Introduction
Cellular oxidative status represents a balance of antioxidant molecules, primarily reduced and oxidized glutathione; antioxidant enzymes such as superoxide dismutase, catalase, glutathione reductase, and glutathione peroxidase; and oxidant molecules, primarily ROS.1,2 Reactive oxygen species are continuously produced by cells as a byproduct of mitochondrial metabolism.2 In health, ROS are readily neutralized by antioxidant molecules and enzymes, and oxidative stress occurs when antioxidant molecules or enzymes are depleted or ROS production increases.2 The production of ROS is increased by various pathophysiologic mechanisms, including immunologic diseases, hypoxia, excess iron, inflammation, drug metabolism, and deficiencies of antioxidant proteins, vitamins, or enzymes.2–5 Reactive oxygen species can induce damage to various membrane structures and enzymatic systems, primarily through their effects on DNA, proteins, and lipids, thus leading to cellular senescence and apoptosis.2,5 In dogs, oxidative stress has been identified with diabetes mellitus, chronic kidney disease, heart disease and congestive heart failure, and anemia.6–12 However, the methods used to evaluate oxidative stress are inconsistent and indirect, typically involving the use of various combinations of assays measuring small molecules, vitamins, heavy metals, enzyme activity, or byproducts of oxidative injury.
Erythrocytes are particularly susceptible to oxidative injury because they have a large number of intracellular hemoglobin molecules that bind oxygen.13 Hemoglobin may undergo auto-oxidation to methemoglobin, which creates an endogenous source of ROS, and the exposure of erythrocytes to other cells of the body increases the likelihood that erythrocytes will encounter exogenously produced ROS.13 Therefore, erythrocytes have developed intricate antioxidant and protective mechanisms to minimize their risk for oxidative injury. Hemoglobin in dogs and people has 2 readily oxidizable sulphydryl groups, such that the risk of ROS-induced erythrocyte damage is also similar.14 Also like people, anemic dogs have erythrocytic oxidative stress for which ROS are implicated, thereby leading to hemolysis and early senescence.8,9,14,15,16
Detection of ROS is appealing because it would allow a direct measure of oxidative status; however, ROS are difficult to measure in tissue because of their short half-life and high reactivity.17 Direct measurement of ROS in dogs is described,5 with its concentration primarily determined with methods that estimate total ROS through the free radical-scavenging ability of the tissue being analyzed.5,18 Yet these methods essentially provide a measurement of total antioxidant capacity, rather than a direct, specific measurement of ROS. In people, flow cytometry has been used to detect oxidative stress in various cells.17,19,20 A fluorochrome, DCFH-DA, is used to help measure ROS through fluorescence of its oxidized derivatives.13 This fluorochrome readily crosses the membranes of erythrocytes and is acetylated and cleaved by intracellular esterases, forming membrane-impermeable H2DCF.17 Oxidation of DCFH-DA by ROS yields the highly fluorescent DCF, which can be detected by fluorescence methods like flow cytometry.17,18 Studies that include measurement of ROS in canine erythrocytes have not been published. Therefore, the objective of the study reported here was to validate the use of DCFH-DA for the detection of ROS in the erythrocytes of healthy dogs.
Materials and Methods
Animals
Fifty skeletally mature healthy dogs were included. Dogs were determined to be healthy on the basis of physical examination and CBC results. Other than a preventive anthelmintic, dogs that received prescription medications, including exogenous antioxidants (combined S-adenosylmethionine and silybin,a S-adenosylmethionineb alone, milk thistle, omega-3 fatty acids, vitamins C or E, or a diet that included antioxidants and was purported to help dogs with joint or kidney disease), were excluded. Additionally, dogs that received an over-the-counter multivitamin were excluded. The study protocol was reviewed and approved by and conducted in accordance with the principles of the Purdue University Institutional Animal Care and Use Committee (protocol No. 1509001296).
Sample collection and preparation
Blood samples were collected from a jugular vein and immediately transferred to blood collection tubes containing EDTA.c On the basis of a published17,21 methodology, samples were allocated to 4 treatment groups as follows: DCFH-DAd plus H2O2d (30% wt/vol; stimulant of oxidative stress [stimulated]), DCFH-DA plus PBSS (nonstimulant [unstimulated]; PBSS [pH 7.4, 137mM NaCl,e 10mM Na2HPO4,d 2.7mM KCl,f and 1.8mM KH2PO4f]), DMSO plus H2O2, and DMSO plus PBSS. Hydrogen peroxide was diluted to a 20mM stock solution with deionized water, and DCFH-DA was diluted to a 50mM stock solution with DMSO. The concentrations of DCFH-DA and H2O2 varied for the specificity portion of the study to determine optimal concentrations. Once determined, consistent concentrations of DCFH-DA and H2O2 were used for the remaining portions of the study.
All samples (except those used to determine assay stability) were analyzed in triplicate within 2 hours of collection. Blood samples were centrifugedg at 3,000 × g for 5 minutes at 4°C. Plasma and buffy coats were removed with a Pasteur pipette. Then, 10 μL of erythrocytes was diluted in 5 mL of PBSSA (PBSS supplemented with bovine serum albuminc [1% wt/vol]). Next, 10 μL of DCFH-DA or DMSO was added to 5-mL round-bottom tubes.c Then, 100 μL of the erythrocyte solution was added to these tubes and incubated at 37°C for 20 minutes.
After incubation, 10 μL of PBSS or H2O2 was added, and solutions were incubated for 20 minutes at room temperature (20°C). Then, the solutions were diluted with 200 μX of 1% PBSSA and immediately analyzed by flow cytometry.h Reactive oxygen species-dependent fluorescence was detected by green fluorescence with an excitation wavelength of 488 nm and gating only around erythrocytes. A total of 50,000 events was collected in the erythrocyte gate for every analyzed sample in each portion of the validation. No data points that were outside of the predetermined erythrocyte gate were considered in the statistical analyses. Flow cytometric analysis was validated according to methods previously described.22
Gating
Fluorescence specificity for erythrocytes was determined on the basis of appropriate removal of contaminant cells, including leukocytes and platelets, and correct gating. For this purpose, 2 blood samples from each of 3 dogs were used; samples were collected into blood collection tubes containing EDTA.
The sample in the first tube was left undisturbed for 20 minutes at 20°C to allow the plasma to separate from the RBCs by gravity. Plasma was then retrieved with a Pasteur pipette, the remaining blood was centrifuged at 300 × g for 10 minutes at room temperature, and the supernatant of platelet-rich plasma was collected.
The sample in the second tube was used for preparation of leukocyte and erythrocyte solutions. First, the tube was centrifuged at 3,000 × g for 5 minutes at 4°C, and plasma was retrieved with a Pasteur pipette. Leukocytes were collected after the buffy coat was retrieved and transferred to tubes with 1 of 10 volumes of an erythrocyte lysis solution (0.15mM NH4Cl,d 10mM NaHCO3,d and 0.1mM EDTAd; pH 7.4). After incubation and continuous agitation at 20°C for 10 minutes, each solution was centrifuged at 3,000 × g for 5 minutes at 20°C, and the subsequent pellet was washed with the same volume of erythrocyte lysis solution. Incubation, continuous agitation, and centrifugation were repeated, and the supernatant was resuspended in 5 mL of PBSSA. For the erythrocyte solution, a pipette tip was passed into the pellet after the buffy coat was removed, and 10 μL of erythrocytes was retrieved. Then, the pipette tip was cleaned with a task wipe, and the retrieved erythrocytes were diluted in 5 mL of PBSSA.
The solutions of platelet-rich plasma, leukocytes, and erythrocytes were then analyzed in triplicate with flow cytometry to establish appropriate cellular gates, with the erythrocyte gate intended to be standardized for the remaining portions of the study. Platelet-rich plasma, leukocyte, and erythrocyte samples from 1 of the 3 dogs were analyzed by a commercial hematology analyzer,i and cytospin preparations were created and evaluated microscopically by a clinical pathologist to confirm the predominant cell type and that contamination was minimal.
To establish gating for erythrocytes positive for the fluorescent enzyme, each erythrocyte solution was divided into the 4 treatment groups as described previously, with samples first incubated with 10 μL of 50μM DCFH-DA or DMSO and then with 10 μL of 2mM H2O2 or 10 μL of PBSS. The samples were analyzed in triplicate by flow cytometry to detect ROS-dependent green fluorescence with an excitation wavelength of 488 nm and gating around only erythrocytes. The concentrations of DCFH-DA and H2O2 for this initial gate were determined on the basis of a previously described methodology.17,21 The unstimulated treatment group (DCFH-DA plus PBSS) in this gating experiment was used to establish expected autofluorescence of the erythrocytes.
Specificity
Assay specificity was determined in a dose-dependent manner by use of the procedure described above. Erythrocyte samples from the same 3 dogs whose samples were used for gating were subsequently incubated with 0 (DMSO only), 5, 10, 25, or 50μM DCFH-DA. Samples were stimulated with 0 (PBSS only), 0.5, 1, 2, or 4mM H2O2 and were analyzed in triplicate for MFI.
Precision
Stimulated (0.5mM H2O2) and unstimulated erythrocyte samples and samples with and without DCFH-DA (25μM) from 3 different dogs than those used for specificity determination were analyzed in triplicate to determine intra-assay precision. These erythrocyte samples were then prepared and analyzed 5 consecutive times to determine interassay precision. The 5 consecutive analyses were achieved by first preparing the samples as described previously and analyzing them in triplicate. Subsequent analyses were achieved by preparing new solutions from the same erythrocyte pellet with the protocol described previously, starting with the dilution of 10 μL of erythrocytes in 5 mL of PBSSA. A CV of ≤ 20% was considered acceptable for intra- and interassay precision.20
Stability
Stimulated (0.5mM H2O2) and unstimulated erythrocyte samples and samples with and without DCFH-DA (25μM) from 3 different dogs than those used for specificity and precision determination were analyzed in triplicate for baseline measurements. Erythrocyte plus PBSSA solutions were kept refrigerated (4°C) and were reanalyzed at 3, 6, 24, 36, and 48 hours after blood collection. Erythrocyte plus PBSSA solutions were stored; however, samples with DCFH-DA plus DMSO or H2O2 plus PBSS were prepared and immediately analyzed at each time point. The process of stimulation and incubation was repeated at each time point, and MFI was determined by flow cytometry. A CV of ≤ 20%, compared with baseline, was considered acceptable.20
Baseline
Erythrocytes from an additional 41 healthy dogs were collected for ROS measurement. Stimulated (0.5mM H2O2) and unstimulated samples with and without DCFH-DA (25μM) from these dogs were analyzed in triplicate for baseline measurements and expected normal MFI.
Statistical analysis
Data were exported from the flow cytometer and analyzed with commercially available software.j,k Median fluorescence intensity was determined, and the logarithmic mean of the 3 MFI values was used for statistical analysis with commercial software.l Normality was assessed by use of the Shapiro-Wilk test. Stimulated and unstimulated samples with increasing concentrations of DCFH-DA were compared by use of a repeated-measures 2-way ANOVA and post hoc Tukey test. Nonlinear regression with fractional polynomial comparisons of normalized data was used to compare the increasing concentration of DCFH-DA in the group with 4mM H2O2 only, and a regression line was identified for MFI and the percentage of DCF-positive cells. Coefficients of variations for precision and stability were calculated. A range of MFI for unstimulated erythrocytes with DCFH-DA from all 50 dogs was calculated by use of a robust method.23 This method used a bootstrapping technique with 10,000 iterations to determine 90% CIs for the lower and upper limits of the reference interval. For all 50 dogs, the Wilcoxon signed rank test was used to compare stimulated and unstimulated samples. Values of P < 0.05 were considered significant.
Results
Animals
Of the 50 dogs included in this study, 27 were female (sexually intact, n = 8; neutered, 19) and 23 were male (sexually intact, 4; neutered, 19), and the median age was 5 years (range, 0.83 to 13 years). Mixed-breed dogs (n = 14) were the most common dog type but 18 breeds were represented, with several that included ≥ 2 dogs (Golden Retriever, n = 6; Australian Shepherd, 5; Rhodesian Ridgeback, 4; Beagle, 3; Pembroke Welsh Corgi, 3; German Shorthaired Pointer, 2; and Labrador Retriever, 2). Baseline CBC data for the 50 dogs were summarized (Supplementary Table S1, available at: avmajournals.avma.org/doi/suppl/10.2460/ajvr.82.5.343).
Gating
Erythrocytes, leukocytes, and platelets were identified as distinct cellular clouds in flow cytometric scatterplots, and their identities were confirmed by hematologic analysis and microscopic evaluation of a blood smear by a clinical pathologist (Figure 1). The cytospin preparation of the RBC pellet predominantly consisted of erythrocytes, with only rare nondegenerate neutrophils and no platelets. The buffy coat predominantly consisted of leukocytes (60% nondegenerate neutrophils, 30% small mature lymphocytes, 7% macrophages, and 3% eosinophils) with minimal erythrocyte contamination and rare platelets. The platelet-rich plasma sample predominantly consisted of small to moderately sized aggregates of platelets with minimal erythrocyte contamination and no leukocytes.
Representative flow cytometric output with logarithmic forward (FSC-A) versus side scatter (SSC-A) plots showing the gates created for a sample from 1 of 50 healthy dogs. A—Depiction of the gates for erythrocytes (blue dots), cells of the buffy coat (mostly leukocytes [red dots]), and cells of the platelet-rich plasma (mostly platelets [orange dots]). B—Scatterplot derived from an erythrocyte sample, with concentric scatter representing the concentration of cells that is lowest at the periphery and highest in the center. This sample also has small numbers of leukocytes and platelets. C—Scatterplot derived from a platelet-rich plasma sample containing mainly platelets and likely a few small erythrocytes. D—Scatterplot derived from a buffy coat sample containing leukocytes (circled), erythrocytes that may be fragmented, and possibly remaining platelets.
Citation: American Journal of Veterinary Research 82, 5; 10.2460/ajvr.82.5.343
On initial flow cytometric gating, platelets were identified as the smallest events and leukocytes as the largest events (Figure 1). In the platelet scatterplot, a few erythrocyte contaminants were seen. In scatterplots of leukocytes, platelets in the buffy coat and a few erythrocyte contaminants were noted. Erythrocytes were isolated and plotted successfully as a single cloud of events larger than the platelets but smaller than the leukocytes. In the scatterplot for erythrocytes, minimal overlap was noted with leukocytes. However, the cloud of erythrocyte events was still distinct. The unstimulated samples with DCFH-DA had an increase in fluorescence, compared with the unstimulated samples without DCFH-DA, likely attributable to endogenous intracellular ROS generation (Figure 2). The gate for determining positivity for DCF was set for any fluorescence greater than this level of endogenous fluorescence.
Histogram for the logarithmic autofluorescence (FLI-A) of unstimulated samples (PBS) from 3 healthy dogs incubated with (blue spike) or without (red spike) 25μM DCFH-DA.
Citation: American Journal of Veterinary Research 82, 5; 10.2460/ajvr.82.5.343
Specificity, precision, and stability
The MFI of treated cells significantly (P < 0.001) increased as concentrations of DCFH-DA increased (Table 1). In the nonlinear regression analysis, the effect of DCFH-DA concentration on MFI and the percentage of cells positive for DCF when stimulated with 4mM H2O2 plateaued for concentrations of DCFH-DA > 25μM (MFI, 25μM vs 0μM [P < 0.001]; DCF-positive cells, 25μM vs 0μM [P = 0.07]; Figure 3). Therefore, 25μM DCFH-DA was used for the remaining portions of the study. The MFI of cells incubated with DCFH-DA significantly (P < 0.001) differed between those that were unstimulated and those that were stimulated with 0.5mM H2O2, but this significance was not detected with increasing concentrations of H2O2; therefore, 0.5mM was used for the remaining portions of the study.
Nonlinear regression curves depicting the effect of DCFH-DA concentration on MFI (A) and the percentage of cells positive for DCF (B) as determined with flow cytometry and triplicate analysis of samples of erythrocytes obtained from 3 healthy dogs and stimulated with 4mM H2O2
Citation: American Journal of Veterinary Research 82, 5; 10.2460/ajvr.82.5.343
Mean (SD) MFI for cells positive for DCF determined from triplicate analysis via flow cytometry of erythrocytes collected from 3 healthy dogs and incubated with increasing concentrations of DCFH-DA and H2O2.
| DCFH-DA concentration (μM) | H2O2 concentration (mM) | ||||
|---|---|---|---|---|---|
| 0 | 0.5 | 1 | 2 | 4 | |
| 0 | 2.12 (0.01)a | 2.56 (0.11)a | 2.60 (0.14)a | 2.59 (0.13)a | 2.53 (0.10)a |
| 5 | 3.10 (0.19)b | 3.61 (0.11)b | 3.61 (0.34)b | 3.19 (0.30)b | 3.42 (0.27)b |
| 10 | 3.13 (0.18)b | 3.84 (0.14)c | 3.77 (0.29)b | 3.76 (0.33)c | 3.65 (0.34)b |
| 25 | 3.40 (0.30)c | 3.87 (0.11)c | 3.86 (0.28)b | 3.89 (0.42)c | 3.89 (0.38)c |
| 50 | 3.29 (0.70)c | 4.06 (0.13)d | 4.01 (0.17)c | 4.00 (0.33)c | 3.89 (0.33)c |
In the same column, superscripted letters that differ indicate a significant (P < 0.05) difference between that value and the preceding value.
Increased fragmentation of erythrocytes was noted for the cells stimulated with 4mM H2O2. Mean intra-assay CV for MFI obtained with flow cytometric analysis of cells stimulated with H2O2 and not stimulated with H2O2 was 8.9% and 11.9%, respectively (Table 2). Similarly, mean interassay CV for stimulated cells was 9.1% and for unstimulated cells was 11.9% (Table 3). Stimulated samples had maximum CV, compared with baseline, of 9.8% and maintained an acceptable CV (≤ 20%) through 48 hours (Table 4). However, unstimulated samples were only stable up to 24 hours with a maximum CV, compared with baseline, of 27.6% at 36 hours.
Intra-assay CV (%) for logarithmic mean MFI obtained for 5 consecutive flow cytometric analyses of erythrocytes from 3 healthy dogs that were or were not stimulated with 0.5mM H2O2. Acceptable CV was ≤ 20%.
| Variable | Unstimulated | Stimulated | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 2 | 3 | 4 | 5 | 1 | 2 | 3 | 4 | 5 | |
| CV for each analysis | 9.9 | 3.3 | 16.8 | 6.6 | 22.0 | 11.2 | 14.3 | 6.3 | 8.8 | 4.1 |
| Overall CV | 11.9 | 8.9 | ||||||||
Logarithmic mean (SD) of the interassay MFI and mean interassay CV (%) for each dog and mean interassay CV per treatment (unstimulated and stimulated cells) for flow cytometric analyses of erythrocytes from 3 healthy dogs. Acceptable CV was ≤ 20%.
| Variable | Unstimulated | Stimulated | ||||
|---|---|---|---|---|---|---|
| Dog 1 | Dog 2 | Dog 3 | Dog 1 | Dog 2 | Dog 3 | |
| MFI | 2.99 (0.04) | 2.95 (0.05) | 2.99 (0.06) | 4.25 (0.04) | 4.36 (0.04) | 4.02 (0.02) |
| CV for each dog | 9.7 | 12.1 | 13.8 | 11.0 | 11.0 | 5.2 |
| Overall CV | 11.9 | 9.1 | ||||
Logarithmic mean (SD) MFI as determined with flow cytometric analyses in triplicate of erythrocytes from 3 healthy dogs that were (stimulated) or were not stimulated (unstimulated) with 0.5mM H2O2; analyses were performed at baseline and after 3, 6, 24, 36, and 48 hours. Acceptable CV was ≤ 20%.
| Time | Unstimulated | Stimulated | ||
|---|---|---|---|---|
| MFI | CV | MFI | CV | |
| Baseline | 2.88 (0.05) | — | 4.13 (0.04) | — |
| 3 | 2.95 (0.04) | 12.2 | 4.19 (0.04) | 8.8 |
| 6 | 2.94 (0.10) | 10.1 | 4.19 (0.04) | 9.8 |
| 24 | 3.02 (0.10) | 19.7 | 4.19 (0.03) | 9.2 |
| 36 | 3.04 (0.12) | 27.6 | 4.18 (0.112) | 7.2 |
| 48 | 3.04 (0.11) | 26.3 | 4.14 (0.003) | 0.6 |
— = Not applicable.
Baseline
The MFI for stimulated and unstimulated erythrocytes was not significantly different between male and female dogs (stimulated, P = 0.37; unstimulated, P = 0.29). For cells incubated with DCFH-DA, logarithmic mean (SD) MFI for stimulated cells (4.14 [0.01]) was significantly (P < 0.001) higher than for unstimulated cells (2.96 [0.02]; Figure 4). For cells incubated with DMSO, MFI did not differ significantly (P = 0.18) between stimulated and unstimulated cells. The median lower and upper limits of the interval for MFI of unstimulated samples determined with the robust method were 2.63 (90% CI, 2.56 to 2.70) and 3.25 (90% CI, 3.17 to 3.32), respectively.
Flow cytometric output from the analyses of erythrocytes from healthy dogs. A—For 1 dog, logarithmic fluorescence (FL1-A) versus cell count depicted by a histogram for erythrocytes incubated with 25μM DCFH-DA and stimulated (red) or not stimulated (blue) with 0.5mM H2O2. B—Box-and-whisker plot for MFI of H2O2-stimulated and unstimulated erythrocytes from 50 healthy dogs that were incubated with DCFH-DA. The horizontal line within each box represents the median, the boxes represent the interquartile (25th to 75th percentile) range, and the whiskers represent maximum and minimum values. Outliers are depicted as individual points. The MFI of stimulated erythrocytes was significantly (P < 0.001) higher than that for unstimulated erythrocytes. The “X” represents the mean. C—Box-and-whisker plot for MFI of H2O2-stimulated and unstimulated erythrocytes from the same dogs in panel B that were incubated with DMSO. The MFI was not significantly (P = 0.18) different between unstimulated and stimulated erythrocytes.
Citation: American Journal of Veterinary Research 82, 5; 10.2460/ajvr.82.5.343
Discussion
Flow cytometric measurement of intraerythrocytic ROS was successfully validated for healthy dogs in the present study. The fluorochrome DCFH-DA reliably diffused into the erythrocytes and formed the impermeable H2DCF, and oxidation of H2DCF by H2O2 yielded the fluorescent product DCF, an indicator of oxidative stress. Proper gating allowed for the isolation of erythrocytes and determination of cellular autofluorescence through the use of the vehicle DMSO. Because DCFH-DA is not cell specific, potential contamination by cells other than erythrocytes was confirmed to be minimal. Leukocyte contamination is the most important to minimize because neutrophils use H2O2 as part of their oxidative burst and phagocytic function.1,24,25 Future studies could investigate the degree to which DCFH-DA may identify ROS in leukocytes with flow cytometry.
Also in the present study, H2O2, even at low concentrations, reliably induced fluorescence in the presence of DCFH-DA. Fluorescence did not increase with increasing concentrations of H2O2, and 4mM H2O2, the highest concentration of H2O2 used, was associated with erythrocytic fragmentation. Fragmentation at H2O2 concentrations of ≥ 6mM has been identified for equine erythrocytes (unpublished data). However, canine erythrocytes were more susceptible to injury and hemolysis at H2O2 concentrations > 2mM. An H2O2 concentration of 4mM was only used to determine the optimal concentration of DCFH-DA needed to reliably detect ROS formation and fluorescence. Remaining portions of the study included stimulation with 0.5mM H2O2 so that the risk of erythrocytic fragmentation was minimized. Because no H2O2 concentrations between 0mM and 0.5mM were investigated, 0.5mM may not have been the optimal concentration; but, for the purpose of validating the flow cytometric assay, this concentration was effective. Future studies should include the investigation of H2O2 concentrations < 0.5mM. Additionally, future studies are needed to ensure that the type of oxidative injury expected with clinical illness is the same as that stimulated by H2O2.
Interestingly, fluorescence was slightly increased for erythrocytes incubated with DMSO versus DCFH-DA. The fluorochrome DCFH-DA was suspended in DMSO, which is a common diluent and cytoprotectant and was the control vehicle for the present study. Possibly, DMSO may have caused cellular fluorescence because DMSO, despite its cytoprotective characteristic, may be toxic to erythrocytes. However, erythrocytic toxicosis occurs at higher concentrations than those used in the present study.26 When used with a different fluorochrome (vs DCFH-DA), DMSO affects flow cytometric results for aquatic species.27 Also possible for this slight increase in fluorescence was a variation in autofluorescence for the cells from the 3 dogs used for this portion of the study. Future studies should include an investigation into the effects of DMSO on erythrocytes and the flow cytometric assay used in the present study to determine whether DMSO is the best control vehicle.
Healthy dogs were used for the present study, and calculated precision implied a linear relationship between oxidative stimulation and DCFH-DA free radical generation. However, this implied linear relationship may not be true for diseased dogs. In people, experimental use of DCFH-DA for detecting cellular ROS does not have a similar linear relationship.17,18 Therefore, a logarithmic scale was used, and doing so may be more sensitive to detect minor changes in intracellular ROS.18,20 Use of a logarithmic scale in the future is recommended, especially when DCFH-DA is selected as the fluorochrome for flow cytometric detection of ROS in disease states like anemia, diabetes mellitus, or chronic kidney disease, in which the degree of oxidative stress is variable or not well described.7,11,16
Unstimulated erythrocyte samples were stable for up to 24 hours after collection, whereas stimulated samples were stable for up to 48 hours. This may be because of variable depletion of antioxidants or hemoglobin auto-oxidation that occurred during cellular storage, although no difference in mean corpuscular hemoglobin concentration was noted between the erythrocytes of unstimulated and stimulated samples.13,18 The observed stability time for unstimulated samples may have represented normal variation in baseline redox status of the cells, whereas the addition of H2O2 may have been a potent stimulus of oxidation that caused a predictable and profound increase in MFI. Because of this difference in sample stability, samples should be analyzed within 24 hours of collection.
One limitation of the present study was that we did not determine assay accuracy; however, no reference standard for the measurement of ROS was available. Therefore, without a reference standard, the outcome of this study simply indicated validation of a new method. Other methods should be investigated in the future to substantiate and compare these findings. The fluorochrome DCFH-DA reliably detected intraerythrocytic oxidative stress, but DCFH-DA requires 2-electron oxidation to fluoresce; therefore, this assay does not detect 1-electron oxidizing species, such as hydroxyl and nitric oxide radicals.18 One-electron oxidation of DCFH-DA results in the formation of an intermediate radical that traps the enzyme in the cell, but this intermediate has the potential to react with stable oxygen or iron to form a superoxide, which will yield additional H2O2 and could artifactually amplify fluorescence.18
Flow cytometry with use of DCFH-DA was successful to identify intraerythrocytic ROS in healthy dogs and was precise with a maximum intra- and interassay CV of 11.9%. Samples were stable up to 24 hours, after which pro- and antioxidant systems were variably altered, making interpretation of the results more challenging. Because of this method's success for direct measurement of oxidative stress in the erythrocytes of healthy dogs, this method may be equally successful for measurement of oxidative stress in the erythrocytes of dogs with various diseases.
Acknowledgments
Funded by internal competitive research funds of the College of Veterinary Medicine, Purdue University.
The authors declare that there were no conflicts of interest.
Abbreviations
| CV | Coefficient of variation |
| DCF | Dichlorofluorescein |
| DCFH-DA | 2‘,7‘-dichlorodihydrofluorescein diacetate |
| DMSO | Dimethyl sulfoxide |
| H2DCF | 2‘,7‘-dichlorodihydrofluorescein |
| H2O2 | Hydrogen peroxide |
| MFI | Median fluorescence intensity |
| PBSSA | Phosphate buffered saline solution with 1% bovine serum albumin |
| ROS | Reactive oxygen species |
Footnotes
Denamarin, Nutramax Laboratories, Lancaster, SC.
Denosyl, Nutramax Laboratories, Lancaster, SC.
Fisher Scientific, Pittsburgh, Pa.
Sigma-Aldrich, St Louis, Mo.
MilliporeSigma, Burlington, Mass.
Mallinckrodt Pharmaceuticals Co, St Louis, Mo.
TS Sorvall ST 40R centrifuge, Thermo Scientific, Pittsburgh, Pa.
BD Accuri C6 flow cytometer, Becton, Dickinson and Co, Franklin Lakes, NJ.
Abbott Cell-Dyn 3700, Abbott Park, Ill.
FlowJo, Ashland, Ore.
GraphPad Software Inc, San Diego, Calif.
StataCorp LLC, College Station, Tex.
References
- 1. ↑
Fibach E, Dana M. Red blood cells as redox modulators in hemolytic anemia. In: Tombak A, ed. Erythrocyte. London: IntechOpen Ltd, 2019.
- 2. ↑
Valko M, Leibfritz D, Moncol J, et al. Free radicals and antioxidants in normal physiological functions and human disease. Int J Biochem Cell Biol 2007;39:44–84.
- 3.
Kadiiska MB, Peddada S, Herbert RA, et al. Biomarkers of oxidative stress study VI. Endogenous plasma antioxidants fail as useful biomarkers of endotoxin-induced oxidative stress. Free Radic Biol Med 2015;81:100–106.
- 4.
McMichael MA. Oxidative stress, antioxidants, and assessment of oxidative stress in dogs and cats. J Am Vet Med Assoc 2007;231:714–720.
- 5. ↑
de Zwart LL, Meerman JH, Commandeur JN, et al. Biomarkers of free radical damage applications in experimental animals and in humans. Free Radic Biol Med 1999;26:202–226.
- 6.
Chansaisakorn W, Sriphavatsarakorn P, Sopakdittapong P, et al. Oxidative stress and intraerythrocytic concentrations of sodium and potassium in diabetic dogs. Vet Res Commun 2009;33:67–75.
- 7. ↑
Ojaimi C, Kinugawa S, Recchia FA, et al. Oxidant-NO dependent gene regulation in dogs with type I diabetes: impact on cardiac function and metabolism. Cardiovasc Diabetol 2010;9:43.
- 8. ↑
Kendall A, Woolcock A, Brooks A, et al. Glutathione peroxidase activity, plasma total antioxidant capacity, and urinary F2-isoprostanes as markers of oxidative stress in anemic dogs. J Vet Intern Med 2017;31:1700–1707.
- 9. ↑
Pesillo SA, Freeman LM, Rush JE. Assessment of lipid peroxidation and serum vitamin E concentration in dogs with immune-mediated hemolytic anemia. Am J Vet Res 2004;65:1621–1624.
- 10.
Kogika MM, Lustoza MD, Hagiwara MK, et al. Evaluation of oxidative stress in the anemia of dogs with chronic kidney disease. Vet Clin Pathol 2015;44:70–78.
- 11. ↑
Silva AC, de Almeida BF, Soeiro CS, et al. Oxidative stress, superoxide production, and apoptosis of neutrophils in dogs with chronic kidney disease. Can J Vet Res 2013;77:136–141.
- 12.
Freeman LM, Rush JE, Milbury PE, et al. Antioxidant status and biomarkers of oxidative stress in dogs with congestive heart failure. J Vet Intern Med 2005;19:537–541.
- 13. ↑
Smith A. Homeostasis of heme in health and disease: current aspects of the structural biology of heme-protein interactions and of gene regulation. DNA Cell Biol 2002;21:245–249.
- 14. ↑
Gibson JS, Wadud R, Lu D, et al. Oxidative stress and haemolytic anaemia in dogs and cats: a comparative approach. Integr J Vet Biosci 2019;3:1–5.
- 15. ↑
Fibach E, Rachmilewitz E. The role of oxidative stress in hemolytic anemia. Curr Mol Med 2008;8:609–619.
- 17. ↑
Amer J, Goldfarb A, Fibach E. Flow cytometric measurement of reactive oxygen species production by normal and thalassaemic red blood cells. Eur J Haematol 2003;70:84–90.
- 18. ↑
Kalyanaraman B, Darley-Usmar V, Davies KJA, et al. Measuring reactive oxygen and nitrogen species with fluorescent probes: challenges and limitations. Free Radic Biol Med 2012;52:1–6.
- 19. ↑
Christov A, Hamdheydari L, Grammas P. Detection of reactive oxygen species by flow cytometry. In: Hensley K, Floyd RA, eds. In: Methods in biological oxidative stress. Totowa, NJ: Humana Press Inc, 2003;175–184.
- 20. ↑
Eruslanov E, Kusmartsev S. Identification of ROS using oxidized DCFDA and flow-cytometry. Methods Mol Biol 2010;594:57–72.
- 21. ↑
Fibach E, Dana M. Oxidative stress in paroxysmal nocturnal hemoglobinuria and other conditions of complement-mediated hemolysis. Free Radic Biol Med 2015;88:63–69.
- 22. ↑
O'Hara DM, Xu Y, Liang Z, et al. Recommendations for the validation of flow cytometric testing during drug development: II assays. J Immunol Methods 2011;363:120–134.
- 23. ↑
Friedrichs KR, Harr KE, Freeman KP, et al. ASVCP reference interval guidelines: determination of de novo reference intervals in veterinary species and other related topics. Vet Clin Pathol 2012;41:441–453.
- 24. ↑
Amer J, Zelig O, Fibach E. Oxidative status of red blood cells, neutrophils, and platelets in paroxysmal nocturnal hemoglobinuria. Exp Hematol 2008;36:369–377.
- 25. ↑
Webb C, Bedwell C, Guth A, et al. Use of flow cytometry and monochlorobimane to quantitate intracellular glutathione concentrations in feline leukocytes. Vet Immunol Immunopathol 2006;112:129–140.
- 26. ↑
Yi X, Liu M, Luo Q, et al. Toxic effects of dimethyl sulfoxide on red blood cells, platelets, and vascular endothelial cells in vitro. FEBS Open Bio 2017;7:485–494.
- 27. ↑
Daly J, Tiersch TR. Sources of variation in flow cytometric analysis of aquatic species sperm: the effect of cryoprotectants on flow cytometry scatter plots and subsequent population gating. Aquaculture 2012;370–371:179–188.
