Hemolysis Assessment and Antioxidant Activity Evaluation Modified in

Feb 6, 2014 - ABSTRACT: Hydrogen peroxide (H2O2)-induced hemolysis is a commonly used model for antioxidant activity evaluation, and the hemolysis ...
0 downloads 0 Views 1MB Size
Download PDF
Article pubs.acs.org/JAFC

Hemolysis Assessment and Antioxidant Activity Evaluation Modified in an Oxidized Erythrocyte Model Xin Xu,*,∥ Jiayi He,∥ Guoyan Liu, Xinyu Diao, Yingying Cao, Qun Ye, Guangxin Xu, and Wendong Mao College of Food Science and Engineering, Hanjiang District, Yangzhou University, 196 Huayang Western Road, Yangzhou 225127, China ABSTRACT: Hydrogen peroxide (H2O2)-induced hemolysis is a commonly used model for antioxidant activity evaluation, and the hemolysis index is often presented as the absorbance of supernatant hemoglobin (Hb), which releases from injured cells. However, in previous studies, as an oxidation-sensitive protein, it has been recognized that Hb easily forms other types or substances, such as metHb, Heinz, and some fluorescent products. This study concerns whether Hb oxidation participated in H2O2-induced hemolysis and confirmed that the destruction of Hb under oxidizing condition had been a novel interfering factor that could reduce the absorbance in Hb quantitative detection. To correct the lower absorbance, the stable fluorescent products found in Hb degradation were selected, and the absorbance correction factor of 6.436 was drawn on the basis of the absorbance and fluorescent intensity. This correction factor obviously altered the results of both dose-dependent hemolysis of H2O2 and antioxidant activity. In addition, the assessment difference was innovatively discussed by altering the sequences of adding antioxidant and oxidant. These different sequences caused variations in hemolysis, indicating that multiple evaluations may be related to the antioxidant pathways which are necessary for more accurate bioactivity data. KEYWORDS: hemolysis, indices correction, fluorescence, adding order

■

INTRODUCTION The erythrocyte is a unique cell model with simple metabolism and sensitivity to oxidation. As a representative biological assay, oxidative hemolysis is widely used and is often combined with other chemical methods (e.g., ORAC, TEAC, and DPPH assays) in the evaluation of antioxidant activity.1,2 In vivo simulation is receiving much more attention and needs to be improved in method establishment, selection, and modification.3 Substrate, dose, reaction time, and many other factors have been considered by previous research. In terms of antioxidant activity, erythrocyte models also have been improved on the basis of physiological and pathological responses. The most common models were created with different free radicals to mimic endogenous oxidation, such as those using H2O2, APPH, hypochlorite, and peroxynitrite.4−7 Different indices for antioxidant have been developed and gradually applied. These indices include the fluorescent intensity of intracellular free radicals, which are captured by some fluorescent probes such as DCFA-DA,8,9 the ratio of intracellular methemoglobin (MetHb) reduction for reflecting the reduced ability of cytosolic antioxidants,10 the turbidity of erythrocyte suspension for evaluating the degree of hemolysis in high throughput microplates,11 and the inhibition of intracellular Heinz body formation for reflecting the oxidative denaturing of hemoglobin (Hb) and predicting hemolysis.12 Moreover, the flux of material addition was introduced in the assessment of the erythrocyte tolerance to oxidative stress by adjusting oxidant dose over different periods of time.13 When erythrocytes serve as an antioxidant model, many other factors should be considered. Index measurement is a strict step for activity evaluation. Hb content in the hemolytic supernatant is usually assessed and determined by spectrophotometry. However, Hb is highly © 2014 American Chemical Society

responsive to oxidative damage and usually transforms into other products, such as MetHb, Heinz bodies, and colorless hemichrome,14 which will increase the variation in Hb absorbance value. Because oxidants, such as H2O2, can easily enter cells and that most oxidants can trigger the cellular oxidation cascade, cell oxidative injury may accompanied with Hb oxidation. If damage causes Hb to turn into colorless products in the erythrocyte model, the quantitation for released Hb by spectrophotometry will be reduced. In this study, we would focus on this potential oxidative response of Hb and determine how much interference the color fading does on the assessment of oxidized hemolysis. In a cell model, cell membrane is the first place that exogenous oxidants arrive to damage and exogenous antioxidants arrive to protect. On the one hand, for presenting an ideal protective activity, an exogenous antioxidant is often assessed especially with a preculture model according to the reported studies, including the hemolysis test. On the other hand, a membrane that serves as a protective barrier is also the bilateral target that can be attacked by both intracellular and extracellular oxidants. Evaluation results obtained from the preculture model cannot reveal the actual effects of antioxidants, such as repair capacity, on those cells that are damaged by extra- or intraoxidants first, because in antioxidant preculture model, there is no particular time for oxidants to damage cells and even no chance for oxidants to enter the cells. We assumed that these potential responses might be related to the reaction order or antioxidant position as shown in Figure 1 Received: Revised: Accepted: Published: 2056

November 6, 2013 February 4, 2014 February 6, 2014 February 6, 2014 dx.doi.org/10.1021/jf4049935 | J. Agric. Food Chem. 2014, 62, 2056−2061

Journal of Agricultural and Food Chemistry

Article

plate, and its fluorescent intensity (FI) was read at the excitation wavelength of 355 nm and the emission wavelength of 460 nm with a Fluoroskan Ascent FL plate reader (Thermo Scientific, USA). Quantitation of Ferric Ions Released from Hb Oxidation by Inductively Coupled Plasma Spectrometry. Ferric Ion Extraction. The extraction method was as follows. The Hb reaction mixture was separated through an ultrafiltration membrane (5000 Da) at 4 °C (Vivaflow 200, Sartorius, Goettingen, Germany), and the filtrate was collected. A 10% (v/v) TCA solution was mixed with the Hb reaction solution at 4 °C for 30 min, and then the supernatant was collected. Finally, the Hb reaction mixture was heated in a water bath at 100 °C for 1 h, and the supernatant was collected. Ferric Ion Detection. Extracted ferric ions were added to a solution of HNO3/HClO4 (v/v, 4:1). The digestion lasted for 2 h at 140 °C, and the solution was heated again at 190 °C to remove acid. The digested ferric ion solution was diluted with ultrapure water, and then the ferric ions were detected by Elan DRC-e inductively coupled plasma spectrometry (Perkin-Elmer, USA). Multiple Assessment of Antioxidant Activity with Hemolysis Tests. A 100 μL solution of erythrocytes was treated with H2O2 of 166.5 μmol/mL of the final concentration and one antioxidant liquor in each test group. The selected antioxidants were GSH of 1.67 mmol/ L, L-c of 1.67 mmol/L, Cys of 1.67 mmol/L, and IWHDC of 0.167 mg/mL, respectively. Both the H2O2 and antioxidant were dissolved in PBS and cultured with erythrocytes at 37 °C for several hours. Three different sequences of antioxidant addition were examined in this work. The first kind of reaction consisted of two steps and was denoted R1. The erythrocytes were preincubated with antioxidant for 15 min and then combined with H2O2 for 4 h to accomplish hemolysis. The second kind of reaction also included two steps and was denoted R2. Antioxidant and H2O2 were fully blended for 15 min prior to dispensing the erythrocytes solution, and the mixture obtained was allowed to sit for 4 h. The third kind of reaction, denoted R3, was slightly different from R1 when H2O2 was added. Briefly, H2O2 was mixed with the erythrocyte solution to initiate the oxidative stress of cells, and after 5 min, antioxidant was injected immediately to inhibit the oxidative hemolysis, and the mixture was allowed to sit for 4 h. Additionally, H2O2 -induced hemolysis in the absence of all antioxidants was performed as the injury control group, and the solution was also maintained for 4 h, during which an equal volume of PBS or other solvent was added instead of the antioxidant liquor. All solutions were freshly prepared and used immediately. Statistical Analysis. Statistical analysis was performed with Microsoft Excel. Statistical significance was tested using the pairedsamples t test. Significant difference between data sets was assessed by a p value of 1 h, and the Hb concentration was at a higher level. Thus, we compared the results of the following tests, one in which the Hb of 0.138 mg was oxidized for 15 min, and the other in which the Hb of 8 mg (H2O2 doses were the same as in the first test) was oxidized for 90 min (Figure 5). In Figure 5B, the value of FI declined when the dose of H2O2 was higher. However, when the reaction time was extended from 90 min to 24 h, the fluorescent products were found to be stable with no significant variation in the FI value. This revealed that many kinds of fluorescent products were generated during Hb oxidized degradation; some of the products could be further degraded, and some were stable, which would be useful for correcting the Hb absorbance. As shown in Figure 5A, the FI value did not decrease at any dose of H2O2, whereas in Figure 5B, a significant decrease of FI was observed at higher doses of H2O2. This difference might result from two situations. One was that no sufficient fluorescent products could be further decomposed during Hb degradation so that there was little effect on the FI value, as shown in Figure 5A. The other was that there was not enough time for Hb to completely degrade. Therefore, to better imitate in vivo reaction, the process in Figure 5B would be implemented in the following study.

Figure 3. Oxidation of free Hb measured with spectrophotometry (solid line) and fluorometry (dashed line). Hb was incubated with various doses of H2O2 for 30, 60, and 90 min.

were other reactions during the oxidation of Hb, for example, color fading. The reducing Abs suggested Hb loss at higher doses of H2O2. The decrease of Abs in Figure 2 probably resulted from the further oxidative destruction of MetHb into hemichrome or other degradation products.14 Curves D1 and D2 in Figure 2 showed that the Abs values at 0 μmol/L of H2O2 were both above 0.1, whereas the Abs at the same abscissa on curve D3 was 0.052 (the supernatant was nearly colorless). This indicated that there was a little free Hb at the beginning of the hemolysis test, and this part of Hb might be attacked first and then degraded by H2O2. To prove whether the initial free Hb presents interferences with the determination of hemolysis, we attempted to add Hb solution to the erythrocyte suspension (the supernatant of this suspension with Abs405 of 0.07 was also nearly colorless) before the hemolysis test. The results coincided with the assumption, showing that in the suspension with extra Hb added, Abs at 2058

dx.doi.org/10.1021/jf4049935 | J. Agric. Food Chem. 2014, 62, 2056−2061

Journal of Agricultural and Food Chemistry

Article

Figure 5. Oxidation curves of free Hb measured with spectrophotometry (solid line) and fluorometry (dashed line). (A) Hb (0.138 mg) was incubated with H2O2 for 15 min, and the relative doses of H2O2 were 0, 1.81, 5.43, 18.10, 180.98, and 361.96 μmol/mgHb. (B) Hb (8 mg) was incubated with H2O2 for 15 min, and the relative doses of H2O2 were 0, 6.12, 61.2, 183.64, and 306.07 μmol/mgHb.

Figure 6. Different assays to extract the ferric ions released from Hb. Hb (8 mg) was incubated with different concentrations of H2O2 for 90 min. The ferric ions were extracted by ultrafiltration membrane (A), boiled water bath, and TCA precipitation (B), respectively.

It was found that Hb degradation not only generated fluorescent products but also induced ferric ion release from heme. We studied this release in three different ways (Figure 6). In the ultrafiltration test, the number of ferric ions after digestion was significantly higher compared with nondigestion (Figure 6A), which could indicate that the released ferric ions were binding with other protein substances. The number of released ferric ions extracted by the TCA and heating assays was significantly higher than that extracted by the ultrafiltration assay (Figure 6B). It was possible that some ferric ion conjugates were trapped by the 5000 Da ultrafiltration membrane due to the decrease in flux or that the ferric ions binding to other substances could not be extracted in this way. At the same time, all three assays showed that the highest ferric ion content was obtained when the H2O2 concentration was 183.6 μmol/mgHb, which was different from the FI values. This explained coincidentally why the decrease of FI values happened at a higher dose of H2O2 (183.6 μmol/mgHb), because there was further degradation. Correction of Abs in Hemolysis. The above discussion indicated that the oxidative degradation of Hb was the key interfering factor in hemolysis evaluation. The following description was about establishing the model of Hb oxidation and correcting the lost Hb detection beyond the visible spectrum. Normal erythrocytes contain 95% Hb, so in either in vivo or in vitro hemolysis, the dose of H2O2 is relatively little and can be calculated in units of micromoles per milligram of Hb. As shown in Figure 7, the H2O2 dose range was between 0 and 61 μmol/mg Hb. Abs and FI showed a good complementary linear relationship. We found that the actual concentration of Hb

Figure 7. Relationship between Abs and FI in the H2O2-induced degradation of Hb.

could be expressed as Hbactual = FI/CC + Abs if FI values divided a constant value (noted as the correction coefficient (CC)). According to the law of conservation of mass, the total Hb content must remain constant before and after degradation. The Abs-M curve in Figure 7 is the calibration curve of Hb with the CC value of 6.436. It can be seen that the Abs of Hb after modification by FI remained close to the initial level with increasing doses of H2O2. This confirmed that the correction of absorbance was effective in hemolysis on the basis of the stable fluorescent products. Effect of the Sequence of Adding Antioxidants on the Assessment of Antioxidant Activity. The previous sections had modified hemolysis measurement, but here we wondered how to improve the protocol of the cell model to imitate the physical oxidation and antioxidation. The widely applied 2059

dx.doi.org/10.1021/jf4049935 | J. Agric. Food Chem. 2014, 62, 2056−2061

Journal of Agricultural and Food Chemistry

Article

Figure 8. Antioxidant activity of IWHDC, GSH, Cys, and L-c under hemolysis induced by 166.5 μmol/L of H2O2: (A) results without modification; (B) results modified by correction factor for FI. Hemolysis reaction was conducted using three different sequences of adding antioxidants (R1, R2, R3), and the control group contained no antioxidant. Each value is the mean ± SD of three experiments. Significant differences are indicated by a, b, and c: a, compared with R1 and R2, p < 0.05; b, compared with R1 and R3, p < 0.05; c, compared with R2 and R3, p < 0.05.

were added in three different sequences. The results showed differences in both the control group and the antioxidant groups for all three sequences may be related to cell resistance to oxidation, membrane transport capacity, and the other effect on cellular metabolism. According to the R1, R2, and R3 groups, more activity pathways can be analyzed, and results from them would have considerable value in the initial stage of antioxidant development. In future studies to find the intracellular oxidation level and explain the membrane roles, we will describe the detail mechanisms observed via fluorescent tracer, Western blot, and Fourier transform infrared spectroscopy techniques.

sequence R1, preincubation with antioxidant and erythrocytes,19,20 reflects the comprehensive activity of exogenous antioxidants because redundant antioxidants can also work outside the cells. Sequence R2, simulating the role of antioxidants on extracellular free radical scavenging, is actually a part of sequence R1 when antioxidants enter the cells.21 Sequence R3 is also a part of R122 and reflects the ability to inhibit free radical chain reaction and repair injury when erythrocytes are in a state of unremitting oxidative stress. These three sequences were tested separately here, but in human blood, they always exist simultaneously and synergistically. Therefore, multidimensional analysis may be better for evaluating antioxidant capacity based on erythrocyte models. Three recognized antioxidants (GSH, Cys, and L-c) and a synthetic peptide (the amino acid sequence IWHDC designed by our laboratory) were evaluated on the basis of their order of addition in this study. First, we compared the difference between the results of unmodified hemolysis (Figure 8A) and those modified by FI (Figure 8B). The antihemolysis activities changed when Abs was modified by FI; therefore, we chose Abs modified by FI for antioxidant activity evaluation. The three sequences (R1, R2, and R3) showed different hemolysis results not only in the antioxidant groups but also in the control group. The difference observed in the control group may be related to the tolerance of erythrocytes to oxidative stress, whereas the differences observed in antioxidant groups may also be related to the antioxidant mechanism and membrane transfer ability in the samples. As an amino acid antioxidant, Cys showed the highest activity in each of the three evaluation sequences, mainly because the rich disulfide bond provided protection from oxidation. The results of the R3 sequence showed that Cys may have entered the cell and controlled intracellular oxidative damage; the R2 sequence showed no activity, which could be directly related to the lack of scavenging H2O2 (data not shown). GSSH generated in the R2 sequence may disrupt the extracellular GSSH/GSH ratio such that the red blood cells exhibit abnormalities of metabolism. The poor performance of L-c in this study may be due to an inappropriate dose. Pentapeptide IWHDC had H2O2-scavenging capacity (data not shown) and showed antioxidative capability in the R1 and R3 groups, but not in the R2 group, suggesting that the oxidation products of IWHDC may disrupt cellular metabolism and reduce antihemolytic activity. On the basis of the modified hemolysis measurement, we investigated how variation in activity assessment if antioxidants

■

AUTHOR INFORMATION

Corresponding Author

*(X.X.) Phone: +86-1895-2731677. Fax: +86-0514-87313372. E-mail: [email protected]. Author Contributions ∥

X.X. and J.H. contributed equally to this study and share first authorship.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We are grateful to our volunteers, Mao WD, Guo ML, and Zhang JY, and to all of the students in the food safety major of the class of 2010 who provided their blood samples for hemolysis testing. This work was supported by Yangzhou University Science and Technology Innovation Fund (2013CXJ085).

■

REFERENCES

(1) Honzel, D.; Carter, S. G.; Redman, K. A.; Schauss, A. G.; Endres, J. R.; Jensen, G. S. Comparison of chemical and cell-based antioxidant methods for evaluation of foods and natural products: generating multifaceted data by parallel testing using erythrocytes and polymorphonuclear cells. J. Agric. Food Chem. 2008, 56, 8319−8325. (2) Buehler, P. W.; Alayash, A. I. Redox biology of blood revisited: the role of red blood cells in maintaining circulatory reductive capacity. Antioxid. Redox Signal. 2005, 7, 1755−1760. (3) Frankel, E. N.; Meyer, A. S. The problems of using onedimensional methods to evaluate multifunctional food and biological antioxidants. J. Sci. Food Agric. 2000, 80, 1925−1941. (4) Stocks, J.; Dormandy, T. L. The autoxidation of human red cell lipids induced by hydrogen peroxide. Br. J. Hamaetol. 1971, 20, 95− 111.

2060

dx.doi.org/10.1021/jf4049935 | J. Agric. Food Chem. 2014, 62, 2056−2061

Journal of Agricultural and Food Chemistry

Article

(5) Landi, L.; Cipollone, M.; Cabrini, L.; Fiorentini, D.; Farruggia, G.; Galli, M. C. Injury of rat thymocytes caused by exogenous peroxyl radicals in vitro. Biochim. Biophys. Acta 1995, 1239, 207−212. (6) Vissers, M. C. M.; Stern, A.; Kuypers, F.; Berg, J. V. D.; Winterbourn, C. C. Membrane changes associated with lysis of red blood cells by hypochlorous acid. Free Radical Biol. Med. 1994, 16, 703−712. (7) Wrobel, A.; Lukaszynska, B.; Kedzierska, J. The effect of peroxynitrite and some antioxidants on the rate of osmotic hemolysis of bovine erythrocytes. Cell. Mol. Biol. Lett. 2003, 8, 445−460. (8) Wolfe, K. L.; Liu, R. H. Cellular antioxidant activity (CAA) assay for assessing antioxidants, foods, and dietary supplements. J. Agric. Food Chem. 2007, 55, 8896−8907. (9) Blasa, M.; Angelino, D.; Gennari, L.; Ninfali, P. The cellular antioxidant activity in red blood cells (CAA-RBC): Aa new approach to bioavailability and synergy of phytochemicals and botanical extracts. Food Chem. 2011, 125, 685−691. (10) Arbos, K. A.; Claro, L. M.; Borges, L.; Santos, C. A. M.; WeffortSantos, A. M. Human erythrocytes as a system for evaluating the antioxidant capacity of vegetable extracts. Nutr. Res. (N.Y.) 2008, 28, 457−463. (11) Takebayashi, J.; Iwahashi, N.; Ishimi, Y.; Tai, A. Development of a simple 96-well plate method for evaluation of antioxidant activity based on the oxidative haemolysis inhibition assay (OxHLIA). Food Chem. 2012, 134, 606−610. (12) Tunsaringkarn, T.; Soogarun, S.; Rungsiyothin, A.; Palasuwan, A. Inhibitory activity of Heinz body induction in vitro antioxidant model and tannin concentration of Thai mimosaceous plant extracts. J. Med. Plants Res. 2012, 6, 4096−4101. (13) Guitton, J.; Servanin, S.; Francina, A. Hexose monophosphate shunt activities in human erythrocytes during oxidative damage induced by hydrogen peroxide. Mol. Toxicol. 2003, 77, 410−417. (14) Nagababu, E.; Rifkind, J. M. Formation of fluorescent heme degradation products during the oxidation of hemoglobin by hydrogen peroxide. Biochem. Biophys. Res. Commun. 1998, 247, 592−596. (15) Hani, A.; Hagai, G. Heme degradation in the presence of glutathione. J. Biol. Chem. 1995, 270, 24876−24883. (16) May, J. M.; Qua, Z. C.; Qiao, H.; Koury, M. J. Maturational loss of the vitamin C transporter in erythrocytes. Biochem. Biophys. Res. Commun. 2007, 360, 295−298. (17) Lochs, H.; Morse, E. L.; Adibi, S. A. Uptake and metabolism of dipeptides by human red blood cells. Biochem. J. 1990, 271, 133−137. (18) Jarolim, P.; Lahav, M.; Liu, S. C.; Palek, J. Effect of hemoglobin oxidation products on the stability of red cell membrane skeletons and the associations of skeletal proteins: correlation with a release of hemin. Blood 1990, 76, 2125−2131. (19) Paiva-Martins, F.; Fernandes, J.; Santos, V.; Silva, L.; Borges, F.; Rocha, S.; et al. Powerful protective role of 3,4-dihydroxyphenylethanol−elenolic acid dialdehyde against erythrocyte oxidative-induced hemolysis. J. Agric. Food Chem. 2012, 28, 135−140. (20) Serem, J. C.; Bester, M. J. Physicochemical properties, antioxidant activity and cellular protective effects of honeys from southern Africa. Food Chem. 2012, 133, 1544−1550. (21) Manna, C.; Galletti, P.; Cucciolla, V.; Montedoro, G.; Zappia, V. Olive oil hydroxytyrosol protects human erythrocytes against oxidative damages. J. Nutr. Biochem. 1999, 10, 159−165. (22) Okoko, T.; Ere, D. Reduction of hydrogen peroxide−induced erythrocyte damage by Carica papaya leaf extract. Asian Pac. J. Trop. Biomed. 2012, 2, 449−453.

2061

dx.doi.org/10.1021/jf4049935 | J. Agric. Food Chem. 2014, 62, 2056−2061