Perturbations of Glutathione and Sphingosine Metabolites in Port Wine Birthmark Patient-Derived Induced Pluripotent Stem Cells

Abstract

Port Wine Birthmarks (PWBs) are a congenital vascular malformation on the skin, occurring in 1-3 per 1000 live births. We have recently generated PWB-derived induced pluripotent stem cells (iPSCs) as clinically relevant disease models. The metabolites associated with the pathological phenotypes of PWB-derived iPSCs are unknown, and so we aim to explore them in this study. Metabolites were separated by ultra-performance liquid chromatography and screened with electrospray ionization mass spectrometry. Orthogonal partial least-squares discriminant, multivariate, and univariate analyses were used to identify differential metabolites (DMs). KEGG analysis was used to determine the enrichment of metabolic pathways. A total of 339 metabolites was identified. There were 22 DMs, among which nine were downregulated-including sphingosine-and 13 were upregulated, including glutathione in PWB iPSCs, as compared to controls. Pathway enrichment analysis confirmed the upregulation of glutathione and the downregulation of sphingolipid metabolism in PWB-derived iPSCs as compared to normal ones. We next examined the expression patterns of the key molecules associated with glutathione metabolism in PWB lesions. We found that hypoxia-inducible factor 1α (HIF1α), glutathione S-transferase Pi 1 (GSTP1), γ-glutamyl transferase 7 (GGT7), and glutamate cysteine ligase modulatory subunit (GCLM) were upregulated in PWB vasculatures as compared to blood vessels in normal skin. Other significantly affected metabolic pathways in PWB iPSCs included pentose and glucuronate interconversions; amino sugar and nucleotide sugars; alanine, aspartate, and glutamate; arginine, purine, D-glutamine, and D-glutamate; arachidonic acid, glyoxylate, and dicarboxylate; nitrogen, aminoacyl-tRNA biosynthesis, pyrimidine, galactose, ascorbate, and aldarate; and starch and sucrose. Our data demonstrated that there were perturbations in sphingolipid and cellular redox homeostasis in PWB vasculatures, which could facilitate cell survival and pathological progression. Our data implied that the upregulation of glutathione could contribute to laser-resistant phenotypes in some PWB vasculatures.

Keywords: Port Wine Birthmark; glutathione; hypoxia; induced pluripotent stem cells; metabolome; sphingolipid; sphingosine.

Figure 1

Characterization of normal skin and PWB-derived iPSCs. (A–D) Normal and PWB disease iPSC colonies under a bright field. (E–H) Stem cell biomarkers Nanog (E), Tra1-60 (F), Sox 2 (G), and Oct4 (H) were used to verify the control and PWB iPSC cells. DAPI: nuclei staining. Scale bar: 100 µm.

Figure 2

An example of a total ion chromatogram (TIC), representing the summed intensity with all detected mass spectral peaks associated with metabolites. (A), TIC from the PWB_4221_6 iPSC line in a negative ion mode electrospray ionization (ESI−); (B), TIC from the same iPSC line in positive ion mode electrospray ionization (ESI+); (C), OPLS-DA model showing scattering scores and cluster tendencies among all samples in ESI + mode; (D), OPLS-DA model showing scattering scores and cluster tendencies among all samples in ESI − mode. Digits on the X or Y axis are eigenvalues of the regression coefficient for the predictive principal component (X) or the orthogonal component (Y), respectively.

Figure 3

Discovery of differential metabolites (DMs) by multivariate analysis. A and B, distribution of significant metabolites detected in ESI− (A) and ESI+ (B) modes using a PLS-DA model resulting in coefficients for the variables in a w*c loading plot. w, PLS-weights for the X-variables; c, PLS-weights for the Y-variables; Red box, metabolites with VIP > 1.5; green box: metabolites with VIP < 1.5; Blue triangle is symbolized by two group dots and X-variables located near a group dot are positively associated with that group; C and D, volcano plots showing clusters of DMs detected in ESI− (C) and ESI+ (D) modes. Green triangle, significantly downregulated metabolites; red triangle: significantly upregulated metabolites; black circle: insignificant metabolites; vertical dash lines: log₂ (fold change) = ± 0.5; horizontal dash line: p value = 0.05.

Figure 4

Identification of glutathione and sphingosine metabolites. (A), identified matched glutathione MS fragments in ESI+ mode; (B), identified matched glutathione MS fragments in ESI− mode; (C), identified matched sphingosine MS fragments in ESI+ mode. The MS chromatograms of glutathione and sphingosine fragments from each sample were extracted and aligned into one chromatogram.

Figure 5

Hierarchical cluster analysis (HCA) and KEGG pathway enrichment of metabolome data. (A,B), the heatmap of significantly DMs identified in ESI− (A) and ESI+ mode (B) in PWB iPSCs as compared to the control ones; C and D, enriched KEGG pathways related to the perturbed metabolic networks involving DMs identified in ESI − (C) and ESI + mode (D); the enriched metabolic pathways are indicated by color (−log₁₀ (p value)) and size (black ball, enrichment ratio). PWB_3921_9d, PWB_4221_3d, and Control_52521_8d were experimental duplications of PWB_3921_9, PWB_4221_3, and Control_52521_8, respectively.

Figure 6

Expressions of key molecules associated with glutathione metabolism in PWB lesions. The IHC assays using antibodies against HIF−1α (A), GCLM (B), GGT7 (C), and GSTP1 (D) show the immunoreactive blood vessels in control skin or PWB lesions. H.M., a higher magnification from the pink boxed area in the left panel showing immunoreactive positive PWB blood vessels for the corresponding antibodies. Scale bar: 50 µm. n, number of blood vessels from PWB (4 subjects) or normal ones (5 subjects). Whiskers: mean ± S.D.; Diamond boxes: IQR; Dotted curves: data distribution. Paired t-test was used for comparing arbitary IHC scores. Blue arrows: control dermal capillaries; red arrows: PWB blood vessels.