Free Radic Biol Med. 2013 Jul;60:49-55.
Albumin oxidation leads to neutrophil activation in vitro and inaccurate measurement of serum albumin in patients with diabetic nephropathy.
Authors: Regina Michelis1*, Batya Kristal2,3, Teuta Zeitun2, Galina Shapiro1, Yoav Fridman1, Ronit Geron2 and Shifra Sela1,3.
1Eliachar Research Laboratory and 2Nephrology Department, Western Galilee Hospital, Nahariya, Israel; 3 Faculty of Medicine in the Galilee, Bar Ilan University, Safed, Israel
BACKGROUND: Previous studies suggest that oxidative modifications of serum albumin lead to underestimation of albumin concentrations using conventional assays. Secondly, oxidation of serum albumin may cause neutrophil activation and further oxidation of albumin, which may result in a series of reciprocal cyclical processes. Since hypoalbuminemia, systemic inflammation and oxidative stress are common in diabetic nephropathy patients, the aim of the present study was to show that albumin modifications and neutrophils activation underlie these reciprocal systemic processes.
METHODS: Blood samples from a cohort of 19 patients with diabetic nephropathy and 15 healthy controls were used for albumin separation. An oxidation-dependent “albumin-detection index”, representing the detection efficacy of the universal bromocresol-green (BCG) assay, was determined for each subject. This index was correlated with serum albumin levels, various markers of oxidative stress or inflammation and kidney function. Activation of separated neutrophils by glycoxidized-albumin was assessed by the release of neutrophil gelatinase-associated lipocalin (NGAL) and myeloperoxidase (MPO).
RESULTS: The albumin-detection index of diabetic nephropathy patients was significantly lower compared to controls, correlating positively with serum levels of albumin and kidney function, and negatively with albumin glycoxidation and inflammatory markers. Glycoxidized-albumin had a direct role in neutrophil activation, resulting in NGAL and MPO release.
CONCLUSIONS: The hypoalbuminemia observed in patients with diabetic nephropathy partially results from underestimation of modified/oxidized albumin using the bromocresol-green assay. However, modified or oxidized albumin may lead to a cycle of accelerated oxidative stress and inflammation involving neutrophil activation. We imply that albumin detection index, a new marker of oxidative stress, may also serve as a biomarker of diabetic nephropathy severity and its progression.
Hypoalbuminemia, defined as human serum albumin (HSA) levels < 3.8g/dl, is a common finding in diseases associated with chronic inflammation and severe oxidative stress (OS), including diabetic nephropathy (DN). In DN, hypoalbuminemia has traditionally been attributed to urinary albumin losses, despite the fact that the liver bears a huge potential albumin production capacity of (~25 g/day). In patients with end stage renal failure treated by hemodialysis (HD) hypoalbuminemia is partially due to oxidative modifications of albumin which impair albumin quantification by the standard laboratory assay bromocresol green (BCG) . One aim of the study was to investigate this possibility in DN patients.
The second aim of the study was to assess the possibility that the oxidized/modified albumin molecules can activate neutrophils to release reactive oxygen species (ROS) and inflammatory mediators. Peripheral neutrophils are a source of reactive oxygen species (ROS) and are activated in diabetes and in chronic kidney disease, thus contributing to OS. If neutrophil activation by oxidized/modified albumin increases ROS and inflammatory mediator release, such findings would indicate the involvement of oxidized/modified albumin molecules in an accelerating cycle of OS and systemic inflammation in these groups of patients.
To study these issues we used an index named “albumin detection index”, defined as the ratio between HSA concentrations determined by BCG to total HSA concentration. Total concentration of HSA was determined by absorbance at 280nm (OD280) or by ELISA. An index value of ~1 indicates native, unmodified HSA molecules, while values <1 indicate the presence of modifications that decrease the detection efficacy of HSA by the BCG assay. Serum contains many proteins besides HSA, which can affect the OD280. Therefore, HSA was separated from other serum proteins by gel-filtration (GF) chromatography, prior to measurements by BCG and OD280.
We found decreased values of the albumin detection index in DN patients (by 25%) compared with healthy control subjects (HC; fig. 1A). In hypoalbuminemic (Hypo) DN patients the detection index was significantly lower then in normoalbuminemic (Normo) patients. The index correlated with the levels of HSA in serum (fig. 1B). When this correlation was evaluated for each group, significance was found in DN patients but not in HC (fig. 1B). * indicates p<0.01 vs HC; # indicates p<0.02 vs normoalbuminemic DN patients.
Decreased index values suggest that HSA quantification is impaired so that actual HSA levels may be higher than measured. To evaluate this possibility we compared patients with nearly identical HSA levels (3.5-3.7 g/dl), divided into two groups: one with low detection index and another with normal index. In this simple experiment sera were separated by SDS-PAGE, and the protein bands were stained and quantified. The results (figure 2) demonstrated increased albumin signals in the group with decreased index, thus indicating the existence of higher HSA levels that are not being represented when using the BCG assay, since the measured HSA values in the clinical biochemistry laboratory were identical in both groups.
Figure 2: Sera of DN patients with HSA levels of 3.5-3.7g% and an albumin detection index of ~0.5 (patients 1-3) or ~1 (patients 4-5) were separated by SDS-PAGE. The gel was silver-stained and the albumin band (67kD) was quantified by densitometry (Densit.).
The correlations of the albumin detection index with OS and inflammation were studied. The index showed an inverse significant correlation with the neutrophil counts (a parameter related to systemic OS and inflammation), with the levels of neutrophil gelatinase-associated lipocalin (NGAL-a marker of neutrophils activation) and with advanced glycation end products (AGE) of HSA (an OS marker). The correlation of the index with AGE-HSA is similar to our previous findings in HD patients .
These correlations prompted us to hypothesize that AGE-albumin (whose levels are elevated in DN) may be directly involved in neutrophil activation. This assumption was investigated by incubating separated neutrophils with AGE-HSA and following the release of NGAL and myeloperoxidase (MPO, another marker for neutrophils activation and an enzyme involved in ROS release). The results indicated activation of HC neutrophils (evident by the release of NGAL and MPO) by AGE-HSA, compared with the native, unmodified-albumin (table 1). In DN neutrophils a similar trend was found. Neutrophil activation suggests a potential role of modified albumin in the acceleration of systemic inflammation and OS.
Table 1: Neutrophils were separated from blood samples of HC and DN subjects and incubated with in-vitro preparations of AGE-modified (AGE), unmodified (U) HSA or without albumin (cells). NGAL and MPO levels were measured in the supernatant. P values are indicated for comparison of AGE-albumin treatment of neutrophils vs. treatment without albumin [p(C)] or with unmodified-albumin [p(U)].
This study addresses, for the first time, the association among systemic inflammation, neutrophil activation, and albumin modifications in a cohort of DN patients. In an environment of increased OS and inflammation such as diabetes, albumin detection efficacy of the routinely used BCG assay is decreased, causing albumin read-out to be lower than its actual level. The albumin detection index demonstrated a significant association with OS and inflammatory processes that parallel the progression of the renal disease. The impairment of albumin measurements in clinical laboratories is not the only predicament arising from this study. Hypoalbuminemia associated with low albumin detection index actually indicates the presence of modified albumin molecules that play a role in a cycle of accelerated OS and inflammation involving the activation of neutrophils, suggesting these modified molecules are a pathogenic factor with the potential to trigger endothelial dysfunction either directly or via activated neutrophils. Hence, the albumin detection index can serve as a potential biomarker of DN severity that reflects systemic OS, inflammation and kidney function.
1. Michelis, R.; Kristal, B.; Snitkovsky, T.; Sela, S. Oxidative modifications impair albumin quantification. Biochem. Biophys. Res. Commun. 401:137-42; 2010.