Oxidative stress and antioxidant status in intensely exercising horses administered nutraceutical extracts



Reasons for performing study: Many nutraceuticals are used as equine supplements without their efficacy having been scientifically tested. Black tea, cranberries, orange peel and ginger are a few of those nutraceuticals that warrant further study.

Objective: To test the effects of single doses of black tea, cranberry, orange peel and ginger extract on markers of oxidative stress and antioxidant status following exercise in horses.

Methods: In Study 1, 9 mature, healthy but unfit Standardbred mares were administered 2 l of a control (water), orange peel extract (30 g extract) or decaffeinated black tea extract (28 g extract). In Study 2 the same mares were administered 2 l of a control (water), cranberry extract (30 g extract) or ginger extract (30 g extract). In each study, mares were given the extracts via nasogastric tube 1 h before performing a graded exercise test (GXT), in a randomised crossover design with at least 7 days between GXTs. Blood samples were collected at rest, at fatigue, and 1 and 24 h post exercise and analysed for lipid hydroperoxides (LPO), total glutathione (GSH-T), glutathione peroxidase (GPx), α-tocopherol (TOC), β-carotene (BC) and retinol. Data were statistically analysed using a repeated measures ANOVA.

Results: In Study 1 there was no effect of treatment for LPO, GSH-T, GPx, TOC or BC. Retinol was higher for both tea (P = 0.0006) and water (P = 0.004) than for orange peel. In Study 2 there was no treatment effect for LPO, GPx, GSH-T, RET, BC or TOC.

Conclusions: The results show that a single dose of various nutraceuticals in exercising horses do not produce an effect on either oxidative stress or antioxidant status and further investigation is needed as to whether long-term supplementation would enhance these effects.


Equine athletes are often given supplements to reduce the purported harmful effects of oxidative stress caused by free radicals generated during exercise. Many antioxidants can be found within natural food products (Bean et al. 2010). The food extracts of cranberry (Vaccinium macrocarpon), ginger (Zingiber officinale), black tea (Camellia sinensis) and orange peel (Citrus sinensis) have yet to be analysed in intensely exercising horses to determine if they are beneficial as nutraceuticals for increasing antioxidant status.

Cranberries (Vaccinium macrocarpon) have long been used for treatment and prevention of human urinary tract infections, but recently have been noted for antioxidant and other properties such as the ability to protect endothelial cells against stress-induced up-regulation of oxidative and inflammatory mediators (Sun et al. 2002; Youdim et al. 2002). Phenolics in cranberries, like quercetin and cyanidin, have highly effective radical scavenging structures (Zheng and Wang 2003). Anthocyanins, flavonoids that cause the red colour found in cranberries, are also powerful antioxidants (Viskelis et al. 2009). A recent study in male New Zealand rabbit kidneys shows that cranberry powder decreases both accumulation of malonedialdehyde (MDA, an indicator of oxidative damage) and inflammation when challenged with an infection of Escherichia coli (Han et al. 2007). Based on these studies showing antioxidant properties of cranberries, administration with cranberry extract could aid in increasing the antioxidant status of exercising horses.

Ginger (Zingiber officinale) has been used to treat a variety of ailments, including rheumatism, sprains, muscular aches, indigestion and infectious diseases (Badreldin et al. 2008). Ginger has also been shown to have antioxidant and anti-inflammatory properties when administered in man (Dugasani et al. 2010) and has also been used to ease the symptoms of arthritis (Srivastava and Mustafa 1992). The major nonvolatile compounds in ginger include [6]-gingerol, [8]-gingerol, [10]-gingerol and [6]-shogoal; all of these can inhibit both superoxide production generated by xanthine/xanthine oxidase and nitric oxide production (Dugasani et al. 2010). Another study showed that ginger acts as an antioxidant by decreasing lipid peroxidation, increasing GSH content and maintaining normal levels of antioxidant enzymes (Ahmed et al. 2000). Given the antioxidant effects that ginger has had on other species it could be speculated that ginger administration could show potential in increasing the antioxidant levels of horses.

Black tea is well known for having many positive health effects (Yang et al. 1997; Sharma and Rao 2009). Black tea contains a variety of polyphenols, the secondary metabolites of plants that areused in the plant's defence system against severe environments (Wang and Ho 2009). The major polyphenols in tea are flavonoids, which include flavanols (in particular catechins and gallic acid), theaflavins and phenolic acids (Sharma and Rao 2009). These flavonoids contribute to the antioxidant capacity of black tea. For example, theaflavin forms complexes with metals such as ferric iron (found in red blood cells), thus protecting red blood cells from damage that can occur during oxidative stress (Sharma and Rao 2009). Black tea theaflavins also prevent cellular DNA damage by reducing oxidative stress and suppressing cytochrome P451A1 in rat liver (Sharma and Rao 2009). In addition, black tea prevents lipoprotein oxidation in mice and inhibits production of both nitric oxide and superoxide in murine peritoneal macrophages (Sharma and Rao 2009). Black tea also has anti-inflammatory actions, as documented by a recent study demonstrating that theaflavin is a strong inhibitor of in vitro gene expression for the inflammatory chemokine IL-8 (Benjamini et al. 2000; Aneja et al. 2004).

Orange peel, the primary waste fraction in the production of orange juice, contains flavonoids associated with antioxidant activity (Kanaze et al. 2008). The glycosides hesperidin and naringin are mainly responsible for the antioxidant activity of citrus peel extracts (Kanaze et al. 2008). Coniferin and phlorin are additional phenols in orange peels found to aid in radical scavenging when administered in the form of orange peel molasses (Manthey 2004). The fractionated phenols found in orange peel protect against damaging biological oxidative stress by acting as radical scavengers.

The goal of this study was to test the hypothesis that various nutraceutical extracts (Study 1= black tea and orange peel extracts; Study 2= cranberry and ginger extracts) would alter markers of exercise-induced oxidative stress or antioxidant status in horses.

Materials and methods

Study design and subjects

The Rutgers University Institutional Animal Care and Use Review Board approved all methods and procedures used in this experiment. Both studies used 9 healthy, unfit Standardbred mares, aged 10 ± 4 years, weighing about 450 kg. One horse in Study 2 was dropped from the third exercise trial due to lameness not related to the experiment. The mares were unfit but accustomed to the lab and running on the treadmill before the start of the experiment. During the studies, the horses were housed as a group on a 3 acre (1.2 ha) exercise dry-lot. Each mare was fed approximately 6 kg/day of a grass hay mix and approximately 3 kg/day of a commercially available total mixed ration hay cube1, divided into 2 feedings (for nutrient composition see Table 1). Hay was tested for flavinols prior to the study with little to no detection, and an estimation of antioxidant content (vitamin E and beta-carotene) was determined from the NRC (NRC 1989, 2007). Horses had ad libitum access to water and a salt block.

Table 1. Nutrient composition of feeds on a dry matter basis
NutrientHay cubesaGrass haya
  • a

    Based on analysis by Dairy One, Inc., Ithaca, New York, USA.

  • b

    Estimated values from the NRC (Anon 1989, 2007). DE = digestible energy.

Dry matter, %90.492.8
Horse DE, MJ/kg11.38.0
Crude protein, %18.77.0
Acid detergent fibre, %26.937.8
Neutral detergent fibre, %35.966.2
Non fibre carbohydrates, %33.020.4
Crude fat, %2.12.7
Calcium, %0.840.23
Phosphorus, %0.380.18
Magnesium, %0.230.12
Potassium, %1.521.75
Sodium, %0.290.003
Iron, ppm173061
Zinc, ppm5020
Copper, ppm117
Manganese, ppm7558
Vitamin E, mg/kgNA11.6b
Beta-Carotene, mg/kgNA47.4b

For each study, horses were divided into 3 treatment groups using a randomised crossover design. For Study 1 they were administered 2 l of either a control (water), orange extract (30 g extract dissolved in 2 l water), or decaffeinated black tea extract (28 g extract dissolved in 2 l water). For Study 2, horses were administered 2 l of either a control (water), cranberry extract (30 g of extract dissolved in 2 l water), or ginger extract (30 g of extract dissolved in 2 l water). The extract dosages given to horses were previously chosen by referring to previous studies in rats and man on a metabolic bodyweight basis. Each administration was given via nasogastric tube on the morning of the study, 1 h prior to exercise. Pharmacokinetic studies were conducted on the extracts prior to the start of the study, and this information has been previously published (Kuang 2005; Reddy 2005). Feed was removed the morning of the study and returned with the afternoon feeding 12 h after extract administration (Streltsova et al. 2006). The exercise trials were conducted between 07.00 and 12.00 h no less than 7 days apart over a 3 week period.

Approximately 1 week before the beginning of Study 1, all animals completed a series of baseline tests. Baseline testing was comprised of an incremental graded exercise test (GXT) to measure maximal aerobic capacity (VO2max; Streltsova et al. 2006; Liburt et al., 2010).

Extract preparation

For Study 1, 30 g of orange peel extract was dissolved in 100 ml of ethanol and 10 g of lecithin was added. The mixture was brought to a boil and then slowly added to warm (49°C) water under high shear (mixing) conditions. This pre-emulsion was then dispersed in 1500 ml of warm water (also under high shear conditions) to yield 2 l of extract for administration (Reddy 2005; Streltsova et al. 2006).

The tea extract was prepared by taking the theaflavin mixture (28 g), dry blending with an equal amount of sugar, dissolving this premix in 500 ml warm water with vigorous stirring, then diluting to 2 l and lastly adding citric acid to reduce the pH below 4.5. The black tea extract was decaffeinated (Kuang 2005; Streltsova et al. 2006).

The extracts for Study 2 were prepared by using either 30 g of cranberry or ginger extract dissolved in 100 ml of ethanol, and 10 g of lecithin was added. The mixture was brought to a boil and slowly added to warm (49°C) water under high shear conditions. This pre-emulsion was then dispersed in 1500 ml of warm water, also under high shear conditions to yield 2 l of extract for administration (Liburt et al. 2010).

Graded exercise test

Prior to every exercise test the horses were weighed and a catheter (Angiocath, 14 gauge)2 inserted percutaneously into the left jugular vein, using sterile techniques and local lidocaine anaesthesia. The horses were then administered their respective treatment and stood quietly for 1 h in a stall. The horses were walked onto the treadmill where they stood quietly for a 10–15 min equilibration period during which post dose, pre-exercise blood samples (10 ml) were obtained.

During the GXT, the animals ran on a high speed horse treadmill (Sato I)3 up a fixed 6% grade. The tests started at an initial speed of 4 m/s for 1 min. Speed was then increased to 6 m/s followed by incremental 1 m/s increases every 60 s (omitting 5 m/s), until the horses reached fatigue. Fatigue was defined as the point where the horse could not keep up with the treadmill despite encouragement. After this, the treadmill was stopped and 5 min of post exercise data were collected. Rectal temperature and bodyweight were obtained again immediately upon completion of the test. Performance indicators such as maximal heart rate, speed at fatigue and time to fatigue were also measured and previously reported (Streltsova et al. 2006; Liburt et al. 2010).

For both studies, blood samples (35 ml) were obtained during the tests at rest (REST), during the last 10 s of the final step (FATIGUE), and 1 and 24 h post exercise. Blood samples were placed into prechilled tubes containing EDTA or sodium heparin (Vacutainer)4 and immediately placed on ice.

Sample analysis

Red blood cells were analysed for total glutathione (GSH-T) and glutathione peroxidase (GPx). Methods for analysis of GSH-T (GSH-420, kit #210235; interassay coefficient of variance [CV] 7.0 %, intra-assay CV 5.6%) and cellular GPx (GPx-340, kit #210175; interassay CV 4.2%, intra-assay CV 5.0%) have been previously described (Williams et al. 2004a).

The lipid hydroperoxide (LPO) concentrations of equine plasma were determined using the PCA-FOX assay (inter-assay CV 3.0%, intra-assay CV 4.6%; Wolff 1994; Gay et al. 1999; Gay and Gebicki 2002). Each plasma sample was divided into a reduced ‘blank’ and an unreduced ‘test’ sample. Catalase was added to eliminate H2O2 interference (Nourooz-Zadeh et al. 1994). TCEP (Tris [2-carboxyethyl] phosphine HCl) was then used to reduce the blank sample, followed by an incubation (Anderson 1989; Nourooz-Zadeh et al. 1994). All samples then received the PCA-FOX assay reagent (xylenol orange and ferrous ammonium iron [II] sulphate in perchloric acid; Gay and Gebicki 2002), and were incubated again. The reaction mixture was then centrifuged and the supernatant aliquoted into a microtitre plate and read at 560 nm with a microplate reader (Spectramax 340)6. Concentrations were determined by dividing the net absorbance by the molar extinction coefficient of LPO in perchloric acid (Gay and Gebicki 2002).

The α-tocopherol (TOC), retinol (RET) and β-carotene (BC) were analysed by Michigan State Diagnostic Laboratories7 using a HPLC (intra-assay CV 1.28%, interassay 2.5%). Samples were run first through a precolumn followed by a reverse phase C-18 HPLC column eluted isocratically at 1.2 ml/min with an injection volume of 50 ul. Absorbance was measured for TOC (292 nm), RET (325 nm) and BC (450 nm).

Statistical analysis

Data are presented as mean ± s.e. unless otherwise noted. Data were analysed for effects of treatment and time of exercise plus potential interactions using an ANOVA for repeated measures in SAS8. Data were also compared for week of exercise and no differences were found. Post hoc comparisons of means were performed using the Tukey test with significance set at P = 0.05. If no significant difference was found between treatments then the data were averaged, unless otherwise stated.


Study 1

The run time, recovery time, plasma lactate and maximal oxygen consumption for this study have been previously published along with markers of inflammation (Streltsova et al. 2006). There was no effect of treatment for LPO, GSH-T, GPx, BC or TOC (Table 2). However, an effect of treatment (P = 0.002) was seen on retinol concentration with tea (P = 0.002; 203 ± 4 ng/ml) and water (P = 0.011; 198 ± 4 ng/ml) being higher than orange peel (180 ± 4 ng/ml; Fig 1). There was an effect of sample time (P<0.0001) for erythrocyte GSH-T, erythrocyte GPx and plasma TOC, all of which peaked at FATIGUE, but returned to baseline by 1 and 24 h. Plasma RET and BC both had an effect of sample time (P<0.0001) with peaks at FATIGUE returning to baseline by 1 h and dropping below baseline at 24 h. There was no effect of sample time for LPO.

Table 2. Concentration of plasma lipid hydroperoxides (LPO), erythrocyte total glutathione (RBC GSH-T), erythrocyte glutathione peroxidase (RBC GPx), plasma α-tocopherol (TOC) and plasma β-carotene (BC) in Study 1, before exercise (REST), at fatigue, and 1 and 24 h after exercise. Due to the lack of a treatment effect treatment groups have been combined
VariableRESTFATIGUE1 h24 hs.e.
  • a,b,c

    Concentrations within rows with different superscripts differ (P<0.05).

Plasma LPO, µmol/l3.633.593.293.670.42
RBC GSH-T, µg/mg protein66.89b82.43a68.81b62.42b3.29
RBC GPx, mu/mg protein29.35b33.24a30.74ab28.47b1.20
Plasma TOC, µg/ml2.80b3.50a2.86b2.73b0.10
Plasma BC, µg/ml0.152b0.182a0.155b0.116c0.008
Figure 1.

Concentration of plasma retinol (RET) in Study 1 for control (water), black tea and orange peel before exercise (REST), at fatigue and 1 and 24 h after exercise. There was an effect of treatment (P=0.002) and means at different time points with different letters differ at P≤0.05 (since there was no interaction, sample time differences are with average of overall treatments).

Study 2

The run time, recovery time, plasma lactate and maximal oxygen consumption for this study have been previously published along with markers of inflammation (Liburt et al. 2010). There was no effect of treatment for LPO, GSH-T, GPx, TOC, BC or RET (Table 3). There was no effect of sample time for LPO or GSH-T. Erythrocyte GPx had an effect of sample time (P = 0.006) where it peaked at FATIGUE and returned to baseline by 1 and 24 h; however, FATIGUE was not different from 24 h. Plasma TOC and RET had an effect of sample time (P<0.0001) where they peaked at FATIGUE and returned by baseline at 1 and 24 h. Plasma BC had an effect of sample time (P = 0.005) where it decreased to its lowest point was at the 24 h sample.

Table 3. Concentration of plasma lipid hydroperoxides (LPO), erythrocyte total glutathione (RBC GSH-T), erythrocyte glutathione peroxidase (RBC GPx), plasma α-tocopherol (TOC), retinol (RET) and plasma β-carotene (BC) in Study 2, before exercise (REST), at fatigue and 1 and 24 h after exercise. Due to the lack of a treatment effect treatment groups have been combined
VariableRESTFATIGUE1 h24 hs.e.
  • a,b

    Concentrations within rows with different superscripts differ (P<0.05).

Plasma LPO, µmol/l3.814.183.893.490.53
RBC GSH-T, µg/mg protein87.8592.9575.1675.407.33
RBC GPx, mu/mg protein23.68b29.29a23.99b25.18ab2.29
Plasma TOC, µg/ml2.46b3.02a2.41b2.49b0.08
Plasma RET, ng/ml193.3b220.7a191.4b186.8b4.2
Plasma BC, µg/ml0.086a0.089a0.078ab0.061b0.010


The horses administered cranberry, ginger, orange peel and black tea did not experience lower oxidative stress with exercise compared to the control horses. Antioxidant status for all horses changed due to exercise. However, the treatments did not have any effect on antioxidant status except for the orange peel group, which had lower concentrations of retinol compared to water.

Neither cranberry nor ginger had an effect on the oxidative stress or antioxidant status in the exercising horses. This is in contrast to previous studies performed in other species that have shown both extracts to have an effect on antioxidant status. In a study performed by Ahmed et al. (2000), oxidative stress was induced in rats with malathion; these rats were then fed a 1% ginger diet for 4 weeks, which decreased lipid peroxidation as compared to the control rats. In another study, rats were split into several groups and fed 3 different diets containing 0.5, 1 and 5% ginger. After one month of supplementation, significant increases in activity of antioxidant enzymes (superoxide dismutase, glutathione peroxidase and catalase) were found in the liver (Kota et al. 2008). These studies suggest that the antioxidant effects of ginger extract may only occur after multiple administrations and might account for the differences found between this study of horses and studies with rats.

It is well known that cranberries have phenolics and anthocyanins that are antioxidants (Zheng and Wang 2003; Dugasani et al. 2010). In this study, the administration of cranberry did not affect oxidative stress or antioxidant status. One possible explanation for this is that the horses were not dosed with cranberry extract repeatedly over time. Another theory is that while exercise had an effect on antioxidant status, it did not affect lipid peroxidation; therefore, the horses did not undergo sufficient oxidative stress to warrant a necessary decrease. This speculation is supported by several studies with healthy human subjects that showed that long-term supplementation with cranberry extracts does not improve antioxidant status (Duthie et al. 2006; Valentova et al. 2007). Healthy female volunteers given cranberry juice (750 ml/day) for 2 weeks did not have changes in total phenol, anthocyanin or catechin concentration of the blood plasma (Duthie et al. 2006). Similarly, Valentova et al. (2007) supplemented healthy women with dried cranberry juice (400 mg/day) for 8 weeks and again, no effect was seen on markers of oxidative stress (superoxide dismutase, glutathione, and glutathione peroxidase). However, the results of these studies are in contrast to those from another study done in rabbits suffering from infection-induced oxidative renal damage, where supplementation with cranberry powder (1 g/kg bwt/day) for 3 weeks resulted in a significant reduction in malonedialdehyde (MDA), an indicator of oxidative stress (Han et al. 2007). While speculative, it may be that cranberry extracts are more effective when the body has undergone intense, long-term oxidative stress, rather than in healthy individuals or in individuals undergoing short-term or acute oxidative stress.

Black tea contains flavonoids and theaflavins that have an antioxidant capacity (Sharma and Rao 2009). In a study conducted by Arent et al. (2010), college-age male athletes consumed 1760 mg/day of black tea extract for 9 days. This increased total baseline GSH compared to controls, which later help buffer exercise-induced oxidative stress. In the present study, there was no effect of black tea on total GSH and one explanation for this difference may be the usage of a single dose vs. multiple dosing. Black tea also did not seem to affect levels of retinol, which differed from other work with tea extracts and retinol levels. Retinol, also known as vitamin A, has been seen to have antioxidant properties. In a study in which healthy rats consumed a liquid diet with 7 g green tea extract/l diet for 5 weeks, vitamin A levels were significantly increased in rats aged 2 and 12 months (Wojciech et al. 2010). Based on the previous study, administering tea extracts increases retinol levels; however, further research will need to be conducted in horses to see if other antioxidant levels could be increased with multiple doses of tea extract.

As stated before, phenols found in orange peel act as radical scavengers that protect against oxidative stress. In the present study, horses administered the orange peel extract appeared to have lower concentrations of plasma retinol compared to the control group. In a study of human patients with arterial disease, supplementation for 4 weeks with orange juice did not affect the activity of the antioxidant enzyme paraoxonase-1 (Dalgard et al. 2007). However, supplementation with hesperidin, a flavanone glycoside found in orange peel, may affect antioxidant status in animals undergoing physiological challenges. In a study done in rats challenged with benzo[α]pyrene, a mutatgen and carcinogen, oral doses of hesperidin (200 mg/kg bwt/day) for 10 days significantly increased levels of superoxide dismutase (SOD), glutathion-S-transferase (GST) and GSH (Arafa et al. 2009). In the same study, healthy rats given the same amount of hesperidin had no changes in antioxidant concentrations (Arafa et al. 2009). Rats treated with acrylonitrile (a volatile toxic liquid) had reduced levels of antioxidants, but rats treated with acrylonitrile and given hesperidin daily (200 mg/kg bwt/day) for 28 days had normal antioxidant levels of SOD, catalase (CAT), GST and GSH in another study (El-Sayed et al. 2008). Similar to the work by Arafa et al. (2009), El-Sayed et al. (2008) found that healthy mice treated with hesperidine had no change in levels of antioxidants. Naringin is another flavanone glycoside found in citrus peel, and in a study conducted with New Zealand White rabbits fed a high cholesterol diet, the rabbits supplemented with naringin (0.5 g/kg diet) for 8 weeks had higher levels of SOD, CAT and vitamin E compared to rabbits fed only a high cholesterol diet (Jeon et al. 2002). Based on these studies supplementation with glycosides found in orange peel extract, may be beneficial only when the body is under severe oxidative stress, and when supplementation is given over time and not as a single administration. More work will need to be conducted, since the results from the current study contradict previously reported work on extracts from orange peel.

No change was seen in LPO for any of the groups. Lipid hydroperoxides are normally seen after intense exercise (Clarkson and Thompson 2000) and it is interesting that there was no difference here. Other studies have shown increased lipid peroxidation when utilising simulated race tests at or above maximal heart rate (Chiaradia et al. 1998; White et al. 2001). The horses did exercise intensely, since in a concurrent study using the same horses and the same exercise protocol, plasma creatine kinase (CK) was measured and found to have exercise-induced increases (Streltsova et al. 2006; Liburt et al., 2010). Additionally, the exercise test performed in the current study had an effect on antioxidant status measured, consistent with results of a similar study performed in the same laboratory utilising the same breed of horse and the same blood analysis (Williams et al. 2008). Exercising horses for a longer duration, without the added intensity, may not create an increase in oxidative stress. This speculation is supported by results from a previous study where endurance horses exercised at a longer duration but more submaximal intensities and no change in oxidative stress occurred (Williams et al. 2004b). One possible reason for the mares not achieving a higher level of oxidative stress could be that the mares are accustomed to the exercise laboratory and running a GXT on the treadmill vs. what may have been seen with a more novel test or subjects.

It has been shown by previous studies that these extracts can increase antioxidant status in other species; however, the supplementation period was generally over multiple days. In addition, it was observed in other studies the extracts may be most beneficial in times of severe oxidative stress, such as a disease state or endurance-type intense exercise. In the future, studies with long-term supplementation and higher intensity, endurance type exercise may be more appropriate in determining the effects of these nutraceutical extracts. Additionally, measurement of the specific antioxidant levels in the extracts before administration may give a clearer indication of the antioxidant intake on the particular day of testing. Baseline levels of the antioxidants found in the horses before dosage would also have helped determine the change in antioxidant levels after dosage administration but before exercise. The lack of both of these measurements can be considered limitations in the current study and should be taken into account. In conclusion, further studies are needed to see if long-term supplementation with these extracts can improve antioxidant status in the exercising horse.

Conflicts of interest

The authors have declared no potential conflicts.

Manufacturers' addresses

1 Eckenberg Farms, Mattawa, Washington, USA.

2 Becton Dickson, Inc., Parsippany, New Jersey, USA.

3 Equine Dynamics, Inc., Lexington, Kentucky.

4 Becton Dickson, Inc., Franklin Lakes, New Jersey, USA.

5 Bioxytech, Oxis Research, Portland, Oregon, USA.

6 Molecular Devices, Sunnyvale, California, USA.

7 Michigan State University, Diagnostic Laboratories, East Lansing, Michigan, USA.

8 Version 9.1, Cary, North Carolina, USA.