Synopsis

Defects in the mitochondrial OXPHOS lead to an extremely heterogeneous group of disorders and common laboratory screening methods are prone to misinterpretations. In this study we report metabolic adaptation of OXPHOS deficient human skin fibroblasts during galactose stress condition. We report a rapid HR-MAS NMR based method for the relative quantification of untargeted metabolome profiling providing containment of over 40 metabolites, which allowed us to observe unique cellular metabolic profiles in OXPHOS deficient fibroblast.

Synopsis

Introduction

Defects in the mitochondrial oxidative phosphorylation (OXPHOS) lead to an extremely heterogeneous group of disorders. Laboratory screening methods are prone to misinterpretations and do not allow for unambiguous diagnosis. New diagnostic methods are therefore needed.

In this study we investigate metabolic adaptation of OXPHOS deficient human skin fibroblasts (FB) during galactose stress condition by combined High-Resolution Magic Angle Spinning (HR-MAS) NMR and metabolic flux analysis. Furthermore, we test metabolic profiling using HR-MAS NMR as a diagnostic tool for mitochondrial diseases.

Methods

Cell culturing: We used human skin fibroblasts (FB) from five controls, two patients with complex 1 deficiency (C1, namely ND6 and NDUFAF2), and from two patients with pyruvate dehydrogenase deficiency (PDH, namely 1Eα defects). Cells were grown for 18h under standard culture condition (Dulbecco's Modified Eagle Medium with 5.5 mM glucose, 10% dialyzed FCS, 2 mM L-glutamine, 1 mM sodium pyruvate, 50μM uridine) or were exposed to galactose based stress medium (10mM Galactose).

NMR Spectroscopy: Sample preparation for ¹H-HR-MAS NMR was performed as described previously⁽¹⁾. Each sample was lysed after collection and additionally heated (70°C). On average 2.2 Million cells in D₂O-based PBS were inserted in 4mm rotors using a 50µl insert. HR-MAS NMR experiments were performed on a 500 MHz Bruker Avance II spectrometer operating at 285K and 3kHz spinning. A PROJECT⁽²⁾ pulse sequence with water presaturation was used for acquiring ¹H-NMR spectra with a TE of 211ms. 768 transients were acquired. Spectral processing was performed using Topspin^TM software, including line-broadening, FFT, phasing and baseline-correction. Chemometrics and individual peak analysis was performed with MatLab, PLS_Toolbox (Eigenvector Research,Inc.), using PQN-normalized individually sized buckets.

Metabolic flux analysis: We investigated metabolic flux in unbuffered culture medium using a Seahorse XF24 cell Analyzer. The oxygen consumption rate (OCR), an assessment of aerobic mitochondrial function, and extracellular acidification rate (ECAR), a detection of extracellular acid production, were measured simultaneously.

Results

Chemometric analysis of NMR spectra

Resonances of the NMR spectra were assigned to approximately 40 metabolites (Fig.1). Partial least squares Discriminant Analysis (PLS-DA) revealed (based on cross-validation) significant separation of metabolomes of control, PDH defects and C1 deficient FB (all P < 0.05). Separation of metabolomes of control and PDH defects under glucose culture conditions were also significant (P < 0.05). C1 deficient FB were not significantly separated under glucose (Fig.2).

Analysis of individual metabolites revealed specific defect dependent differences (see Fig.3). PDH defects showed around 2-fold decrease in choline (Cho), phosphocholine (PC) and glycerophosphocholine (GPC) under glucose, whereas only GPC was significantly decreased under galactose. In contrast, C1 defects showed around 1.5-fold increase in Cho and PC under galactose. Furthermore, PDH deficient FB were separated by a pattern of increased accumulation of branched-chain and ketogenic amino acids under both culture conditions.

Energetic phenotypes using metabolic flux analysis

Metabolic flux analysis revealed clear separation of C1 and PDH deficient FB from controls (Fig.4) confirming expectation. PDH defects were characterized by an increased ECAR and shifted under galactose to an even more stressed phenotype (higher OCR and ECAR). In contrast, C1 showed a decreased metabolic activity (low OCR and ECAR) and shifted towards a more aerobic phenotype under galactose (higher OCR), which represented the physiological adaptation in control FB.

Discussion

OXPHOS deficiency resulted in distinct changes in several Cho metabolites, which were oppositional in the two studied defects. Changes in Cho metabolites may be associated with altered cellular membrane turnover, possibly including mitochondrial membranes⁽³⁾. Altered Cho catabolic and anabolic pathways might thus represent interesting potential biomarkers for mitochondrial dysfunction.

Furthermore, accumulation of amino acids might be caused by increased protein degradation in order to fuel krebs cycle and/or inhibited catabolism. Interestingly, branched-chain amino acids are also involved in metabolic signalling⁽⁴⁾. Further investigations are necessary in order to elucidate the association between PDH defects and amino acid dysmetabolism.

Conclusion

Acknowledgements

References

(1) G. Diserens, D. Hertig, M. Vermathen, B. Legeza, C.E. Fluck, J.M. Nuoffer, and P. Vermathen, Analyst 2017, 142(3): p. 465-471.

(2) J. A. Aguilar, M. Nilsson, G. Bodenhausen, G.A. Morris, Chemical Communications 2012, 48, 811-813.

(3) A.T. Baykal, M.R. Jain, H. Li, Metabolomics 2008, 4(4):347-356.

(4) C.J. Lynch, S.H. Adams, Nat Rev Endocrynology 2014, 10(12):723-736.