Urinary neutrophil gelatinase-associated lipocalin (NGAL) is associated with severity of renal disease in proteinuric patients

Sir, Persistent proteinuria is an expression of renal damage caused by numerous factors, but it also represents an important cause of renal injury progression that can lead to chronic renal failure (CRF) [1]. Effectively, the presence of plasmatic proteins within tubular lumen is a source of further harm to renal tubule cells, probably determined by complement cascade activation [2,3] and many authors have hypothesized that this condition plays a basic role in the appearance and subsequently the progression from isolated macro-proteinuria to CRF [4]. Neutrophil Gelatinase-Associated Lipocalin (NGAL) is a protein produced by neutrophils, but also released by many epithelial cell types including kidney tubular cells [5]. Its synthesis and urinary excretion increase in many pathological conditions, such as ischemia, infections and toxicosis [6–8]. For these reasons, we aimed to analyse the urinary concentrations of NGAL (uNGAL) in 33 patients (18 M,15 F; mean age 55 ± 26; GFR 59.3 ± 17.2 ml/min) affected by persistent macroproteinuria associated with primitive membranous (n = 17) and membranoproliferative glomerulonephritis (n = 16). During the study, none of the patients was treated with steroids, immuno-suppressors, ACE-inhibitors or angiotensin 2 receptor blockers (ARB). All the patients showed proteinuria >1 g/day for at least 6 months. Twenty healthy subjects were enrolled as controls. The protocol was approved by the local Ethics Committee and informed consent was obtained from all subjects. NGAL was measured in urine using ELISA commercial available kit (Antibody Shop, Gentofte, Denmark) and its levels were expressed as nanogram per millilitre. Statistical analysis of the data was carried using GraphPad Prism (version 4.0) package. Unpaired two-tailed t-test and Pearson correlation coefficient were used respectively to test comparisons and correlations. P < 0.05 was considered significant. The results of our study showed, for the first time in the literature that in patients with macroproteinuria, uNGAL concentrations were significantly higher than in controls (378.28 ± 111.13 vs 7.38 ± 3.26 ng/ml; P = 0.01) (Fig 1) and furthermore that these levels were directly correlated with the entity of urinary protein loss (r = 0.294, P = 0.01) and inversely with residual renal function (uNGAL/GFR: r = −0.528, P = 0.04; uNGAL/Creatinine: r = 0.588, P = 0.02) (Table 1). Differences in uNGAL excretion between patients and controls (*P = 0.01). Comparison and correlations (r) of uNGAL sCr, serum creatinine; GFR, glomerular filtration rate (Cockcroft-Gault); uNGAL, urinary NGAL; P24, proteinuria; MN, Membranous glomerulonephritis; MPN, Membranoproliferative glomerulonephritis. ar = 0.588, P = 0.02; br = −0.528, P = 0.04; Correlation with uNGAL: cr = 0.294, P = 0.01; dP = 0.01 vs controls; eP not significant vs MPN group. Comparison and correlations (r) of uNGAL sCr, serum creatinine; GFR, glomerular filtration rate (Cockcroft-Gault); uNGAL, urinary NGAL; P24, proteinuria; MN, Membranous glomerulonephritis; MPN, Membranoproliferative glomerulonephritis. ar = 0.588, P = 0.02; br = −0.528, P = 0.04; Correlation with uNGAL: cr = 0.294, P = 0.01; dP = 0.01 vs controls; eP not significant vs MPN group. In the first place, the condition could be explained by the fact that a greater quota of the fraction of NGAL produced at a systemic level is lost through damaged glomerulus, as happens in a similar way with other plasmatic proteins. The evidence of a strict correlation between uNGAL and the degree of proteinuria confirms this hypothesis, even though the possibility of holding other factors responsible is not excluded. For example, bearing in mind that NGAL is released copiously from the renal epithelial after injuries of various kinds [6–8], it should not be excluded that a significant quota of uNGAL can be attributed to a response of such tissue to the condition of chronic tubular stress linked to the transit, elevated and prolonged, of haematic proteins. It is well-known, in fact, that a persistent macroproteinuria itself constitutes a factor of progressive renal damage caused by the toxicity of the plasmatic protein on tubular renal cells, particularly proximal ones [1,4]. Such toxicity promotes apoptosis, causing at first an increased turnover of the epithelial and, subsequently, tubular atrophy and interstitial fibrosis, as experimental models of nephropathy have demonstrated after prolonged proteic overload [9–11]. A fundamental role, in that sense, seems to be linked to intra-tubular activation of the complement cascade as previous studies have found a strict correlation between urinary excretion of some factors of the complement and the extent of daily proteinuria [12–14]. The increased production of NGAL by tubular cells could in this context constitute a compensatory mechanism, with defensive intent, based on the already described inhibition of apoptosis processes by this protein, but it should not be excluded that the same cell damage from chronic tubular injury causes increased urinary excretion of NGAL: this result would concur with past observations in experimental models of protein-overload nephropathy, for another ‘tubular stress’ protein, KIM-1 [15]. Our hypothesis, that levels of NGAL are in some way an expression of subacute renal suffering, is confirmed by the fact that in our patients NGAL also showed a significant correlation to the residual renal function. Other authors have also highlighted the same correlations in other pathologies that hesitate in a state of CRF, such as nephropathic processes in the paediatric field [16]. In the latter context, Devarajan et al. [17] have even hypothesized as to the possibility of using serum NGAL as a new marker of renal function, on the basis of evidence that the protein correlated with effective GFR better than creatinine and cystatin C. The exact mechanism at the basis of that correlation has not yet been well-defined. The most likely hypothesis still remains that of anti-apoptotic tubular compensatory mechanism to chronic renal damage, while the lower reduction of renal clearance would not, in the first place at least, be held responsible. In conclusion, although we have highlighted a close link between uNGAL and severity of renal disease in our proteinuric patients, the findings reported cannot clarify the pathophysiological significance of this relationship. The hypothesis formulated should therefore be confirmed by further in-depth studies. Conflict of interest statement. None declared.