Considering the utility of urinary amino acids for early identification of non-diabetic chronic kidney disease

Abstract

Full text

To the Editor,

The excess excretion of amino acids in urine signifies loss of nutrients and metabolic dysregulation, which are pathophysiological processes involved in chronic kidney disease (CKD). Urinary amino acids have previously been investigated as a novel non-invasive metabolomic biomarker for early identification of diabetic kidney disease (DKD) [1–3]. However, the utility of urinary amino acids has not been studied in the context of non-diabetic early CKD, which is also a prevalent CKD population at risk of adverse clinical outcomes.

Our group conducted a cross-sectional pilot study, performing urinary amino acid profiling in 20 adult non-diabetic patients (Table 1). All 20 patients had kidney biopsy performed and CKD status was determined here as per interstitial fibrosis and tubular atrophy (IFTA) grading. There were 10 patients with IFTA 0%–10% (i.e. no CKD) and 10 patients with IFTA 10%–25% (i.e. early CKD). The two groups were statistically similar in age, a median age 67 years (interquartile range (IQR) 41–71 years) in the no CKD group versus 62 years (IQR 57–68 years) in the early CKD group, *P *= .880; sex, all 20 patients were male; and estimated glomerular filtration rate (eGFR), a median eGFR 90 ml/min/1.73 m^2^ (IQR 90–90 ml/min/1.73 m^2^) in the no CKD group versus 88 ml/min/1.73 m^2^ (IQR 86–90 ml/min/1.73 m^2^) in the early CKD group, *P *= .246. In the early CKD group, there were seven patients with micro- or macroalbuminuria. Urine samples were collected immediately before kidney biopsy and processed through HPLC and mass. Specific metabolites were identified, and urinary amino acid concentration (reflecting absolute abundance) and mole% (reflecting relative abundance) were determined via bioinformatics analyses. The differences in median urinary amino acid concentration (in µg/ml) and mole% between the non-diabetic early CKD and no CKD groups were examined.

Eighteen amino acids were identified and characterized across the non-diabetic early CKD and no CKD groups (Table 2). The median urinary concentration (3.6 vs 0.8 µg/ml, *P *= .012) and mole% (1.18% vs 0.69%, *P *= .034) of valine and the median urinary mole% of alanine (6.95% vs 5.12%, *P *= .049) were significantly higher in the non-diabetic early CKD compared to the no CKD group. Conversely, the median urinary mole% of histidine (11.2% vs 15.3%, *P *= .041) was significantly lower in the non-diabetic early CKD compared to the no CKD group.

Previous studies observed significant associations between urinary levels of valine and DKD progression [1, 3]. The exact mechanisms on how valine, a branched-chain amino acid (BCAA), contributes to DKD pathophysiology is still being investigated for both Type 1 and 2 diabetes mellitus (DM). It is now widely considered that increased BCAA levels is associated with insulin deficiency, resistance, and dysregulation in glucose metabolism, all risk factors for DKD development [4]. Otherwise, urinary alanine levels are found to be predictive of DKD [2]. The complex interplay between alanine, its associated enzyme alanine aminotransferase, and CKD/DKD is not yet fully explored, but nonetheless deemed to be potentially correlated with mechanisms involving mitochondrial dysfunction and oxidative stress [5]. Decreased plasma and urinary levels of histidine were observed in patients with DKD compared to DM patients and healthy controls [3]. Linked to protein synthesis as well as anti-oxidation and anti-inflammation, significant associations between decreased histidine levels and protein-energy wasting, oxidative stress and inflammation in CKD/DKD, and adverse CKD/DKD outcomes were established [6].

In summary, our pilot study demonstrated by using HPLC-MS that urinary levels of specific amino acids differed between individuals with non-diabetic early CKD and no CKD. There is potential in utilizing the urinary levels of valine, alanine, and histidine as non-invasive biomarkers for early identification of non-diabetic CKD. Further delineation on the roles of these amino acids in non-diabetic CKD pathophysiology is needed going forward.