M ethodology For reprint orders, please contact [email protected]
Detection and quantification of dermorphin and selected analogs in equine urine Background: Dermorphin, a hepta-peptide with potent analgesic properties, is classified as a doping agent in equine racing. Since its discovery, a number of biologically active structural analogs have been synthesized and made commercially available so there is a need for reliable methods of detection. Methodology/Results: A sensitive detection method was developed for dermorphin and six analogs in equine urine. Peptide enrichment was achieved using weak cation exchange with subsequent separation and detection by nano-UHPLC–MS/MS. Method validation parameters included: specificity, linearity (5–10000 pg/ml), recovery (58–93%), intra and inter-assay repeatability, LOD (5–50 pg/ml) and matrix effects. Conclusion: The presented method will facilitate the control of the abuse of dermorphin and selected analogs in equine sports. Dermorphin and [Hyp 6 ]-dermorphin were first
isolated in 1981 from the skin of frogs (genus Phyllomedusa) found in South America [1,2]. These peptides are potent analgesics that have selective affinity for μ-opioid receptors [1,3,4]. With no accepted medical use and the pharmac ological potential to alter the performance of an animal, the use of these peptides is now prohibited in horse racing . It is envisaged that low-level detection of these peptides will be necessary as they exhibit a potency of 30–40-times greater than morphine with respect to their pharmacological effect . The detection of low levels of these peptides in equine urine, along with ‘designer’ analogs, requires sensitive analytical techniques that provide unequivocal identification of the peptide. The minimal sequence, as identified by structure–activity relationship studies of dermorphin analogs, required for agonistic activity at µ-opioid receptors is the N-terminal tetrapeptide  and the presence of a d-isomer amino acid (d-Ala) located between the first and third amino acid residue, is essential for this agonistic activity . The rat model has shown that cleavage enzymes in the blood, liver, brain and spinal cord degrade peripherally administered dermorphin to the deamidated N-terminal tetra-peptide (dermorphin-[1–4]-OH)  and administration to six horses indicated that dermorphin was detected in plasma up to 4 h post-intravenous administration, up to 6 h postintramuscular administration and, in urine, up to 36 h post-intravenous administration . To
date, there is no data published detailing the pharmacological fate of dermorphin in the horse. Dermorphin and its analogs were found to retain effectiveness as µ-opioid receptor agonists after peripheral injection in mice . Since the discovery of dermorphin, a large number of analogs have been synthesized and their structure–activity relationships have been established using standard animal models [1,4]. It has been proposed that dermorphin analogs containing C-terminal amidation may have greater agonistic activity for µ-opioid receptors , and that with replacement of the d-Ala 2 residue with d-Arg 2 results in tetra-peptides resistant to enzymatic metabolism . Although investigations into the effectiveness of dermorphin and its analogs as analgesics have been extensively conducted, they have not been employed as pharmaceutical agents for therapeutic purposes. They are, however, readily available from commercial sources and have the potential to be employed as a doping agent in equine racing. Enrichment of peptides and proteins is often achieved through the use of antibodies, however, production of antibodies for peptides smaller than eight amino acids is uncommon, since small peptides tend not to be immunogenic [10,11]. To enhance immune response conjugation of peptides to carrier proteins may be used, however this creates additional expense. For laboratories to respond promptly to the emergence of new peptides and synthetic analogs in the sporting arena, alternative approaches must be used to isolate these small peptides from biological matrices. In 2011, suspicion arose as to the illicit use of dermorphin in the USA, and in response
10.4155/BIO.13.281 © 2013 Future Science Ltd
Bioanalysis (2013) 5(24), 2995–3007
Stacey L Richards*1,2 , Adam T Cawley1 & Mark J Raftery3 Australian Racing Forensic Laboratory, Racing NSW, Sydney, NSW, Australia 2 School of Biotechnology & Biomolecular Sciences, University of NSW, Sydney, NSW, Australia 3 Bioanalytical Mass Spectrometry Facility, University of NSW, Sydney, NSW, Australia *Author for correspondence: E-mail: [email protected]
M ethodology | Key Terms Dermorphin: Natural opioid agonist hepta-peptide that binds to and activates opioid receptors.
µ-opioid receptors: One of
four major subtypes of opioid receptors with activation resulting in analgesia, respiratory depression, euphoria and physical dependence.
Animal models: Depression
of electrically evoked contractions of mouse vas deferens or guinea pig ileum myenteric plexus-longitudinal muscle preparations have been used as sensitive models for studying the pharmacology of µ-opioid, d-opioid and k-opioid agents.
Richards, Cawley & Raftery Guan et al. developed an analytical method for the detection and quantification of dermorphin in equine plasma and urine . This procedure is based on a mixed-mode SPE technique combining reverse-phase and strong cation exchange chromatography. Kwok et al. published the extraction of dermorphin along with other bioactive peptides from equine plasma employing mixed-mode strong anion exchange chromatography . In addition, Wang et al. investigated the use of mixed-mode strong cation exchange chromatography to isolate dermorphin and Hyp 6 -dermorphin from equine plasma . Preliminary investigation of these procedures for the detection of dermorphin analogs indicated that while both strong cation exchange and strong anion exchange are exceptionally efficient at isolating dermorphin from equine urine and plasma, they are not ideal for the isolation of all dermorphin analogs targeted in this study. This paper presents a simple SPE procedure combining reverse-phase and weak cation exchange chromatography to isolate dermorphin and a number of dermorphin analogs from equine urine. Materials & methods Chemicals & materials [d-Ala 2]-dermorphin, [l-Ala 2]-dermorphin, [Hyp 6 ]-dermorphin, dermorphin analog, dermorphin analog (1–4) and [d-Arg 2]-dermorphin (1–4) tetra-peptide amide, with purity ≥95%, were purchased from Phoenix Pharma ceutical Inc. (CA, USA). [d-Arg 2 ,Lys 4]-dermorphin (1–4) amide and the IS deltorphin-1, with purity ≥95%, were purchased from Synbiosci Corporation (CA, USA). Primary solutions of each peptide at 1.0 mg/ml were prepared in acetonitrile/water/acetic acid (50:50:0.1, v/v/v). Acetic acid (100%), formic acid (>98%) and trifluoroacetic acid (>98%) were from Sigma-Aldrich (Sydney, Australia), and heptafluorobutyric acid (HFBA) was purchased from Thermo Scientific (Victoria, Australia). Orthophosphoric acid (≥85%), dipotassium hydrogen orthophosphate and potassium dihydrogen orthophosphate were purchased from Ajax Fine Chemicals (NSW, Australia). Ammonia (25%), methanol (LC grade) and acetonitrile (LC grade) were from Merck (Darmstadt, Germany), and urea (PlusOne, >99.5%) was purchased from GE Healthcare (NSW, Australia). The Oasis ® WCX cartridges (30 mg, 1 ml) were purchased from
Bioanalysis (2013) 5(24)
Waters (NSW, Australia). Deionized water was produced from a water purification system (Milli-Q®, Molsheim, France). Calibration
standards & QC samples A combined stock solution containing all peptide targets was prepared at 1 ng/µl in acetonitrile/water/acetic acid (20:80:0.1, v/v/v) and stored for 1 month at -20°C. Spiking solutions were prepared daily by dilution of the stock solution. Equine urine samples were spiked at 5, 10, 25, 50, 100, 250, 500, 1000, 2500, 5000, 10,000 and 20,000 pg/ml, with calibration curves prepared in seven different lots of equine urine. The IS (deltorphin-1) was added to achieve 250 pg/ml. QC samples with each analog at 25, 250 and 2500 pg/ml in urine were similarly prepared. Initial investigations indicated that the instrument response to [Hyp 6 ]-dermorphin was significantly lower than the other target peptides. As such, the QC samples of [Hyp 6 ]-dermorphin employed for validation of the method were prepared at concentrations of 50, 500 and 5000 pg/ml. Sample
preparation Urine (1 ml) in a glass culture tube (13 × 100 mm) was spiked with deltorphin-1 (250 pg) as an IS. The Oasis ® WCX cartridges were preconditioned with methanol (1 ml), water (1 ml) and pH 6 phosphate buffer (0.05 M, 0.5 ml). The urine samples were diluted with H 3PO 4 (4%, 0.5 ml) and urea (8 M, 0.5 ml) prior to loading onto the preconditioned Oasis WCX cartridges. The purpose of pretreatment of urine with urea was to denature proteins present in the sample that may have bound to the target peptides. The cartridge was washed with aqueous ammonia (5%, 1 ml), aqueous methanol (20%, 1 ml) and dried for 1 min. The aqueous ammonia wash assists in removing acidic matrix from the sorbent while retaining the peptides weakly bound by hydrophobic interactions, and enhances the ionic interaction of the positively charged peptides with the negatively charged sorbent. The cartridge was eluted with 0.5 ml of acetonitrile/water/formic acid (75/20/0.1, v/v/v) into a microcentrifuge tube. The eluate was evaporated to dryness at 40°C for 1.5 h using a vacuum concentrator (SpeedVac™ Plus, Savant). The residue was reconstituted in 100 µl of aqueous reconstitution solvent (1% formic acid/0.05% HFBA) and sonicated for 5 min followed by centrifugation at future science group
Detection & quantification of dermorphin & selected analogs in equine urine 13000 × g for 10 min. HFBA increases the peptide hydrophobicity through protonation of the carboxylic groups and formation of ion pairs with basic residues . A 90 µl filtrate was then transferred to an autosampler vial for nano-UHPLC–MS/MS analysis. Unless otherwise stated, all the apparatus used were plasticware. Nano
UHPLC–MS/MS instrumentation All UHPLC–MS/MS analyses were performed on a TSQ Vantage triple quadrupole mass spectrometer (Thermo Scientific) with a nano electrospray source coupled to a Dionex® Ultimate 3000 nano UHPLC with a Switchos loading pump (LC Packings, Idstein, Germany). The LC separation was carried out on a capillary column, packed in-house using a Waters BEH C18 (100 × 0.1 mm I.D., 1.7 µm), with an Acclaim PepMap100 (C18, 5 × 0.3 mm I.D., 5 µm, 100 Å) trap. The mobile phase was composed of 0.1% formic acid in distilled water (solvent A) and 0.1% formic acid in 80/20 acetonitrile/distilled water (solvent B). The 20 µl injection volume was delivered to the trap column at 15 µl/min using an aqueous solution of acetonitrile and TFA (2%:0.25% v/v) via the loading pump. The solute was concentrated on the PepMap™ trap while salts and other matrix components were flushed to waste for 4 min. Switching the ten-port valve under gradient elution (0.25 µl/min flow rate) allowed the enriched peptides to be back flushed onto the analytical column. The linear gradient program used is presented in Table 1. Between injections, the autosampler needle was washed with 200 µl of methanol:water:formic acid (10:90:0.1, v/v/v). Detection of the target peptides was performed by ESI+ in SRM mode with the spray voltage set to 2000 V and a capillary temperature of 290°C. The scan time and scan width respectively were set to 0.02 s and 0.01 m/z and the isolation width of both Q1 and Q3 were set at 0.7 m/z. The SRM transitions, S-Lens voltage and collision energies were optimized by infusion of the target peptides into the TSQ Vantage. The instrument was calibrated using the manufacturer’s calibration solution (polytyrosine-1,3,6) with argon as the collision gas. The selected characteristic ion transitions and optimized MS parameters are summarized in Table 2 . Transitions in bold type were used for quantification. future science group
| M ethodology
performance assessment The assay was validated for quantitative purposes in accordance with the guidelines outlined by the National Association of Testing Authorities Technical Note  for specificity, sensitivity, linearity, accuracy, precision, recovery, matrix effects and analyte stability. Analyzing extracts, in duplicate, from five different sources assessed the assay selectivity. No endogenous peaks containing all SRM transitions, at the expected retention time of the target peptides, were observed in any of the urine lots evaluated. In addition, the procedure proposed by Matuszewski et al.  was used to assess the crosstalk between the monitored MS/MS channels. Specifically, each peptide was injected separately at the highest QC concentration (2500 pg/ml), and the response in all other channels was monitored. A blank urine sample spiked only with the IS was injected and the response in the target analyte channels was monitored. No crosstalk was observed. Sensitivity was evaluated by establishing a LOD and LLOQ for each analyte. The LOD was defined as the lowest concentration of analyte that met the following criteria: all SRM transitions with a S/N ratio of >3; relative abundance of three SRM transitions within 10% absolute or 30% relative, when compared with a reference standard run contemporaneously and the LC relative retention time tolerance of ± 2% . The LLOQ was the lowest concentration that met the LOD criteria but with a S/N ratio of ten, together with acceptable accuracy (80–120%) and precision (40%) indicates significant relative matrix interference with this assay. This finding indicates potentially limited robustness of the method across different sources of equine urine with regard to quantification of the exact levels of these peptides. The consequences of this enhanced effect on the analytical results will depend on whether the effect is more pronounced in the blank urine used for construction of the calibration curve or in the suspect equine urine samples being analyzed by the laboratory. In the former situation, the determined concentrations in the sample urines will be artificially low, while in the latter situation the determined concentrations will be artificially high. Ideally, any blank urine samples being utilized by the laboratory for construction of the calibration curve should be assessed for matrix interference prior to being employed in these procedures. While it is undesirable to have matrix interference occurring, the intra- and inter-individual variation between horse urine means that it is impossible to eliminate it entirely. The use of isotopically labeled analogs would assist in reducing this level of interference, however, as these peptides are exogenous to horses, the very presence of any confirmable level would indicate prohibited use.
| M ethodology
40 20 0
Figure 4. Product ion chromatograms of [ d -Arg2,Lys4 ]-dermorphin (1-4) amide. Obtained from analysis of (A) urine sample spiked with [d-Arg2,Lys4]-dermorphin (1-4) amide at 10 pg/ml; and (B) [d-Arg2,Lys4]dermorphin (1-4) amide reference standard. The relative chromatographic peak ratios, as percentages, are listed for each peak.
d-dermorphin and the analog peptides were found to be stable when stored at -80°C for 30 days with losses of