and MMC versus MMC + VPA. (d) Enumeration of PAX3⁺ cells in neural-derived healthy AFSCs and MMC-AFSCs at day 15 of neural induction, exposed or not with MTX and VPA, shown as percentage. N = 4 and values are indicated as mean ± SD. NS, not significant. ^∗∗P < 0.01 Healthy + VPA versus MMC + VPA; ^∗∗∗∗P < 0.0001 Healthy versus MMC, MMC versus MMC + VPA, and MMC + MTX versus MMC + VPA.