Background and Aim: Several forms of adrenal hypertension, including Pheochromocytoma (PPGL) and Primary Aldosteronism (PA) have been described, and they all share an abnormally high blood pressure as a symptom coupled with an endocrine underlying cause. Even though these diseases are associated with cardiovascular and metabolic complications, current diagnostic approaches often fail to detect them in time. Nuclear Magnetic Resonance Spectroscopy (NMR) has proven to be a valuable tool in the discovery and quantification of disease-related biomarkers, but a reform of the traditionally used protocol was necessary for applying NMR metabolomics on a large set of plasma samples collected for studying adrenal hypertension. To this end, we compared the results from several methods aimed at focusing on small polar metabolites and preparing spectra for multivariate data analysis. The resulting protocol consists of a novel combination of a NMR experiment, a seldom used internal standard for metabolite quantitation, and a data processing routine.

Materials and Methods: Plasma was collected from four patients, two of which suffered from PA and the other two from PPGL. Pooling the plasma from each set of two patients and subsequently aliquoting it, resulted in the creation of two groups, each consisting of 10 replicate samples. These were either subjected to the traditional ultrafiltration approach to remove proteins and lipids altogether, or left unfiltered for analysis with NMR experiments CPMG and LED for removing only the signals originating from the macromolecules. A new internal standard, Maleic Acid, was added to the buffer solution for absolute quantification of metabolites, and its peak compared to that of the commonly used internal standard, TSP, which is known to broaden in the presence of macromolecules, thus hindering quantification. Spectra were recorded on a Bruker Avance III, operating at 500 MHz, and next underwent data reduction using the SPEAQ R package for peak picking and alignment. Results were assessed using the broker Topspin software for spectral inspection and the SIMCA-P software for Principal Component Analysis (PCA).

Results: Both CPMG and LED spectra were deprived from high signals arising from proteins and lipids, indicating the capability of both methods to suppress macromolecule peaks. Maleic acid did not display any peak broadening, in contrast to the traditional quantification standard. PCA models show the metabolic signature of the LED method being closer than CPMG to the ultrafiltration. The LED approach was also able to successfully differentiate the groups of samples, with an adequately low within group variance and a high between group variance.

Conclusions: Overall, our results indicate that by adding Maleic Acid as an internal standard to unfiltered plasma samples, using LED NMR spectroscopy and the SPEAQ data processing pipeline, we were able to achieve a clear separation of PPGL from PA samples according to statistical models with a high degree of variation explained and a high predictive ability. The LED experiment was capable of suppressing protein and lipid signals, indicating its applicability in plasma metabolomics. The fact that Maleic Acid gave rise to a peak that was unaffected by the presence of macromolecules, shows that it is a reliable standard for quantification. With its speed, affordability and ease of use, this complete approach is well suited for large scale metabolomics studies in group differentiation, sample stratification, and putative metabolite quantification. This research is expected to prove valuable in providing the means for upgrading the diagnosis of the different forms of adrenal hypertension.