Nuclear magnetic resonance detects phosphoinositide 3-kinase/Akt-independent traits common to pluripotent murine embryonic stem cells and their malignant counterparts

Abstract

Pluripotent embryonic stem (ES) cells, a potential source of somatic precursors for cell therapies, cause tumors after transplantation. Studies of mammalian carcinogenesis using nuclear magnetic resonance (NMR) spectroscopy have revealed changes in the choline region, particularly increased phosphocholine (PCho) content. High PCho levels in murine ES (mES) cells have recently been attributed to cell pluripotency. The phosphoinositide 3-kinase (PI3K)/Akt pathway has been implicated in tumor-like properties of mES cells. This study aimed to examine a potential link between the metabolic profile associated with choline metabolism of pluripotent mES cells and PI3K/Akt signaling. We used mES (ES-D3) and murine embryonal carcinoma cells (EC-F9) and compared the metabolic profiles of 1) pluripotent mES (ESD0), 2) differentiated mES (ESD14), and 3) pluripotent F9 cells. Involvement of the PI3K/Akt pathway was assessed using LY294002, a selective PI3K inhibitor. Metabolic profiles were characterized in the extracted polar fraction by (1)H NMR spectroscopy. Similarities were found between the levels of choline phospholipid metabolites (PCho/total choline and PCho/glycerophosphocholine [GPCho]) in ESD0 and F9 cell spectra and a greater-than five-fold decrease of the PCho/GPCho ratio associated with mES cell differentiation. LY294002 caused no significant change in relative PCho levels but led to a greater-than two-fold increase in PCho/GPCho ratios. These results suggest that the PCho/GPCho ratio is a metabolic trait shared by pluripotent and malignant cells and that PI3K does not underlie its development. It is likely that the signature identified here in a mouse model may be relevant for safe therapeutic applications of human ES cells.

Figure 1

Phosphatidylcholine metabolism. The tinted square highlights the pathway of the choline cycle investigated in the study. CDP-choline indicates cytidine diphosphate choline; PKC, protein kinase C; PKA, protein kinase A.

Figure 2

Pluripotency and differentiation. (a) Western blot analysis for Oct 4 shows strong expression in ESD0 and F9 cells and its absence from ESD14 cells. Expression of β-actin confirms equal loading. (b) Semiquantitative RT-PCR analysis shows expression of transcripts of TBra, AFP, and NF-68, markers of mesoderm, endoderm, and ectoderm, respectively, in differentiated ESD14 cells. In contrast, these transcripts are not detected in undifferentiated ESD0 and F9 cells. β-actin was used as loading control.

Figure 3

PI3K/Akt signaling. (a)Western blot analysis for pAkt shows its strongest expression in F9 cells. For mES cells, Akt phosphorylation is high in ESD0 cells and decreases with cell differentiation (ESD14). (b) Semiquantitative RT-PCR for Eras transcript shows its strongest expression in F9 cells and weakest in differentiated ESD14 cells. (c) Treatment of cells with 10 µM LY294002 for 24 hours caused an almost complete attenuation of Akt phosphorylation.

Figure 4

NMR spectra. (a) Representative ¹H NMR spectrum of a mES cell extract. Predominant metabolite resonances are labeled on the spectrum. Val indicates valine; Ile, isoleucine; Leu, leucine; Lac, lactate; Ala, alanine; Gluc, glucose; Cr, creatine; AMP, adenosine monophosphate. (b) Section of ¹H NMR spectra (expanded between 3.195 and 3.245 ppm) of mES cells at day 0 with (ESD0 + LY, red line) and without (ESD0, blue line) LY treatment, mES cells after 14 days of growth (cyan line) and teratocarcinoma cell line F9 with (F9 + LY, black line) and without (F9, green line) LY294002 treatment.

Figure 5

Choline phospholipid metabolites. (a) PCho and total choline (Tot choline) ratios. PCho / (GPCho + PCho + Cho) for mES cells at day 0 with (ESD0 + LY) and without (ESD0, ESD0V) LY294002 treatment, mES cells after 14 days of differentiation (ESD14) and teratocarcinoma cells with (F9 + LY) and without (F9, F9V) LY treatment. Differentiation is associated with a significant reduction of PCho/Tot (total) choline (ESD0 vs ESD14, P < .05). In both cell lines, treatment with LY294002 caused observable but not significant increases in the PCho/Tot choline ratio. Vehicle alone (ESD0V; F9V) did not have any impact on PCho/Tot choline ratios. ESD0V and F9V—cells treated with vehicle alone. (b) PCho and GPCho ratios. PCho/GPCho for mES cells at day 0 with (ESD0 + LY) and without (ESD0, ESD0V) LY294002 treatment, mES cells after 14 days of differentiation (ESD14) and teratocarcinoma cells with (F9 + LY) and without (F9, F9V) LY294002 treatment. In differentiated mES cells, there is a significant decrease in PCho/GPCho ratios compared with undifferentiated cells (ESD0 vs ESD14, P < .0001). In both pluripotent cell lines (ESD0 and F9), treatment with LY294002 caused a significant (P < .001) increase in PCho/GPCho ratios. Vehicle alone (ESD0V and F9V) did not have any impact on PCho/GPCho concentrations.