A validated method for the quantification of IOX-2, a potent prolyl hydroxylase inhibitor in equine urine and plasma using liquid chromatography–high-resolution mass spectrometry

Ezra Mikhail1 | Erik Siccardi2 | Ali Bawazir1 | Ambika Rajesh1 |
Seetharani Prathyush1 | Duaa Mohammad Kamal Al Wazani1 | Muhammed Sabeek1 | Thomas John1

1Veterinary Sports and Research Centre, Wadi Al Safa, Dubai, UAE
2Al Nasr-2 Equine Clinic, Dubai, UAE

Correspondence
Ezra Mikhail, Veterinary Sports and Research Centre, Wadi Al Safa, Dubai, UAE.
Email: [email protected]

Funding information
Zabeel Equestrian office

1 | INTRODUCTION

Use of performance-enhancing substances has always been an issue in sports. Acknowledging the facts of performance-enhancing and del- eterious side effects, the Federation for Equestrian Sports (FEI) has banned such drugs in sports.1 Still numerous pharmacological strate- gies to simulate the effects of hypoxia at the cellular level and

increase expression of hypoxia-inducible factor (HIF) and its down- stream targets such as erythropoietin (EPO) are being used for perfor- mance enhancement. HIF is the most significant regulator of O2 sensing and homeostasis,2 which was discovered in 19923 and subse- quently found to tightly regulate itself via degradation initiated by prolyl hydroxylase (PHD) enzymes.4–6 HIF is responsible for the tran- scription of many gene products involved in respiration, metabolism,

angiogenesis, erythropoiesis, and many other functions at the cellular and organ levels. HIF-α, the regulatory subunit of the HIF dimer, is constitutively produced. It is degraded by enzymatic hydroxylation on

proline residues leading to recognition of the hydroxyl form by the von Hippel–Lindau tumor suppressor protein (pVHL) and subsequent ubiquitin-mediated proteasomal destruction. The primary enzymes responsible for the hydroxylation of HIF are the three PHD isoforms: PHD1, PHD2, and PHD3. PHDs are members of the a-keto acid- dependent nonheme iron-containing family of hydroxylases. These dioxygenases require O2, iron, and an a-keto carboxylic acid, in this case, 2-oxoglutarate (2-OG) for activity and mediate the C4 trans-
hydroxylation of HIF-α initiating the path to protein degradation.7
Stabilization of HIF-1R through inhibition of PHD has been exam- ined as a potential treatment for ischemic diseases including anemia, myocardial infarction, and stroke in humans. In December 2018, Roxadustat has been the first-in-class PHD inhibitors to be approved globally for therapeutic use in humans, namely, for the long-term treatment of anemia in chronic kidney disease (CKD) patients.9 Although HIF stabilizers have promising therapeutic application, they are equally prone to misuse in amateur and elite sports due to their erythropoietic properties, as recently proven by several cases of adverse analytical findings in doping control testing. Doping with Roxadustat was first detected in 2015 and with Molidustat in 2017.8
Similar to roxadustat, IOX-2 is an inhibitor of prolyl-hydroxylase protein PHD. IOX-2 recently became a focal point in January 2020, due to its detection in a pair of Standardbred horses racing at Yonkers Raceway.9 For a drug that is neither approved for human nor for vet- erinary use to be found in a race horse’s system is not a surprise in this high-stakes game of cat and mouse and necessitates racing labo- ratories to include this substance in their ever expanding screening list.
Earlier this year, Görgens et al.10 performed a microdose elimina- tion study of IOX-2 in humans. Considering the potential misuse and lack of in vivo studies in horses, we decided to develop a sensitive and efficient liquid chromatography (LC)–high-resolution mass spec- trometry (HRMS) method for quantitative analysis of IOX-2 in equine urine and plasma and investigate its elimination profile after a single IV administration.

2 | EXPERIMENTAL

2.1 | Materials

IOX-2 (Figure 1) was purchased from Cayman Chemical (USA). Flunixin-D3 (Figure 2) used as internal standard was purchased from Sigma (Germany). Potassium hydroxide was purchased form Merck. Potassium phosphate monobasic, ammonium formate, and phosphoric acid were bought from Sigma Aldrich. Methanol, methyl tert-butyl ether, and formic acid were purchased from Fisher Chemicals. Bond Elut ABS Elut-Nexus 60 mg/3 ml was purchased from Agilent (USA). Deionized water was obtained from Thermo Scientific, GenPure ultra- pure water system.

FIG UR E 1 Structural formula of IOX-2

FIG UR E 2 Structural formula of Flunixin-D3

2.2 | Instrumentation

2.2.1 | LC conditions

A Dionex Ultimate 3000 RS liquid chromatographic system (Thermo Scientific, Germering, Germany) coupled to a Q-Exactive Orbitrap high-resolution mass spectrometer (Thermo Fisher Scientific, Bremen,
Germany) was used for analysis. Separation was achieved on a Zorbax Eclipse Plus C18, 150 × 2.1 mm i.d. column with 1.8-μm particle size (Agilent Technologies, USA) protected by the corresponding Zorbax Eclipse Plus C18, 5 × 2.1 mm i.d. guard column with 1.8-μm particle size. Chromatography was carried out with mobile phases consisting
of deionized water with 1-mM aqueous ammonium formate and 0.01% formic acid (solvent A) and methanol with 1-mM ammonium
formate and 0.01% formic acid (solvent B). The column compartment was maintained at 50◦C. A gradient elution program at a flow rate of
0.3 ml/min was employed in the following pattern: 0 to 0.5 min, 20%
B; from 0.5 to 10.0 min, to 95%B; during 4 min 95%B, from 14.0 to
14.1 min, to 20%B; 14.1 to 18.0 min, 20%B. Injection volume was 10 μl.

2.2.2 | MS conditions

High-resolution mass spectrometry (HRMS) was carried out with a Q-Exactive Orbitrap mass spectrometer equipped with a HESI source operating in positive voltage (3.5 Kv), and the capillary temperature
was set to 320◦C. The sheath and auxiliary gas were set to 45 and
15 units, respectively, were supplied from a specialty nitrogen genera- tor (Peak Scientific, Billerica, MA, USA). The s-lens RF level was set to 50, and the sweep gas flow rate was 0.
Data were obtained using parallel reaction monitoring (PRM) mode (Figure 3), with the mass resolution was set to 70,000 at m/z 200 with an automatic gain control (AGC) of 1e6. The maximum IT

FIG U R E 3 Electrospray ionization (ESI) product ion mass spectra of protonated IOX-2

was set to 50 ms. Isolation window was set at 1 m/z. All systems were controlled with Xcalibur 4.2.47 (Thermo Fischer).

2.3 | Drug administration and sample collection

Two horses, an Arabian mare & gelding, were selected for the study; prior to administration, the horses underwent physical examination, complete blood count and a serum biochemistry panel. Horses did not receive medication for 2 weeks prior to the commencement of the study. IOX-2 was administered as a single intravenous dose of
200 mg. Drug was administered in the right jugular vein using a 14-gauge catheter. Both the horses followed a regular exercise regi- men during the study period. Blood samples were collected by veni- puncture into a lithium heparin vacutainer. One set of blood and urine samples were collected prior to administration. Post administration blood samples were collected after 0.5, 1, 2, 4, 6, 8, 10, 12, 14, 16, 18, 24, 30, 36, 42, 48, 54, 60, 66, 72, 78, 84, 90, 96, 108, 120, 132, 144, 156, 168, 192, 216, and 240 h of administration. Urine samples were collected as free catch. Blood samples were centrifuged and plasma separated after collection. Both urine and plasma samples
were stored at −20◦C until analysis.

2.4 | Sample extraction

2.4.1 | Extraction of urine sample

One milliliter of urine was spiked with 0.3 ng of the ISTD Flunixin-D3 by adding 30 μl of a methanolic solution containing 10 ng/ml stock.

After gentle vortexing, the sample was diluted with 2 ml of 0.1-M phosphate buffer (pH 6.0) prior to SPE which was carried out using positive pressure manifold (UCT, Inc., Bristol). The sample was loaded onto an Agilent Bond Elut Nexus SPE cartridge that had been condi- tioned with 1 ml of methanol and 1 ml of water. After loading, the car- tridge was washed with 3 ml of water, and the excess water was removed by drying the column for 5 min. Elution was done with
1 × 2 ml of methanol. The eluate was dried under gentle stream of nitrogen at 45◦C and reconstituted in 40 μl of methanol followed by
60 μl of 1-mM ammonium formate in 0.01% formic acid, and the mix-
ture was transferred to a vial for analysis.
2.4.2 | Extraction of plasma samples
A 500 μL of plasma was spiked with 0.15 ng of the ISTD Flunixin-D3 by adding 15 μl of an aqueous solution containing 10 ng/ml stock. After gentle vortexing, 20 μl of 1-M phosphoric acid was added and
vortexed again. A 5 ml of methyl tert-butyl ether (MTBE) was added, and the tube was rotated at 40 rpm for 20 min. The sample was cen- trifuged at 5,500 rpm for 10 min, and the organic layer was trans- ferred to a fresh container. It was dried under gentle stream of
nitrogen at 45◦C and reconstituted in 40 μl of methanol followed by
60 μl of 1-mM ammonium formate in 0.01% formic acid, and the mix- ture was transferred to a vial for analysis.
2.5 | Method validation

Method is validated for use in equine plasma and urine for quantita- tive determination of IOX-2. Due to the unavailability of a deuterated
analogue, various trails were performed to select the most suitable internal standard. As most precise data were obtained using Flunixin-D3, it was used for the study. The guidelines used for
method validation and ion ratio criteria were based on “General
Accreditation Guidance — Validation and verification of quantitative and qualitative test methods” by NATA and “Guidelines for the Minimum Criteria for Identification by Chromatography and Mass Spectrometry” by AORC.

2.5.1 | Selectivity and specificity

Selectivity is the ability of the bioanalytical method to measure unequivocally and to differentiate the analyte(s) in the presence of components, which may be expected to be present. Typically, these might include metabolites, impurities, degradants, and matrix components.11
The selectivity of the method was established by investigating blank plasma and urine samples collected from six different horses for interfering peaks at the respective retention time of analyte and IS. Method specificity was ensured by confirming that the maximum difference in retention time of the reference standard and the test compound is within ±50% of the half height-peak width or 3 s, which- ever is larger and the mass of the diagnostic ions from the extracted ion chromatogram was identified within a mass resolution of
±5 ppm.12
2.5.2 | Linearity

Linearity was determined by analyzing matrix fortified calibration standards. Plasma calibration standards were prepared at the concentration of 0.25, 0.5, 1, 10, 25, 50, and 100 ng/ml, and urine calibration standards were prepared at the concentration of 0.125, 0.25, 0.5, 1, 10, 25, 50, and 100 ng/ml. A calibration model was established using seven replicates at each concentration level.
The presence of heteroscedasticity was determined by calculat- ing the probability that the variance of measurements at the highest calibration standard was equal to or smaller than the variance of measurements at the lowest calibration standard using F-test value.13
Variance evaluation was performed by first applying each weighting scheme to the measurements to calculate the concentration levels normalized weighted variances, which were then used to calcu- late the total normalized weighted variance.14
Relative error and lack of fit test was performed for re- gression model selection. After choosing the calibration model, Kolmogorov–Smirnov (KS) was used to test whether the standard- ized residual distribution is significantly different from normal distribution.
Once the calibration model was established and validated, same model and single standard points were used for further analysis.15

2.5.3 | Intraday and interday precision and accuracy (bias)

Intraday precision and accuracy were evaluated by analyzing seven matrix spiked replicates at three different concentrations levels on the same day.
Interday precision and accuracy were evaluated by analyzing seven matrix spiked replicates at three concentration levels on three different days. The matrix spikes were prepared at 0.5, 10, and 50 ng/ml levels for both plasma and urine.
The accuracy of a measurement result describes how close the result is to its true value and therefore includes the effect of both pre- cision and trueness (expressed in the form of bias).16 In this paper, the
term accuracy will be used in the sense of bias “The difference
between the expectation of the test results and an accepted reference value”.17

2.5.4 | Limit of detection

The estimated limit of detection (LOD) for a target analyte represents the lowest evaluated spiked concentration that gave a signal to noise ratio greater than 3:1 in the product ion chromatogram. We took a stringent approach to consider LOD where along with the above criteria, the ion ratios of the diagnostic ions should also match. Thus, the actual LOD could be much lower.
2.5.5 | Stability and carryover

Freeze thaw stability of IOX-2 in both urine and plasma was deter- mined in triplicates at 10 ng/ml following three freeze/thaw cycles.
All samples were stored at −20◦C prior to analysis. Data from the
freshly spiked sample were compared with the frozen samples.
The stability of the reconstituted samples, stored in vials in the auto sampler tray at 12◦C, was determined after 19 hours. Corresponding data were compared with the freshly reconstituted
sample.
Carryover was assessed by analyzing blank sample after mid QC of 10 ng/ml.

2.5.6 | Dilution integrity

To demonstrate the ability to dilute and analyze samples with IOX-2
concentration above the highest calibrator, plasma samples spiked with IOX-2 (8 μg/ml) in triplicates were diluted hundred times with blank plasma (to 80 ng/ml), and urine samples spiked with IOX-2 (30 μg/ml) were serially diluted thousand times (to 30 ng/ml) with blank urine prior to sample preparation. To estimate the dilution preci-
sion and accuracy, calculated mean concentration from the analysis was further converted to the original sample concentrations and com- pared with nominal concentration.

3 | RESULTS

3.1 | Method validation

3.1.1 | Selectivity and specificity

Chromatogram of blank plasma and urine samples collected from six different horses did not show any interfering peaks at the respective retention time of analyte and ISTD. The relative retention time (rRT) of the analyte in the test sample did not vary from that in the refer- ence sample and thereby fulfilled the AORC requirement. The method demonstrated a rRT CV% of <0.04% and mass of the diagnostic ions from the extracted ion chromatogram was identified within a mass resolution of ±5 ppm.

3.1.2 | Linearity & LOD

Calibration curve was constructed with seven fortification levels for plasma and eight fortification levels for urine. The calibration model was established using seven replicates at each concentration level. The presence of heteroscedasticity was determined by using an F-test. P value of 1.3e-15 and 2.16e-13 was observed for urine and plasma, respectively, indicating that the data sets are heteroscedastic

and weighting must be applied. Variance test scores for urine and plasma confirmed that a 1/x2 weighting factor is most suitable for the calibration model. Based on the results from the lack of fit (LOF) and percent relative error, the final calibration model of linear 1/x2 for plasma and quadratic 1/x2 model for urine was chosen. The calibra- tion model was validated using KS normality testing of the standard- ized residuals. KS test provide Dn value of 0.095 for urine and 0.1 for plasma, which is smaller than Dn,a (0.2, where a = 0.05), confirming that there is no significant difference between the data and data that are normally distributed. Data are provided in Tables 1 and 2.
The estimated LOD in plasma was 0.075 ng/ml and in urine was
0.025 ng/ml.

3.1.3 | Intraday and interday precision and accuracy (bias)

Precision and accuracy were determined at 0.5, 10, and 50 ng/ml levels for both plasma and urine. The intraday and interday precision
were determined as per the calculation provided in “Scientific
Working Group for Forensic Toxicology (SWGTOX) Standard Prac- tices for Method Validation in Forensic Toxicology”.18 Intraday preci- sion for plasma and urine were within 15% and accuracy in plasma
and urine were within ±15%. Data are provided in Table 3.

TA BL E 1 Factors determining weight selection for equine urine and plasma

Equine urine (n = 7 per level) Equine plasma (n = 7 per level)
Heteroscedasticity check F test P(F ≤ f) one-tail 1.30e-15 2.16e-13
Weighting None 1/x 1/x2 None 1/x 1/x2
Normalized weighted variance 111.02 1.04e-02 6.20e-05 0.15 1.32e-05 3.93e-07

TA BL E 2 Factors determining the selection of curve fitting procedure for equine urine and plasma

Equine urine (n = 7 per level) Equine plasma (n = 7 per level)
Weighting 1/x 1/x2 1/x 1/x2 1/x 1/x2 1/x 1/x2
Fitting Linear Linear Quadratic Quadratic Linear Linear Quadratic Quadratic
Σ % relative error 38% 41% 34% 26% 52% 51% 52% 64%
Percentage of standards 100% 100% 100% 100% 100% 100% 87.5% 87.5%
that passed
P value for lack of fit 0.2 0.7 1.0 1.0 0.8 0.7 0.6 0.2
TA BL E 3 Interday and intraday precision and accuracy

Matrix
Spiked concentration (ng/ml) Intraday (n = 7) Precision (%) Interday (three sets, n = 7 per set) Precision (%)
Average accuracy (%)
.4 | Stability and carryover

Concentration of IOX-2 in freshly spiked samples and samples after freeze thaw cycle and auto sampler stability is provided in Tables 4 and 5, respectively. P values determined using t-test were greater
than 0.05 indicating with 95% confidence that the concentrations of the freshly prepared samples and samples assayed after freeze thaw cycles were the same. Similar results were observed for the stability in auto sampler.
Less than 1% of carryover was observed in urine and plasma.

TA BL E 4 Freeze thaw stability of IOX-2 in equine urine and plasma

Days in storage No. of freeze thaw cycle Sample type (n = 3 per cycle) Spiked conc. (ng/ml) P value (a = 0.05)

TA BL E 5 Auto-sampler stability of IOX-2 in equine urine and plasma

TA BL E 6 Effect of dilution on bias and precision

Matrix Original concentration (μg/ml) (n = 3) Nominal concentration after dilution (ng/ml)
Calculated concentration in the diluted sample (ng/ml) Mean calculated concentration in the diluted sample (ng/ml)
Accuracy (%)
Precision (%RSD)
Plasma 8 80 72 74 93 4.53
FIG U R E 4 Elimination profile of IOX-2 in equine plasma

FIG U R E 5 Elimination profile of IOX-2 in equine urine

FIG U R E 6 Extracted ion chromatogram from the parallel reaction monitoring (PRM) mode showing the confirmation ions and their respective ion ratios in (A) preadministration plasma, (B) plasma spiked with 10 ng/ml IOX-2, and (C) plasma 24 h post administration

3.1.5 | Dilution integrity

Precision and accuracy of diluted samples was less than 15% and within ±15% of the nominal concentration. Summary of dilution accu- racy and precision is provided in Table 6.

3.2 | Post administration

Elimination profile of IOX-2 in equine plasma and urine after a single IV administration is provided in Figures 4 and 5, respectively. In post administration plasma, IOX-2 peak concentration was

2 μg/ml, observed at 30 min. The concentration dropped below
the LOD after 54 h. In post administration urine, the peak concen- tration was 25 μg/ml observed at 7 h, which dropped below LOD after 151 h.
Representative extracted ion chromatograms (EIC) showing the presence of IOX-2 in post administration plasma and urine samples are shown in Figures 6 and 7. Plasma and urine samples collected prior to administration did not show any response for IOX-2.
Görgens et al10 reported the detection of IOX-2 and its metabo- lites in urine of humans for doping control purposes and their respec- tive fragments. Similar metabolic profile was observed in equine urine, after intravenous administration of IOX-2, with the phase II metabo- lites being IOX-2 glucuronide and hydroxylated IOX-2 glucuronide. However, in equine plasma, the only metabolite that could be identi- fied was hydroxylated IOX-2. These metabolites were identified by the accurate mass of the protonated molecule and their corresponding product ions (Table 7). Peaks for hydroxylated IOX-2 were observed at three different RTs (Figure 8) indicating different locations of hydroxylation. Three chromatographically unresolved peaks were observed for glucuronides (Figure 9), whereas two sets of peaks were obtained for hydroxy glucuronides (Figure 10). A single peak at 9.0 and three chromatographically unresolved peaks at 9.9. The product ion spectra of hydroxylated IOX-2, IOX-2 glucuronide, and hydroxylated IOX-2 glucuronide are given in Figures 11, 12, and 13, respectively. Although these metabolites could be identified with their distinguishing fragments, due to the absence of their reference standards they were not quantified. A plot of the ratio of the area of the metabolite to the area of ISTD against time is given in Figure 14.

FIG U R E 7 Extracted ion chromatogram from the parallel reaction monitoring (PRM) mode showing the confirmation ions and their respective ion ratios in (A) preadministration urine, (B) urine spiked with 10 ng/ml IOX-2, and (C) urine 47 h post administration

Moreover, the relative abundance of their corresponding peaks was not only lower in both matrices but also their detection window was shorter compared with parent IOX-2. The hydroxylated IOX-2 could be detected in plasma and urine for 16 and 55 h, respectively, post administration. The phase II metabolites glucuronidated IOX-2 and

hydroxylated IOX-2 glucuronide were detected in urine only for 33 h post administration. Therefore, we conclude that IOX-2 predomi- nantly exists in free form in plasma and urine of equine athletes and parent IOX-2 could be used to control the abuse of IOX-2 in equine sports. Hence, no further attempt was made to elucidate the isomeric

TA BL E 7 Accurate mass of the precursor and the product ions

Precursor Product ions

Elemental Theoretical m/z Observed m/z Δ Theoretical m/z Observed m/z Δ
Analyte composition [M + H]+ [M + H]+ ppm [M + H]+ [M + H]+ ppm
IOX-2 C19H16N2O5 353.1132 353.11227 −2.6 296.09173 296.09109 −2.2
278.08117 278.08057 −2.2
307.10772 307.10693 −2.6
Hydroxy- IOX-2 C19H16N2O6 369.10811 369.10751 −1.6 312.08665 312.08752 2.8
294.07608 294.07523 −2.9
323.10263 323.10172 −2.8
IOX-2 glucuronide C25H24N2O11 529.14529 529.14478 −1.0 307.10772 307.10696 −2.5
296.09173 296.09113 −2.0
278.08117 278.08051 −2.4
Hydroxy-IOX-2 C25H24N2O12 545.1402 545.13965 −1.0 312.08665 312.086 −2.1
glucuronide 294.07608 294.07544 −2.2
323.10263 323.10202 −1.9

FIG U R E 8 Extracted ion chromatogram of protonated hydroxylated IOX-2 and its confirmation ions in post administration urine

FIG U R E 9 Extracted ion chromatogram of protonated IOX-2 glucuronide and its confirmation ions in post administration urine

FIG U R E 10 Extracted ion chromatogram of protonated hydroxylated IOX-2 glucuronide and its confirmation ions in post administration urine

FIG U R E 11 Electrospray ionization (ESI) product ion mass spectra obtained from parallel reaction monitoring (PRM) experiment of protonated hydroxylated IOX-2 in post administration urine

FIG U R E 12 Electrospray ionization (ESI) product ion mass spectra obtained from parallel reaction monitoring (PRM) experiment of protonated IOX-2 glucuronide in post administration urine

metabolite peaks. The present study is based on a single intravenous administration and the metabolite profile, and their detection window might be different had the drug been administered orally, requiring further studies.

The dosage used in the current study was approximately
0.4 mg/kg, which is relatively low hence little or no pharmacological and performance enhancing effects are expected in horse. In spite of such a small dose, a current method is capable of detecting IOX-2 for

FIG U R E 13 Electrospray ionization (ESI) product ion mass spectra obtained from parallel reaction monitoring (PRM) experiment of protonated hydroxylated IOX-2 glucuronide in post administration urine

FIG UR E 14 Excretion profiles of hydroxylated IOX-2 in plasma and hydroxylated IOX-2, IOX-2 glucuronide, and hydroxylated IOX-2 glucuronide in urine

up to 54 h in plasma and up to 151 h in urine. Therefore, the devel- oped method is suitable to control any potential misuse in horse racing.

4 | CONCLUSION

A sensitive and reliable LC–MS–MS method was successfully devel- oped and validated for the detection and quantification of IOX-2, a

selective inhibitor of prolyl-hydroxylase protein PHD, in equine urine and plasma. The method was developed using ultra high-performance liquid chromatography coupled to high-resolution accurate mass spec- trometer. The developed method was also used to successfully ana- lyze plasma and urine samples collected following a single-dose intravenous administration. The proposed protocol had acceptable ranges of accuracy and repeatability. Low-estimated LOD proves high sensitivity of the method. In both plasma and urine, the detection win- dow of IOX-2 is longer, and its abundance is much higher than the
metabolites. Therefore, IOX-2 predominantly exists in free form in equine urine and plasma and the parent molecule remains the analyte of choice to be monitored for any potential abuse in horse racing. In spite of the dosage being small, it was sufficient to allow the detection and quantification of IOX-2 for a reasonable period of time after administration IOX2

ACKNOWLEDGEMENTS
This research was funded by Zabeel Equestrian office, under the patronage of H. H Sheikh Hamdan bin Rashid Al Maktoum. The authors are thankful to Mr Humaid Saeed Muroshed Director, Zabeel Equestrian Office for valuable suggestions and support. The authors acknowledge the support and assistance of Mr Saravana Kumar Natarajan in sample collection and storage.

CONFLICT OF INTEREST
The authors declare that there is no conflict of interest.

ETHICAL APPROVAL
The study was approved by Animal Welfare and Protection commit- tee of Zabeel Equestrian Office, Dubai, UAE.

ORCID
Ezra Mikhail https://orcid.org/0000-0001-7231-2650

REFERENCES
1. Federation for Equestrian Sports (FEI). 2020 Equine Prohibited Sub- stances List. (n.d.) Available at: https://inside.fei.org/sites/default/ files/2020%20Prohibited%20Substances%20List_0.pdf
2. Wang GL, Jiang BH, Rue EA, Semenza GL. Hypoxia-inducible factor 1 is a basic-helix-loop-helix-PAS heterodimer regulated by cellular O2 tension. Proc Natl Acad Sci U S A. 1995;92(12):5510-5514. https:// doi.org/10.1073/pnas.92.12.5510
3. Semenza GL, Wang GL. A nuclear factor induced by hypoxia via de novo protein synthesis binds to the human erythropoietin gene enhancer at a site required for transcriptional activation. Mol Cell Biol. 1992;12(12):5447-5454. https://doi.org/10.1128/mcb.12.12.5447
4. Ivan M, Kondo K, Yang H, et al. HIF alpha targeted for VHL-mediated destruction by proline hydroxylation: implications for O2 sensing. Science. 2001 Apr 20;292(5516):464-468. https://doi.org/10.1126/ science.1059817
5. Jaakkola P, Mole DR, Tian YM, et al. Targeting of HIF-alpha to the von Hippel-Lindau ubiquitylation complex by O2-regulated prolyl hydroxylation. Science. 2001 Apr 20;292(5516):468-472. https://doi. org/10.1126/science.1059796
6. Bruick RK, McKnight SL. A conserved family of prolyl-4-hydroxylases that modify HIF. Science. 2001 Nov 9;294(5545):1337-1340. https:// doi.org/10.1126/science.1066373
7. Rabinowitz MH, Barrett TD, Rosen MD, Venkatesan H. Inhibitors of HIF prolyl hydroxylases. In: John E. Macor, ed. Annual Reports in

Medicinal Chemistry. 45. San Diego, USA. Elsevier Inc.; 2010:123-139. https://doi.org/10.1016/S0166-526X(10)56008-3
8. Sciesielski LK, Kirschner KM. ExActa HIF prolyl hydroxylase inhibitors—the new lifestyle drug? Acta Physiologica. 2019;227(3): 123-139, e13370. https://doi.org/10.1111/apha.13370
9. Thoroughbred daily news. Synthetic doping agent detected in two standardbreds. Available at: https://www.thoroughbreddailynews. com/synthetic-doping-agent-detected-in-two-standardbreds (Accessed: 23 January 2020).
10. Görgens C, Guddat S, Bosse C, Knoop A, Geyer H, Thevis M. Imple- mentation of the HIF activator IOX-2 in routine doping controls—pilot study data. Drug Test Anal. 2020 Aug 18;12(11-12):1614-1619. https://doi.org/10.1002/dta.2914
11. Shah VP, Midha KK, Findlay JW, et al. Bioanalytical method validation—a revisit with a decade of progress. Pharm Res. 2000 Dec; 17(12):1551-1557. https://doi.org/10.1023/a:1007669411738
12. Association of Official Racing Chemists (AORC). Guidelines for the minimum criteria for identification by chromatography and mass spectrometry. 2015. http://www.aorc-online.org/documents/aorc- ms-criteria-modified-23-aug-16/
13. International Organization for Standardization (ISO). Water quality— calibration and evaluation of analytical methods and estimation of performance characteristics part 1: statistical evaluation of the linear calibration function. 1990:8466–1. Available at: https://www.iso.org/ obp/ui/#iso:std:iso:8466:-1:ed-1:v1:en
14. Desharnais B, Camirand-Lemyre F, Mireault P, Skinner CD. Procedure for the selection and validation of a calibration model I—description and application. J Anal Toxicol. 2017;41(4):261-268. https://doi.org/ 10.1093/jat/bkx001
15. Peters FT, Maurer HH. Bioanalytical method validation and its impli- cations for forensic and clinical toxicology—a review. Accreditation and Quality Assurance. 2002;7(11):441-449. https://doi.org/10.1007/ s00769-002-0516-5
16. National Association of Testing Authorities (NATA). General accreditation guidance—validation and verification of quantitative and qualitative test methods. 2018. https://www. nata.com.au/phocadownload/gen-accreditation-guidance/Validation- and-Verification-of-Quantitative-and-Qualitative-Test-Methods.pdf
17. International Organization for Standardization (ISO). Accuracy (true- ness and precision) of measurement methods and results. 1994; 5725-1 to 5725-3. Available at: https://www.iso.org/obp/ui/#iso:std: iso:5725:-1:ed-1:v1:en
18. Scientific Working Group for Forensic Toxicology. Standard practices for method validation in forensic toxicology. J Anal Toxicol. 2013; 37(7):452-474. https://doi.org/10.1093/jat/bkt054