1. In recent years, evolutionary ecologists have become increasingly interested in antioxidants and oxidative stress. Information on redox systems can provide new insights into our understanding of life-history variation and animal responses to environmental stressors.
2. A common approach of ecological studies to the study of antioxidant capacity of animals has been measurement of the total antioxidant capacity of serum or plasma. Some of these studies have suggested that most of the antioxidant capacity measured in plasma is made up of uric acid and, therefore, estimates of antioxidant capacity should be corrected for the concentration of uric acid.
3. Here, I show that (i) the correlation between plasma concentration of uric acid and plasma antioxidant capacity is method dependent and (ii) different assays for the quantification of circulating antioxidant capacity can provide information on different components of the antioxidant machinery.
4. To determine whether measurements of antioxidant capacity need to be corrected for the uric acid concentration in the sample, it is therefore important to take into account the biochemical properties of the assay used.
The term “antioxidant”, while often not clearly and consistently defined, refers to molecules (e.g. antioxidant enzymes, dietary antioxidants, repair systems) or structures (e.g. composition of cell membranes) that prevent or minimize oxidative damage (see also Halliwell & Gutteridge 2007). The analyses of antioxidants in ecological research have basically focussed on two approaches: quantification of single classes of antioxidants (e.g. vitamin E, carotenoids) or estimates of the total antioxidant capacity of a biological matrix (e.g. tissue, yolk). The second approach allows the estimation of how the whole suite of antioxidants present in a certain biological matrix might respond to an in vivo oxidative challenge.
Although the analysis of a tissue’s total antioxidant capacity by itself is not sufficient to make inferences about the level of oxidative stress (Costantini & Verhulst 2009), it is still informative, as it provides a crude estimate of how the organism could be expected to respond to an oxidative insult (e.g. through upregulation or remobilization of antioxidants) and of the potential costs associated with the antioxidant response itself. Mounting an antioxidant response requires the consumption of energy for the synthesis of antioxidant enzymes, or the use of exogenous antioxidants which may be very limiting in the diet.
There are several colorimetric assays on the market that can be used to measure total antioxidant capacity [e.g. ferric reducing ability of plasma (FRAP), oxygen radical absorbance capacity (ORAC), OXY, trolox equivalent antioxident capacity (TEAC)], some of which have been used in recent years for the analysis of antioxidant capacity of serum or plasma of birds. Some studies suggested that most of the antioxidant capacity measured in avian plasma by the TEAC or AOP-490 assay is made up of uric acid (Cohen, Klasing, & Ricklefs 2007; Horak et al. 2007). Uric acid is a waste product of nitrogen metabolism, which has antioxidant properties: it can, for example, inhibit lipid peroxidation (Smith & Lawing 1983) and protect DNA from free radical damage (Cohen, Aberdroth, & Hochstein 1984). It is unclear, however, whether changes in uric acid concentration in the blood reflect an adaptive response to oxidative stress, a passive response to a change in protein metabolism or the onset of a disease. For these reasons, it has been suggested that estimates of antioxidant capacity should be corrected for the concentration of uric acid (e.g. by use of residuals) (Cohen, Klasing, & Ricklefs 2007). These general statements, however, may give the impression that uric acid is the most significant molecule to characterize the antioxidant barrier in the circulating system and that a correction for uric acid is always needed whatever assay of antioxidant capacity is used.
My aim here is to make the points that (i) the importance of the contribution of uric acid to measurements of circulating antioxidant capacity is method dependent and (ii) different assays can provide different information on antioxidant capacity that can be combined to produce a better estimate of that capacity. I have therefore analysed the plasma antioxidant capacity of a well-studied bird using two common assays (OXY-absorbent test and FRAP test) together with the plasma concentration of uric acid.
Materials and methods
A sample of blood was taken from the brachial vein of 26 male adult zebra finches Taeniopygia guttata (2–3 years old) using microhaematocrit heparinized capillary tubes (Vetlab Supplies Ltd, Broomers Hill Park, Pulborough, West Sussex, UK). Blood samples were maintained on ice until centrifugation, which occurred 2 h later. Plasma samples were stored at −70°C. Laboratory analyses were carried out 2 months later. Three samples were not included in the laboratory analyses because the plasma was turbid.
The OXY-adsorbent test (Diacron International, Grosseto, Italy) quantifies the ability of the plasma non-enzymatic antioxidant compounds to cope with the in vitro oxidant action of hypochlorous acid (HOCl; an oxidant endogenously produced). This assay is able to quantify the contribution of several kinds of antioxidants because HOCl can, for example, react with proteins, thiols, ascorbate, vitamin E and carotenoids. The procedure was carried out as in previous studies (Costantini et al. 2006; Costantini & Dell’omo 2006a,b). Plasma samples were diluted 1 : 100 with distilled water. A 200-μL aliquot of HOCl solution was incubated with 5 μL of the diluted plasma for 10 min at 37°C. The same relative volumes were used for the reference standard and blank (i.e. water). At the end of incubation, 5 μL of the chromogen N,N-diethyl-p-phenylenediamine was added. An alkyl-substituted aromatic amine dissolved in the chromogen is oxidized by the residual HOCl and transformed into a pink derivative. The intensity of the coloured complex is inversely related to the total plasma antioxidant capacity. The absorbance was read with a Thermo Scientific Multiskan Spectrum (ThermoFisher, Vantaa, Finland) at a wavelength of 490 nm (similar results were obtained at 505 and 546 nm). Measurements are expressed as mM of HOCl neutralized according to the following formula:
where Abs indicates absorbance of the values and Std the reference standard. Similar results were obtained using a calibration curve as a reference. Analyses were run in duplicate, and the mean coefficient of variation between replicate assays was 4·48%.
The FRAP test quantifies the ferric reducing ability of serum or plasma, so providing an estimate of the non-enzymatic antioxidant power (Benzie & Strain 1996). Ferric to ferrous ion reduction at low pH causes a coloured ferrous–tripyridyltriazine complex to form, whose intensity is directly related to FRAP potential. On the analysis day, FRAP reagent and standard calibration solution were freshly prepared. The FRAP reagent is composed by a solution of Fe3+ solution, 10 mM 2,4,6-tripyridyl-s-triazine and acetate buffer (pH 3·6) in a volume ratio of 1:1:10. For the reference standard, 0·0834 g of ferrous sulphate heptahydrate (FeSO4 × 7H2O) was dissolved in 100 mL of distilled water to have a final concentration of 3 mM Fe2+. Plasma samples were diluted 1 : 1 with distilled water. Five microlitres of the diluted plasma was pipetted into wells (96-well plates; Corning Incorporated, Corning, NY, USA) with 15 μL of distilled water and 150 mL of FRAP reagent. The same relative volumes were used for the calibration curve and blank. The reaction was incubated for 20 min, and the absorbance was read with a Thermo Scientific Multiskan Spectrum at a wavelength of 593 nm. FRAP values are expressed as mM Fe2+. Analyses were run in duplicate, and the mean coefficient of variation was 3·26%.
Plasma uric acid concentration
The plasma concentration of uric acid was measured by Uric Acid Assay kit (BioVision Research Products, Linda Vista Avenue, Mountain View, CA, USA). In each well (96-well plates, Corning Incorporated), 5 μL of plasma, 45 μL of uric acid assay buffer, and 50 μL of reaction mix (46 μL uric acid assay buffer, 2 μL uric acid probe, 2 μL uric acid enzyme mix) were pipetted. The same relative volumes were used for the calibration curve and blank. The reaction was incubated for 30 min, and the absorbance was read with a Thermo Scientific Multiskan Spectrum at a wavelength of 570 nm. Analyses were run in duplicate, and the mean coefficient of variation was 2·25%.
Statistical analyses were performed using spss Version 15.0 (SPSS Inc., Chicago, Illinois, USA). I used Pearson correlation to evaluate how the values of OXY, FRAP and uric acid concentration correlated with each other. Given that FRAP and uric acid were strongly correlated , I calculated the residual FRAP from a linear regression of FRAP values on uric acid.
Plasma uric acid concentration was strongly correlated with the FRAP assay measurement of total antioxidant capacity (r = 0·91, n = 23, P < 0·001; Fig. 1a), but was not correlated with the equivalent measurement from the OXY test (r = 0·24, n = 23, P = 0·15; Fig. 1b). OXY and FRAP values were not significantly correlated (r = 0·23, n = 23, P = 0·17; Fig. 1c), nor were OXY measurements correlated with residual values for FRAP after controlling for uric acid concentrations (r = 0·02, n = 23, P = 0·47; Fig. 1d).
The results of this study show that the correlation between plasma uric acid concentration and plasma antioxidant capacity is method dependent. In general, it cannot be concluded that a correlation between uric acid and plasma antioxidant capacity, whatever method is used, indicates that uric acid is actively contributing to that capacity. In fact, such a correlation could simply indicate a pattern of independent covariation between the two variables. For the FRAP test, however, it can be stated that the correlation between antioxidant capacity and uric acid concentration is really mirroring the ferric reducing ability of uric acid itself. In fact, Benzie & Strain (1996) show that uric acid is the molecule that most strongly contributes to total FRAP in serum or plasma. On the other hand, the lack of correlation between OXY and uric acid is in line with recent results obtained on captive pigeons Columba livia domestica (Costantini 2010a) and wild marmots Marmota marmota (D. Costantini et al. in preparation). It is therefore clear that it is not necessary to correct OXY values for uric acid concentration.
Assays that measure the antioxidant capacity of a biological matrix are many and none is ideal. Most importantly, assays can measure different components of the antioxidant system (Prior & Cao 1999; Del Rio, Serafini, & Pellegrini 2002), and so a combination of different assays could improve our ability to estimate the real antioxidant capacity of that matrix. In contrast to the FRAP assay, the OXY test uses a pro-oxidant (HOCl) that is endogenously produced by organisms. The OXY test is not specific and can measure the contribution of both hydrophilic and lipophilic antioxidants. HOCl can react with several kinds of antioxidants (e.g. vitamins A and E, ascorbate; see e.g. Halliwell & Gutteridge 2007), but the values obtained by the OXY test do not always correlate with concentrations of a specific antioxidant. For example, Palleschi et al. (2007) found that in humans OXY may be positively correlated with the concentration of total thiols (r = 0·65–0·67) and vitamins A (r = 0·59) and C (r = 0·43), but this correlation is not always significant. Mixed results were also found for the correlation between OXY and circulating thiols in birds (Costantini 2010a,b; Costantini & Bonadonna 2010; Costantini, Carello, & Fanfani 2010). These data suggest that the contribution of one antioxidant class to OXY could vary in relation to the concentration of other antioxidants present in the sample and to their antioxidant potential.
The FRAP assay involves neither a pro-oxidant nor an oxidizable substrate. Instead, it measures the ability of a compound to reduce Fe3+ to Fe2+. Although it is theoretically possible that the Fe2+ ion reacts in vivo with H2O2 or hydroperoxides to produce new free radicals, the FRAP test provides an indirect estimate of the antioxidant capacity, which however is very repeatable and reproducible. The results of this study suggest that the OXY and FRAP tests can provide independent results of plasma antioxidant capacity, which can be evaluated together to produce a more rounded picture of antioxidant defences. It may be possible, however, that the two methods may produce correlated results under different circumstances than those of the present study (i.e. adult male birds that never reproduced).
Another assay commonly used for the quantification of the plasma or serum antioxidant capacity is the ORAC assay. This test responds to numerous antioxidants present in the biological matrix (e.g. albumin, vitamin E, ascorbic acid; Cao & Prior 1998) and like the OXY test quantifies the capacity of antioxidants to cope with the in vitro activity of a pro-oxidant. Results from the ORAC assay have been found to correlate weakly with FRAP values (r = 0·35; Cao & Prior 1998) and strongly with OXY values (r = 0·98; Bonanni et al. 2007).
Finally, I recognize that measuring a biomarker of plasma antioxidant capacity provides a reductionist modelling of the in vivo situation, and therefore, caution is needed in the interpretation of results. Furthermore, the extent to which the antioxidant capacity of a tissue reflects that of another in the same individual is still to be fully determined. Blood is routinely the only tissue available in non-terminal ecological research. Considering that the contribution of antioxidant enzymes to biomarkers of plasma antioxidant capacity is likely to be negligible to absent, results should be interpreted as reflecting only circulating antioxidants, such as vitamins, proteins, bilirubin or carotenoids, as previously suggested by other authors (Cohen, Klasing, & Ricklefs 2007).
In conclusion, this study shows that the correlation between plasma uric acid and plasma antioxidant capacity is method dependent. Therefore, to determine whether values of antioxidant capacity need to be corrected for the uric acid concentration, it is important to take into account the biochemical properties of the assay used. This study also suggests that the combination of different assays for the quantification of antioxidant capacity could help to produce a better estimate of tissue antioxidant status.
I thank two anonymous reviewers for valuable comments that helped me improve the manuscript, Neil Metcalfe for valuable comments on the manuscript and for improving the English, all members of the technician staff for taking care of animals and William Mullen for advice on the FRAP method. I was supported by a NERC postdoctoral research fellowship (NE/G013888/1).