In this study we used dermal fibroblasts

In this study we used dermal fibroblasts derived from three different patients with distinct mutations in the TAZ1 gene (Houtkooper et al., 2009) and generated BTHS patient-specific iPSCs. We fully characterized them in terms of pluripotency both in vitro and in vivo. Similar to human ESCs, all generated iPSCs show the self-renewal capacity and are able to differentiate into derivatives of all three embryonic germ layers. Importantly, BTHS–iPSCs exhibit the typical defects in CL remodeling: lower CL levels, accumulation of MLCL, and a shift of CL and MLCL towards more saturated species. It is important to note that the CL composition of cultured iPSCs differs from the CL composition in heart tissue, in which tetralinoleyl-CL is the predominant form (Houtkooper et al., 2009). Cultured iPSCs contained the main CL clusters found in heart, but the linoleic histone methyltransferase inhibitor content is lower. One likely explanation is that cultured iPSCs rely predominantly on glucose as their major source of energy (in contrast to the heart); therefore, the CL composition in iPSCs does not need specific adaptation to tetralinoleyl-CL for optimizing high mitochondrial activity. Similar hypothesis has been suggested for the CL composition in the brain by Cheng et al. (2008). Another likely explanation is that the lipid composition in cultured cells highly depends on the fatty acids available in the culture medium. A direct correlation of CL fatty acid composition and fatty acid supplementation in the growth media has been shown previously for patient fibroblasts (Valianpour et al., 2003). Further studies should investigate the CL composition in cultured cardiomyocytes derived from iPSCs. In our iPSC model of BTHS, we observed structural rearrangements of respiratory chain complexes in all iPSC lines derived from three different patients, which most dramatically affect the large supercomplexes (respirasomes) (Figs. 6A and B), consistent with data from Barth patient derived lymphoblasts (Mckenzie et al., 2006). In these studies, the I–III2 complexes was the largest respiratory complex which was stable in Barth patients. This form is hardly visible in Figs. 6A and B, however, extended exposures confirmed it to be stable in BTHS patient iPSCs (data not shown). Steady state level protein analysis indicated that the stability of most constituents of the respiratory chain complexes is not affected (Fig. 6C). Instead, smaller assembly forms of respiratory chain complexes, such as the dimer of complex III2, monomeric complex IV and the III2–IV heterooligomer become more prominent in BTHS–iPSCs. Interestingly, these complexes showed a slightly faster migration behavior in TAZ13 iPSCs. The faster migration of the monomeric complex IV has been observed before in lymphoblast of Barth patients (Mckenzie et al., 2006) and can be either explained by differences in the solubilization of the complexes due to the absence of cardiolipin or by the dissociation of single components. Structural data revealed that components, such as COX6A form the periphery of the cytochrome c oxidase complex (Tsukihara et al., 1996) and biochemical data indicate that dissociation of peripheral subunits occurs with decreasing CL content (Sedlák and Robinson, 1999). The formation of respiratory chain supercomplexes is essential for efficient energy transformation. To evaluate if the structural reorganization, observed in BTHS patients will affect mitochondrial respiration, we analyzed the metabolism of BTHS–iPSCs. Patient TAZ10 and TAZ13 iPSCs showed a strong defect in basal respiration as well as in their maximal respiratory capacity, indicative of reduced energy conversion by the respiratory chain complexes. An assessment of cellular growth rate excluded the possibility that a difference in cell numbers is responsible for this effect. A determination of mitochondrial proteins of a normalized amount of cells also excluded the possibility that impaired respiration in these cells is a result of decreased mitochondrial proliferation. In line with our findings, previous reports showed that mitochondrial content in BTHS fibroblast and lymphoblast is rather increased than decreased (Xu et al., 2005; Barth et al., 1996). We therefore conclude that the structural remodeling of supercomplexes results in a decreased mitochondrial respiration, in agreement with earlier reports on a decreased activity of complexes III and IV (Xu et al., 2005; Barth et al., 1996). Interestingly, full uncoupling using FCCP induces respiratory rates comparable to basal respiration. This is in contrast to what has been observed for differentiated cells such as fibroblasts. We speculate that basal respiration of iPSCs is close to maximal respiration and decreases after initiation of differentiation. We propose that the mutation in TAZ13 in a splice acceptor site at the intron 1–exon 2 boundary will result in a non-functional form of the tafazzin protein, explaining the severity of this mutant. Patient TAZ15 shows a slightly milder phenotype in respiratory measurements. Recently, a yeast mutant resembling the mutation of Barth patient TAZ15 was described (Whited et al., 2012). The yeast mutant protein was stable and localized in the IMS within mitochondria. When the enzymatic transacylase activity was determined, only prolonged exposure to restrictive conditions decreased enzymatic activity, indicative of a temperature sensitive phenotype of this mutant. We therefore conclude that differences in the severity of the mutation might be reflected in the extent of respiratory deficiency. Clinically, the severity of manifestations of BTHS varies among patients. Larger sets of BTHS–iPSC lines harboring different mutations in TAZ1 are therefore valuable to further validate the patient-specific and mutation site-specific disease phenotype and compare pathogenetic mechanisms in diverse forms of the disease.