High Glycolytic Activity Enhances Stem Cell Reprogramming of Fahd1-KO Mouse Embryonic Fibroblasts

Abstract

Mitochondria play a key role in metabolic transitions involved in the reprogramming of somatic cells into induced pluripotent stem cells (iPSCs), but the underlying molecular mechanisms remain largely unexplored. To obtain new insight into the mechanisms of cellular reprogramming, we studied the role of FAH domain-containing protein 1 (FAHD1) in the reprogramming of murine embryonic fibroblasts (MEFs) into iPSCs and their subsequent differentiation into neuronal cells. MEFs from wild type (WT) and Fahd1-knock-out (KO) mice were reprogrammed into iPSCs and characterized for alterations in metabolic parameters and the expression of marker genes indicating mitochondrial biogenesis. Fahd1-KO MEFs showed a higher reprogramming efficiency accompanied by a significant increase in glycolytic activity as compared to WT. We also observed a strong increase of mitochondrial DNA copy number and expression of biogenesis marker genes in Fahd1-KO iPSCs relative to WT. Neuronal differentiation of iPSCs was accompanied by increased expression of mitochondrial biogenesis genes in both WT and Fahd1-KO neurons with higher expression in Fahd1-KO neurons. Together these observations establish a role of FAHD1 as a potential negative regulator of reprogramming and add additional insight into mechanisms by which FAHD1 modulates mitochondrial functions.

Keywords: FAHD1; glycolytic activity; iPSCs; mitochondria; neuronal differentiation; oxidative phosphorylation; reprogramming.

Figure 1

Increased iPSC-like colony formation in Fahd1-KO MEFs during reprogramming. (A) Merged phase-contrast and fluorescent images showing the difference in the fluorescent colony formation during the course of reprogramming from day 3 (D3) to D7 of WT and Fahd1-KO MEFs. (B) Merged tile scan images of phase contrast and fluorescent images covering most of the surface of a well of WT and Fahd1-KO reprogrammed MEFs, respectively, at D7. (C) Counting the fluorescent and non-fluorescent iPSC-like colonies at 3, 4, 5, and 7 days in vitro shows a clear and significant increase in the number of colonies formatted in Fahd1-KO as compared to WT, * p < 0.05, n = 3. (D) Growth curve comparison between WT and Fahd1-KO MEFs for 37 days in vitro. No significant difference was observed. MEFs: mouse embryonic fibroblasts, D: day in vitro, WT: wild type, KO: knock-out, PDL: cumulative population doubling level. Scale bars, 200 µm.

Figure 2

Germ layer and neural differentiation showed similar cell fate between WT and Fahd1-KO iPSCs. (A) Phase contrast imaging showing typical iPSC-like round colonies when cultured on inactivated MEFs. (B) Immunofluorescence showing the expression of the pluripotency markers Oct4 and Sox2 in both WT and KO iPSCs. Cell nuclei were stained with DAPI. (C) After germ layer differentiation, immunofluorescence on the differentiated cells showed the expression of the endodermal marker AFP, the mesodermal marker SMA, and the ectodermal marker Tuj1. (D) Neural differentiation showed typical neural rosette formation both in WT and Fahd1-KO, as it is shown in the phase contrast images. (E) Relative mRNA expression by qRT-PCR of the pluripotency marker Nanog, and the neural progenitor markers Pax6 and Nestin in WT and Fahd1-KO MEFs, iPSCs, and neural progenitors. The results showed a downregulation of the pluripotency and an upregulation of neural markers similarly in both WT and Fahd1-KO. MEFs: mouse embryonic fibroblasts, iPSCs: induced pluripotent stem cells, WT: wild type, KO: knock-out, ND: not detected. Scale bars, 100 µm.

Figure 3

Increased glycolysis in Fahd1-KO MEFs. (A) ECAR measurements in mpH/min/cell using Seahorse XFp analyzer and following glycolysis stress test protocol showed a highly significant increase in basal glycolysis, glycolytic capacity, non-glycolytic acidification, and glycolytic reserve in Fahd1-KO MEFs as compared to WT, *** p < 0.001, n = 6. (B) OCR measurements in pmol/min/cell using Seahorse XFp analyzer and following MitoStress test protocol showed a trend to decreased OCR in Fahd1-KO vs. WT MEFs, whereas maximal respiration and spare respiratory capacity parameters were significantly decreased in Fahd1-KO, * p < 0.05, ** p < 0.01, n = 3. (C) Average fold-change of mtDNA/nDNA ratio showing the expression of the mt genes 16S rRNA and ND1 as normalized to two housekeeping genes Gapdh and Hk2. The DNA copy number of these genes was determined in WT and Fahd1-KO MEFs. No significant difference was observed, n = 6. MEFs: mouse embryonic fibroblasts, iPSCs: induced pluripotent stem cells, WT: wild type, KO: knock-out, ECAR: extra cellular acidification rate, OCR: oxygen consumption rate, mt: mitochondrial, n: nuclear.

Figure 4

Mitochondrial remodeling during reprogramming is affected by Fahd1 deficiency. (A) ECAR measurements in mpH/min using the Seahorse XFp analyzer and following glycolysis stress test protocol showed that basal glycolysis, as well as glycolytic capacity, were weakly but significantly increased in Fahd1-KO iPSCs as compared to WT, * p < 0.05, n = 3. (B) OCR measurements in pmol/min using the Seahorse XFp analyzer and following MitoStress test protocol showed a significant increase of OCR for almost all parameters in Fahd1-KO iPSCs, *** p < 0.001, n = 3. (C) Average fold-change of mtDNA/nDNA ratio showing the relative concentration of mitochondrial vs. nuclear DNA, determined by qPCR amplification using primers for the mt genes coding for 16S rRNA and ND1, respectively, which were separately normalized to nuclear DNA amplified using primers for nuclear genes Gapdh and Hk2. qPCR analysis was performed in WT and Fahd1-KO iPSCs, as indicated, * p < 0.05, ** p < 0.01, *** p < 0.001, n = 3. (D) Relative mRNA expression of markers known to be involved in mt biogenesis, Pparg, Ppara, Slit1, Creb1, Nrf1, and Junk1. Their mRNA expression was evaluated in WT and Fahd1-KO iPSCs. Pparg and Ppara mRNA expression was significantly increased in Fahd1-KO iPSCs compared to WT, * p < 0.05, n = 3. (E) OCR normalized to mitochondrial copy number was slightly higher in WT vs. Fahd1-KO iPSCs, * p < 0.05, ** p < 0.01, *** p < 0.001, n = 3. iPSCs: induced pluripotent stem cells, WT: wild type, KO: knock-out, mt: mitochondrial, n: nuclear, OCR: oxygen consumption rate.

Figure 5

A high degree of mitochondrial biogenesis is maintained in Fahd1-KO iPSC-derived neurons. (A) Immunocytochemistry against the neuronal marker Tuj1 and the nuclei marker DAPI showed an efficient neuronal differentiation in both WT and Fahd1-KO iPSCs. (B–D) Relative mRNA expression of mt biogenesis markers, Pparg, Ppara, Slit1, Creb1, Nrf1, and Junk1. (B) Upregulation of mt biogenesis markers in iPSC-derived neurons relative to WT iPSCs. (C) Upregulation of mt biogenesis markers in Fahd1-KO iPSC-derived neurons relative to Fahd1-KO iPSCs. (D) Expression of mt biogenesis markers in neurons produced from Fahd1-KO and WT iPSCs. All the analyses were performed on iPSC-derived neurons after 20 days of differentiation, p < 0.05, ** p < 0.01, n = 3. iPSCs: induced pluripotent stem cells, WT: wild type, KO: knock-out.

Figure 6

Role of FAHD1-mediated oxaloacetate decarboxylation on TCA cycle flux. When FAHD1 is functional (left panel), excess levels of oxaloacetate (OAA) are converted to pyruvate and CO₂, keeping OAA concentration below the threshold required for efficient inhibition of succinate dehydrogenase (SDH), thus enabling sustained flux through the TCA cycle. When the FAHD1 gene is deleted, OAA may accumulate to an extent sufficient to inhibit SDH activity, resulting in the accumulation of succinate, reduced levels of fumarate, and reduced flux through the TCA cycle, together leading to an impaired mitochondrial function, the extent of which differs between cell types (see also main text) (Figure adapted from Etemad et al., Mechanisms of Ageing and Development 177 (2019) 22–29). Severe mitochondrial dysfunction due to FAHD1 deficiency will be compensated for by increased mtDNA copy number (as shown here for FAHD1-deficient iPSC; see also main text). SDH: succinate dehydrogenase. This figure as well as the graphical abstract were created with BioRender.com (accessed on 5 July 2021).