1
Department of Physics, Chemistry and Biology
Final Thesis
Determination of testosterone esters in serum by liquid
chromatography – tandem mass spectrometry
(LC-MS-MS)
Erica Törnvall
Final Thesis performed at
National Board of Forensic Medicine
2010-06-03
LITH-IFM-EX--10/2263--SE
Department of Physics, Chemistry and Biology Linköping University
2
Department of Physics, Chemistry and Biology
Determination of testosterone esters in serum by liquid
chromatography – tandem mass spectrometry
(LC-MS-MS)
Erica Törnvall
Final Thesis performed at
National Board of Forensic Medicine
2010-06-03
Supervisors
Yvonne Lood
Martin Josefsson
Examiner
Roger Sävenhed
3 Datum Date 2010-06-03 Avdelning, institution Division, Department Chemistry
Department of Physics, Chemistry and Biology Linköping University
URL för elektronisk version
ISBN
ISRN: LITH-IFM-EX--10/2263--SE
_________________________________________________________________
Serietitel och serienummer ISSN
Title of series, numbering ______________________________
Språk Language Svenska/Swedish Engelska/English ________________ Rapporttyp Report category Licentiatavhandling Examensarbete C-uppsats D-uppsats Övrig rapport _____________ Titel Title
Determination of testosterone esters in serum by liquid chromatography – tandem mass spectrometry (LC-MS-MS) Författare Author Erica Törnvall Nyckelord Keyword
LC-MS-MS, MRM, testosterone, testosterone esters
Sammanfattning
Abstract
Anabolic androgenic steroids are testosterone and its derivates. Testosterone is the most important naturally existing sex hormone for men and is used for its anabolic effects providing increased muscle mass. Testosterone is taken orally or by intramuscular injection in its ester form and are available illegally in different forms of esters. Anabolic androgenic steroids are today analyzed only in urine. To differentiate between the human natural testosterone and exogenous supply the quote natural testosterone and epitestosterone is used. Detection of testosterone esters in serum is an unmistakable proof of exogenous supply of testosterone. The aim of this thesis was to find a method for determining testosterone esters in serum and to study an extraction method possible for quantification of testosterone esters in serum.
The technique used to separate and identify the Testosterone esters was Liquid Chromatography Tandem Mass Spectrometry Electro Spray Ionisation. Parameters for chromatography and mass detection were optimized for nine testosterone esters and evaluated according to selectivity, resolution and intensity. A method that could be used for determination of testosterone esters in serum was found. The MS-method was set and at least three possible transitions for each testosterone ester were found. The best choice of column proved to be the C18 column where all the esters were separated and seven of them were base-line separated. The C18 column along with methanol and ammonium acetate buffer, 5 mM, pH 5 showed the highest sensitivity for Multiple Reaction Monitoring-detection. A gradient profile for a total runtime of 5.6 minutes was established. Two alternative extraction procedures were tested, with tert-butylmethylether or diethyl ether/ethyl acetate and both seemed to work satisfactory. Analysis of serum proved to work well and no severe interference occurred. Results from the linearity tests indicate that future quantification method in serum will be possible.
4
Abstract
Anabolic androgenic steroids are testosterone and its derivates. Testosterone is the most important naturally existing sex hormone for men and is used for its anabolic effects
providing increased muscle mass. Testosterone is taken orally or by intramuscular injection in its ester form and are available illegally in different forms of esters. Anabolic androgenic steroids are today analyzed only in urine. To differentiate between the human natural
testosterone and exogenous supply the quote natural testosterone and epitestosterone is used. Detection of testosterone esters in serum is an unmistakable proof of exogenous supply of testosterone. The aim of this thesis was to find a method for determining testosterone esters in serum and to study an extraction method possible for quantification of testosterone esters in serum.
The technique used to separate and identify the testosterone esters was Liquid
Chromatography Tandem Mass Spectrometry Electro Spray Ionisation. Parameters for
chromatography and mass detection were optimized for nine testosterone esters and evaluated according to selectivity, resolution and intensity. A method that could be used for
determination of testosterone esters in serum was found. The MS-method was set and at least three possible transitions for each testosterone ester were found. The best choice of column proved to be the C18 column where all the esters were separated and seven of them were base-line separated. The C18 column along with methanol and ammonium acetate buffer, 5 mM, pH 5 showed the highest sensitivity for Multiple Reaction Monitoring-detection. A gradient profile for a total runtime of 5.6 minutes was established. Two alternative extraction procedures were tested, with tert-butylmethylether or diethyl ether/ethyl acetate and both seemed to work satisfactory. Analysis of serum proved to work well and no severe interference occurred. Results from the linearity tests indicate that future quantification method in serum will be possible.
5 Abbreviations MeOH Methanol ACN Acetonitrile LC Liquid chromatography MS Mass spectrometry
TIC Total ion chromatogram
AAS Anabolic androgenic steroids MRM Multiple reaction monitoring CID Collision induced dissociation
C18 Octadecyl
Rt Retention time
ESI Electrospray ionization
WADA World Anti-Doping Agency
T/E Testosterone glucuronide/Epitestosterone glucuronide
TA Testosterone acetate TB Testosterone benzoate TC Testosterone cypionate TD Testosterone decanoate TE Testosterone enanthate TP Testosterone propionate TPh Testosterone phenylpropionate TI Testosterone isocaproate TU Testosterone undecanoate
6
Table of contents
Abstract
Abbreviations
1. Introduction 8
1.1. Anabolic Androgenic Steroids 8
1.2. Testosterone and Testosterone Esters 8
1.3. Methods of Analysis 9
2. Methodology 10
2.1. Liquid Chromatography 10
2.2. Tandem Quadrupole Mass Spectrometry 11
2.3. Multiple Reaction Monitoring 11
2.4. Aim 12
3. Experimental 13
3.1. Chemicals and Reagents 13
3.2. Solutions 13
3.3. Instrumentation 13
3.4. Sample Preparation 14
3.4.1. Infusion Study 14
3.4.2. Mixed Standards 14
3.4.3. Sample Preparation by LLE 14
3.5. Optimization
3.5.1. MS-Specificity
3.5.2. Chromatographic Selectivity 3.5.3. Linearity and Sensitivity
14
14 15 15
4. Results and Discussion 16
4.1. Tandem-MS Detection 16
4.2. Gradient Profile 17
4.3. Mobile Phase Composition 20
4.4. Stationary Phase 22
4.5. Final Gradient Profile 24
4.6. Matrix Test 26 4.7. Linearity 30 4.8. Final Method 32 5. Conclusion 34 6. Acknowledgement 35 References 36
7
Appendix A. Names and Structures of Testosterone Compounds 37
Appendix B. Transitions in the MS-method Tested with Reference Solutions 38
Appendix C. Transitions Based on the Serum Analysis 42
Appendix D. Linearity Study 45
D.1. Chromatograms from the Linearity Study D.2. Linearity Evaluated by Using Transition 97
45 47
Appendix E. Transitions Used in the Final Study 49
E.1. Final Transition Method E.2. Tests of Final Transitions
49 50
8
1. Introduction
1.1. Anabolic Androgenic Steroids
Anabolic androgenic steroids (AAS) are testosterone and its derivates. All AAS have both anabolic properties such as increased muscle hypertrophy and androgenic such as
masculinisation [2].
Strength training is widely used to increase performance in sports with high physical demands. The use of drugs to further enhance the performance happens via forbidden substances, methods and manipulations. AAS are wide spread among athletes and the youth today in gyms. The effects of these drugs on physical performance are documented [1]. AAS increase the muscle hypertrophy induced by strength training further for athletes involved in doping. The number of nuclei per muscle fibre increases [2]. Those who have withdrawn from anabolic steroid usage and training for several years still have a remaining high number of myonuclei [1].
AAS are used in medical purpose as substitution treatment for men with no natural production of testosterone. Testosterone is predominantly administrated as intramuscular injection but is also available as gel and plaster. In forensic investigation AAS are involved in violent
behaviour, depression and criminality and could cause more serious harm such as sudden cardiac death, damaged liver function and disturbances in the lipid metabolism [1, 2]. In Sweden it is restricted by law the use of synthetic AAS, testosterone and its derivates, growth hormone and chemical substances increasing production and secretion of testosterone and its derivates or growth hormone [3].
1.2. Testosterone and Testosterone Esters
Testosterone is produced in the Leydig cells in the testicles and even in females by the ovaries in small quantities. Testosterone is naturally secreted to urine [4]. Men produce 6-10 mg testosterone daily of which approximately 1% is excreted in urine [1]. Because of the short half-life of only one hour exogenous intake of pure testosterone do not have any effect, since only 2% of oral intake reaches the muscles. To slow down the metabolism and receive better effect the testosterone molecule has been modified at its 17-position. This modification creates stronger anabolic effect and weaker androgen effect as well [5].
O
C H3
CH3 O
R
Figure 1. Structure of Testosterone with an R-group at its 17h position.
9
Testosterone esters have varying half-life. Testosterone cypionate and testosterone enanthate injected intra muscularly have a half-life for 4.5 days, testosterone propionate 0.8 days and testosterone undecanoate 21-34 days. The esters are preferably administrated by intra muscular injection to directly reach the target cells and delay the liver metabolism [6].
1.3. Methods of Analysis
For forensic laboratories as for doping laboratories it is important to be able to determine intake of AAS. To differentiate between exogenous and endogenous testosterone the
measured urinary ratio testosterone glucuronide/ epitestosterone glucuronide (T/E) indicates the use of exogenous intake of testosterone. The naturally occurring epitestosterone is constant since it is not affected by intake of testosterone. A T/E ratio above 6 indicates testosterone doping and levels above 4.0 being considered suspicious according to World Anti-Doping Agency (WADA) and the corresponding ratio used by Department of Forensic Toxicology is 12 [4]. The chosen value is due to statistical reasons and studies indicate that the urinary T/E ratios vary between individuals influenced by genetic factors [2, 4].
A study of samples from Swedish and Korean people predicted different effects of
testosterone intake on the T/E ration in the two ethnic groups. The Swedish people had 16-times higher excretion of testosterone than the Koreans. Recent findings indicate that the gene UGT2B17 influences the testosterone pattern. All individuals homozygous for the UGT2B17 gene have negligible or no excretion of TE. The genotype is seven times more common in Asians than in European people [2]. It is a common polymorphism with an allele frequency of 29% in Swedes and 78% in Koreans. The sensitivity and specificity of the T/E test could be markedly improved by using genotype-based cut off ratios [4]. Studies have showed that even after a single dose of 360 mg testosterone, 40% of the subjects homozygous for the UGT2B17 deletion never reaches the T/E cut off ratio of 4.0. East Asians such as Japanese, Chinese and Koreans, have considerably lower T/E ratios than Europeans increasing the risk of false negative test results, challenging the accuracy of the test [4].
Determination of intact testosterone esters in serum is an unmistakeable proof of exogenous intake and would avoid the problems related with the varying T/E ratios between individuals in urine analysis [7].
10
2. Methodology
2.1. Liquid Chromatography
In order to avoid the time-consuming sample processing and derivatization needed for gas chromatography often used as a method for steroid analysis a simple and rapid liquid chromatography-tandem mass spectrometry (LC-MS/MS) method with electrospray
ionization (ESI) has been tested. The method has proven to be an alternative choice in steroid analysis [8, 9].
High-performance liquid chromatography (HPLC) uses high pressure to force solvents, mobile phase, through columns containing very fine particles, stationary phase, that give high resolution separations. The HPLC system consists of a solvent delivery system, a sample injection valve, a high pressure column, a detector, and a computer to display results and control the system [10].
The compounds in a sample are separated due to their difference in size and affinity for the stationary phase [10]. The chromatography is most commonly reversed-phase in which the stationary phase is less polar than the mobile phase. Typically mobile phases for reverse-phase chromatography are based on acetonitrile or methanol in combination with aqueous buffer. C18 is the most non-polar and common column and other non-polar alternatives are C8 and phenyl column. Drugs of interest are mostly less polar and are therefore better retained by the reversed phase [11]. The chromatogram provides both qualitative and quantitative information. Each compound in the mixture has its own elution time under a given set of conditions. Both the area and the height of each peak are proportional to the amount of the corresponding substance [11].
The growing demand for high-throughput separations in many fields, including forensics, clinical chemistry and doping require faster separations. The need for enhanced productivity and a large number of analyses require mandatory rapid analytical procedures. Ultra
performance liquid chromatography (UPLC) provides faster analyses with the same resolution as HPLC (figure 2) due to the decrease in particle size and column length and increase in pressure making the mobile phase flow rate faster. The theoretical plate is higher for UHPLC than HPLC (fig. 2) [12].
11
Figure 2. Changes in efficiency due to linear velocity depicted as a van Deemter plot. Columns packed with 1.7 μm diameter particles perform better independent of flow rate. (waters.com)
2.2. Tandem Quadrupole Mass Spectrometry
Mass spectrometry embraces detection of analytes by ionizing molecules and then sorting the fragments according to their mass-to-charge (m/z) ratios. Electrospray ionization (ESI) is a very mild ionization technique especially developed for LC-MS which provide little or no fragmentation, making it keep the precursor ion intact. ESI produces charged ions directly from an aqueous/organic solvent system by creating a stream of charged droplets in the presence of a strong electric field. In the quadrupole (Q1) for tandem-MS the ions transferred into the vacuum travel through their respective regions, but only the ions with the selected m/z are detected. Tandem mass spectrometry have two quadrupoles and a collision cell in between which allows selection of a specific m/z in Q1 to further fragmentation in the collision cell (Q2) for selection of another specific m/z in the second quadrupole (Q3) is called multiple reaction monitoring, MRM [13].
Figure 3. Multiple Reaction Monitoring
2.3. Multiple Reaction Monitoring
MRM provides a function for selecting specific m/z and ignores all other fragments. In this mode selected transitions between the precursor ion and a single fragment are monitored. The selected precursor ions are selected in Q1 of tandem mass spectrometers, fragmented in Q2 by
12
collision with an inert gas and the fragments are analysed in the Q3. Unlike the scan function were each m/z is scanned shortly, MRM provides prolonged scan time for each transition. Sensitivity is therefore increased and the signal to noise ratio increased, whereas the spectra specific for the selected precursor contains less chemical noise or interferences.
Fragmentation energy in the collision cell and cone voltage is optimized to obtain reproducible spectra for a large group of compounds [14].
2.4. Aim
To prove intake of testosterone esters in a suspected user, it is desirable to be able to identify the intact esters. The testosterone esters chosen in this study are due to their presence on the market and the most frequent seizure by the police. The primary aim of this thesis was to find a method for liquid chromatography tandem mass spectrometry to determine testosterone esters in serum. The parameters were optimized according to selectivity, sensitivity and resolution and included selection of transition, column, gradient and mobile phase. The secondary aim was to study an extraction procedure possible for quantification of testosterone esters in serum.
13
3. Experimental
3.1. Chemicals and Reagents
Methanol (MeOH), acetonitrile (ACN), acetic acid, formic acid, ammonium acetate and ammonia were purchased from Merck/VWR International.
The reference substances testosterone acetate (TA), testosterone benzoate (TB), testosterone cypionate (TC), testosterone decanoate (TD), testosterone undecanoate (TU), testosterone enanthate (TE), testosterone propionate (TP), testosterone phenylpropionate (TPh) and testosterone isocaproate (TI) were purchased from Steraloids Inc.
Ammonium formiate was purchased from Fluka/Sigma.
Tert-butylmethylether and diethyl ether and ethyl acetate were purchased from Merck.
Milli-Q H2O was produced in house.
Negative serum was obtained from the blood center at the University Hospital in Linköping.
3.2. Solutions
All standard solutions were dissolved in ACN at 200 µg/mL and 10 µg/mL.
Mixed standard solutions containing TA, TB, TC, TD, TE, TP, TPh, TI and TU was prepared in ACN at 1 µg/mL and 0.5 µg/mL, 0.1 µg/mL, 0.05 µg/mL, 0.01 µg/mL, 0.005 µg/mL and 0.001 µg/mL. Mobile phases A ammonium formiate buffer, 5 mM pH 3, were prepared from 1 M stock solutions of formic acid and ammonium formiate, and ammonium acetic buffer, 5 mM pH 5, pH 7.8 were prepared from 1 M stock solutions of Acetic acid and ammonium acetate. Mobile phase B was ACN with 0.05% formic acid and MeOH with 0.05% formic acid.
3.3. Instrumentation
An electrospray liquid chromatography tandem-mass spectrometry system (ESI-LC-MS-MS) for gradient chromatography was used. The instrumentation consisted of an Acquity, Ultra Performance Liquid Chromatographic system (UPLC), equipped with a solvent manager, a sample manager and a column manager for handling of four columns (Waters, Milford, MA). Mass detection was performed on a Quattro Premier XE tandem-MS (Waters, Milford, MA) operating in positive ion mode. The following instrument conditions were used; capillary voltage, 0.9 kV, extractor voltage 3V, RF lens voltage 0.1 V, multiplier voltage 680 V, source temperature 120 C, oven temperature 60C, desolvation gas temperature 400C, cone gas flow 50 L/hr, desolvation gas flow 1100 L/hr, collision gas flow 0.54 L/hr, ion energy (1) 0.5 V, ion energy (2) 0.2 V at LM and HM resolution (1) of 13 and (2) of 15, collision entrance and exit potential of 1 V. Cone voltage and collision energy were optimized individually for each testosterone ester, appendix 5. Instrument control was performed using MassLynx 4.1 and integration, processing and calculation were performed using the TargetLynx software. Infusion experiments for multiple reaction monitoring (MRM) optimizations and ion
suppression studies were performed with an integrated Hamilton syringe pump at a flow rate of 10l/min.
14
High-resolution liquid chromatographic separation was performed on an AQUITY BEH C18 and an AQUITY BEH phenylUPLC-column both with the dimensions 502.1 mm i.d. with 1.7 µm particles (Waters, Milford, MA). An injection volume of 2 µL was used followed of a strong wash of 500 µL ACN, isopropanol, MeOH and water 25:25:25:25 (v/v/v/v) with 0.2% HFo and a weak wash of 900 µL MeOH: 0.1 M HFo 50:50 (v/v). The mobile phases consisted of 5 mM ammonium formiate buffer pH 3.0 or ammonium acetate buffer pH 5.0 or 7.8 for phase A and MeOH with 0.05% HFo or ACN with 0.05% HFo for phase B. Different gradient profiles were tested. A flow-rate of 0.6 ml/min at 60°C was used. Reference chromatograms and the retention times are shown in Results section and the Appendices.A MRM method was prepared including the three or four most intense transitions for each analyte (appendix 5).
3.4. Sample Preparation
3.4.1. Infusion Study
To find suitable transitions for the MS-method, solutions of each ester had to be infused and the distinctly appearing signals examined. The solution infused was made of 50 µL standard solution (10 µg/mL), 450 µL MeOH and 500 µL 20 mM ammonium formiate buffer pH 3. The sample was infused into the LC-MS/MS at a flow rate of 10 mL/min and instrument parameters were optimized individually.
3.4.2. Mixed Standards
Reference mixtures of equal amounts of testosterone esters in each sample were injected to the LC at each chromatography. 100 µL of 10 µg/mL of each ester was mixed to a final concentration of 1 µg/mL in ACN. A 2 µL aliquot was injected into the LC-MS/MS for studies of different chromatographic conditions.
3.4.3. Sample Preparation by LLE
To 500 µL serum 25 µL ester mixture in ACN (10 µg/mL) was added. Serum was extracted for 5 min with 3 mL of tert-butylmethylether or 3 mL diethyl ether/ethyl acetate (70/30) [7]. The phases were separated by centrifugation at 4000 rpm for 10 min and the upper organic phase was transferred to a clean 10-mL conical glass tube and solvents were evaporated to dryness under nitrogen at room temperature. Residues were reconstituted in 200 µL ACN. A 2 µL aliquot was injected into the LC-MS/MS and interferences from the matrix were studied.
3.5. Optimization
3.5.1. MS-Specificity
Suitable testosterone ester transitions were selected for MRM determination by performing infusion experiments. Molecular ion weight for the precursor was easily found knowing the theoretical value for the molecular weight while using the scan mode. Finding product ions
15
required several infusion experiments at various collision energies at daughter ion scan mode for detection of specific and common fragments for the testosterone esters. A preset of selected transitions was tried out using reference substances. Fragments with the highest intensities for each testosterone ester were chosen and set in the MRM-method.
3.5.2. Chromatographic Selectivity
Based on that testosterone esters are non-polar a phenyl column and C18 column were tested. Phenyl columns are not as non-polar as the C18 column and were assumed to give the best separation, due to testosterone esters long fatty tails that was suspected to get very retained in the column and difficult to elute.
3.5.3. Linearity and sensitivity
By future interest in possible quantification a study for linearity and sensitivity was made based on serum analysis. Linearity for concentration of each testosterone ester is based on transition 97.0 or 97.1 at 1.0 µg/mL, 0.5 µg/mL, 0.1 µg/mL, 0.05 µg/mL, 0.01 µg/mL, 0.005 µg/mL and 0.001 µg/mL.
Mobile phase and pH affect the sensitivity for detection at the interface. Mobile phase B were ACN and MeOH and A were ammonium formiate pH 3 and ammonium acetate pH 5 and pH 7.8. Combined in different combinations at different gradient profiles of mobile phase A and B the highest sensitivity at the detection was to be found.
16
4. Results and Discussion
The aim of this study was to find a method for identifying the nine most frequently appearing testosterone esters at the market in serum and test the quantifying range and linearity with LC-MS/MS. Focus was on selection of the most suitable column, mobile phase, gradient and MRM-transitions. The optimisation was made with reference substances and after a suitable method was selected it was tested on spiked serum samples. Transitions found by infusion tests were set up in a method by which following reference substances and serum sample tests were analysed and evaluated. Extraction of serum samples was based on two different
methods.
4.1. Tandem-MS Detection
In LC-MS fragmentation is generated by collision induced dissociation (CID). All
testosterone esters are based on the testosterone molecule and have therefore CID-fragments in common and also fragments specific for each testosterone ester based on the ester structure. Regulations for LC-MS analysis recommend at least two transitions for identifications of drugs. Three or four fragments were chosen for each testosterone ester in this study. Common product ions from the testosterone structure are 97 and 109 (table 1). Specific product ions could not be found for all of the esters. As can be seen in table 1, fragment 1 and 2 are in common for all the testosterone esters and are therefore assumed to origin from the testosterone structure. By evaluations and comparison of mass spectra it was found that several other fragments were in common as well (eg. 149.1, 163.2 and 175.2). Although unique fragments not were available for most testosterone esters, unique transitions could be selected since all testosterone esters had unique precursor ions [M+1].
Table 1. Fragment ions and their intensities.
Analyte MW [M+1] Fragment 1 Fragment 2 Fragment 3 Fragment 4
TA 330.5 331.2 97.0 1.61e6 109.0 2.25e6 135.1 7.40e4
TB 392.5 393.2 97.1 4.83e5 109.1 1.11e5 105.0 8.94e5
TC 412.6 413.3 97.1 5.51e5 na 125.1 1.89e5 163.2 6.20e4
TD 442.7 443.3 97.1 5.45e5 109.0 4.77e5 119.2 7.75e4 123.1 4.26e4
TE 400.6 401.3 97.0 4.28e5 109.0 2.97e5 113.1 3.02e5
TPh 420.6 421.3 97.1 8.43e5 109.0 7.51e5 163.2 1.24e5 173.1 4.52e4
TP 344.5 345.1 na 109.0 1.83e6 123.1 5.84e4 187.2 4.49e4
TI 386.6 387.5 97.1 2.24e5 109.1 1.29e5 149.1 1.03e4 175.2 2.36e4
TU 456.7 457.3 97.1 8.81e5 109.1 6.84e5 169.2 3.70e5
Fragments were selected based on their sensitivity during chromatography and the highest intensities were chosen and set in the MS-method. One fragment from the infusion test
showed no signal in the method (fig. 4) and was replaced. TC, seen in figure 4, shows a signal at 2.01 min which origins from the mobile phase (appendix 2). Final transitions table is found in appendix 5.
17
Figure 4. Transitions tested for TC showing an incorrect transition for 413.3 > 301.2.
4.2. Gradient Profile
The chromatographic gradient profile affects the elution and separation of the testosterone esters. A steep gradient profile may elute the testosterone esters earlier in the chromatography. For a final method the testosterone esters were preferred in a wide range retention times for better separation and determination of the peaks. Short time of analysis within less than 10 minutes for is preferred for routine analysis. Testosterone esters need therefore to elute in the chromatography and before the wash-period begins. Fairly high content organic solvent in the mobile phase was needed to elute the testosterone esters.
Initially different gradient profiles were tested on a phenyl column. The first trials had a total time of 7 minutes. The mobile phase was ACN pH 3 and the flow rate is 0.6 mL/min. In order to avoid band broadening it was important to begin at a low start concentration of mobile phase B. The gradient profiles were set at a start concentration of 10-40% (figure 5. A-D). The higher start concentration of mobile phase B the wider the range in retention time between the testosterone esters was seen and all nine testosterone esters were visible but not baseline separated in all of the gradient profiles tested (fig. 5). Gradient C and D proved to elute the most suitable for fast analysis at a first retention time of approximately 2 minutes.
413.4 > 301.2
413.4 > 163.2 413.4 > 125.1 413.4 > 97.1 TIC
18
Gradient A Gradient B
Gradient C Gradient D
Figure 5. Phenyl column and mobile phases ACN pH 3. Gradient profiles starting at (A)10%, (B) 20%, (C) 35% and (D) 40% of mobile phase B.
7 minutes analyzing time is a fairly long time for routine analysis and a reduction in time to 5.6 minutes was performed. The slope of gradient C with an initial composition of 35% B-solvent was set as a template and the maximum concentration of gradient C was never reached in order to save time and instead enter the wash-period directly after the last testosterone ester had eluted.
Four gradients were tested with MeOH on the phenyl column at pH 3, starting with the result made by gradient C as a template (fig. 6 Gradient E). Gradient E elute the testosterone esters late in the chromatography and the start concentration was increased until a suitable gradient was found based on elution of the first testosterone ester. The higher the start concentration of mobile phase B, the longer the range in retention time between the testosterone esters are, which is sought for. The long range in retention time results in seven base line separated
19
analytes. The change in mobile phase B from ACN to MeOH led to two analytes eluting at the same retention time, TI and TPh. Gradient H was chosen to be the most suitable gradient profile for the phenyl column. To determine how to go further in selection of gradient profile the most suitable column must be selected.
Gradient E Gradient F
Gradient G Gradient H
Figure 6. Change in gradient profile to centre the analytes for a decrease in chromatography to 5.6 minutes. Gradient (E) starting at mobile phase B of 45% MeOH, (F) starting at mobile phase B of 50% MeOH, (G) starting at mobile phase B of 55% MeOH and (H) starting at mobile phase B of 60% MeOH on a phenyl column with mobile phase A as pH 3.
20
4.3. Mobile Phase Composition
Three different pHs’ were tested on the phenyl column. The phenyl column with pH 3, 5 and 7.8 as mobile phase A and ACN as mobile phase B was tested with a mixture of nine
testosterone esters.
Testosterone esters were separated at pH 3 but only five of them were base line separated, TA, TP, TB, TU and TD (fig. 7). The intensities vary but were fairly equal besides the peaks at 1.89 minutes, TA and 3.68 minutes, TU. The mobile phase was more sensitive to TA and least sensitive to TD (fig. 7, table 2).
Figure 7. Phenyl column with ACN as mobile phase B and mobile phase A of pH 3.
Both pH 5 (fig. 8) and 7.8 (fig. 9) show an increase in intensity for the first peak at 1.89 minutes. The intensities for the other peaks are fairly good. Separation is on the same level as for pH 3 (fig. 9). Chromatograms show that a change in pH does not have any effect on selectivity (fig. 7-9).
Comparing pH 3 for MeOH and pH 3, 5 and 7.8 for ACN on the phenyl column shows the highest sensitivity at pH 5. Sensitivity was the lowest for pH 7.8 and was therefore abandoned for further analysis.
21
Figure 8. Phenyl column with ACN as mobile phase B and mobile phase A of pH 5.
Figure 9. Phenyl column with ACN as mobile phase B and mobile phase A of pH7.8
Table 2. Intensity based on mobile phase A and B on a phenyl column. (* optimum conditions)
Analyte ACN pH 3 ACN pH 5 ACN pH 7.8
TA 5.64e6 7.26e6* 5.92e6
TB 4.58e5 7.98e5* 4.71e5
TC 1.37e6* 1.26e6 7.41e5
TD 4.60e5 1.04e6* 6.96e5
TE 5.36e5 1.20e6* 5.37e5
TPh 1.23e6* 1.13e6 7.45e5
TP 2.84e6* 1.89e6 1.06e6
TI 1.44e6 1.80e6* 1.24e6
22
4.4. Stationary Phase
Selection of stationary phase material affects the selectivity for the testosterone esters. Thus C18 and as well as a phenyl column were tested with a mobile phase consisting of MeOH and ammonium acetate pH 5.
Eight peaks were visible and seven were base-line separated on the phenyl column (fig. 10A). TPh and TI have the same retention time and are hidden in the same peak (fig. 10B). The order of elution is based on the size and polarity of the esters. TI eluted slightly before TPh probably due to the π-interaction created between the phenyl-groups in TPh and the stationary phase.
A B
Figure 10. (A) Phenyl column with MeOH as mobile phase B and mobile phase A pH 5. (B) Transition 97 for all testosterone esters.
A change of column from phenyl to C18 for increased retardation of the testosterone esters were expected and could give a better separation. Nine peaks were shown and seven of them were base line separated using the C18 column (fig. 11). The order of elution has changed, TPh eluated before TI due to the increased affinity for TI and the stationary phase after the change in column. C18 column with MeOH as mobile phase B provides better selectivity for testosterone esters than a phenyl column with MeOH as mobile phase B of the same
dimensions. TI TPh TU TD TPh TC TE TB TI TP TA
23
A B
Figure 11. C18 column with MeOH as mobile phase B and mobile phase A of (A) pH 5 and (B) transition 97.
A change in column to C18 and MeOH as mobile phase B shows nine peaks and almost seven of the testosterone esters are base line separated (fig. 12). Sensitivity is higher for MeOH than ACN (table 3). TA is the most sensitive to the method and TD is least sensitive in both pH 3 and 5, to be compared with MeOH were the lowest sensitivity varied depending on pH, TU at pH 3 and TD at pH 5 (fig. 13). The selectivity is the same for MeOH and ACN, but the sensitivity is much higher for MeOH (table 3). Compared to the phenyl column (table 2) the C18 column shows higher intensities for mobile phases MeOH and pH 3 (table 3). MeOH has the best sensitivity at both pH 3 and pH 5 compared to ACN (table 3). MeOH and pH 5 shows the highest intensity for each one of the testosterone esters.
A B
Figure 12. C18 column with ACN as mobile phase B and mobile phase A of (A) pH 3 and (B) pH 5. TI TPh TU TD TPh TC TE TB TI TP TA
24
A B
Figure 13. C18 column with MeOH as mobile phase A and mobile phase B of (A) pH 3 and (B) pH 5.
Table 3. Intensity for transition 97 based on pH of mobile phase A on a C18 column.(*optimum conditions)
Analyte MeOH pH 3 ACN pH 3 MeOH pH 5* ACN pH 5
TA 3.36e7 8.13e6 4.01e7 6.37e6
TB 6.06e6 1.25e6 7.76e6 1.21e6
TC 1.05e7 1.23e6 1.30e7 1.74e6
TD 3.50e6 1.12e6 7.03e6 1.12e6
TE 7.96e6 1.51e6 9.93e6 1.42e6
TPh 1.22e7 1.32e6 1.43e7 1.23e6
TP 1.66e7 1.79e6 1.91e7 1.99e6
TI 1.24e7 1.46e6 1.49e7 2.27e6
TU 4.36e6 1.55e6 1.07e7 2.76e6
4.5. Final Gradient Profile
After selection of the C18 column as the better choice, the gradient profiles for both MeOH and ACN were tested. 60% of MeOH as start concentration equals the same eluent strength as 45% of ACN [9]. Gradient H (fig. 6) gave room for even earlier elution and the start
concentration for ACN was therefore set at 50%.
The gradient with ACN for 5.60 minutes had the first testosterone ester, TA eluting early in the chromatography and the last two at maximum concentration of mobile phase B, TD and TU (fig. 14). All nine testosterone esters were separated and seven of them are base-line separated.
25
The gradient with MeOH for 5.60 minutes (fig. 15) has its first testosterone ester, TA eluting at approximately the same time as H (fig. 6) but the last two testosterone esters, TD and TU elutes later when mobile phase B reached its maximum percentage.
Compared to the gradient for MeOH, the gradient for ACN provides a larger range in
retention times between the testosterone esters, 3.22 versus 3.01 (table 4). The gradients differ due to the higher percentage needed for ACN at the start level. The two esters eluting after wash-period has started could result in interference with the serum in the serum analysis following.
Figure 14. C18 column with mobile phase A at pH 3 and mobile phase B starting at a concentration of 50 % ACN.
Figure 15. C18 column with mobile phase A at pH 3 and mobile phase B starting at a concentration of 60 % MeOH.
TA TP TB TPh TI TE TC TD TU
26 Table 4. Retention time based on mobile phase A. (*Rt last peak – Rt first peak)
Analyte ACN pH 3 MeOH pH 3
TA 1.73 2.04 TP 2.14 2.47 TB 2.72 3.20 TPh 2.84 3.36 TI 3.19 3.61 TE 3.63 4.04 TC 3.73 4.13 TD 4.64 4.94 TU 4.95 5.05 Range* 3.22 3.01 4.6. Matrix Test
Transitions and parameters were optimized for testosterone esters in reference solution. A change of matrix to serum could show interference with serum and extraction solutions and also suppress the sensitivity for the testosterone esters at the detection. To find out if serum gives any interferences it is important to pre-run a serum analysis with a blank of mobile phase A and blank serum. Two extraction methods are tested, tert-butylmethylether and (70:30) diethyl ether/ethyl acetate and two serum blanks. The serum analysis involves serum spiked with the nine testosterone esters at the same concentration before extraction as the reference solution.
Blank of mobile phase pH 5 showed a system peak at 1.33 and an increase in intensity at 4.06 and 6.14 minutes (fig. 16). The high intensity at 1.33 is a system peak and the high intensity at 6.14 minutes is due to the wash-period where MeOH is at 95%.
27
The blank serum extracted with diethyl ether/ethyl acetate and tert-butylmethylether both shows interference at 4.1 minutes (fig. 17, 18). Transitions selected for TA showed interference at 4.15 minutes (fig. 16, 17).
A B
Figure 17. Blank serum extracted with diethyl ether/ethyl acetate. (A) Chromatogram (B) Transition 97 for all testosterone esters.
A B
Figure 18. (A) Blank serum extracted with tert-butylmethylether. (B) Transition 97 for all testosterone esters.
TU TD TPh TC TE TB TI TP TA TU TD TPh TC TE TB TI TP TA
28
The interference in serum is shown for TA (fig, 17, 18) at transition 97 and 109 (fig. 19) during a test for all transitions for TA. This could be due to carry over or a similar natural testosterone hormone with the same precursor and some transitions as TA.
A B
Figure 19. A) Blank serum extracted with diethyl ether/ethyl acetate and B) tert.butylmethylether.
Both extraction methods are equally working (fig. 20). All testosterone esters were separated and sensitivity for each ester was good enough for determination (appendix 4). The sensitivity is still higher for TA and least sensitive for TD and TU.
A B
Figure 20. Serum spiked with mixture of the nine testosterone esters extracted with (A) diethyl ether/ethyl acetate and (B) tert.butylmethylether run on a C18 column.
29
Since TA showed an increase in intensity at 4.1 (fig.19) which was the retention time for TC an investigation whether it is due to carry over or not, two blanks following one blank matrix from the diethyl ether/ethyl acetate method were run. Four more transitions for TA were set in the MS-method to find out if more transitions show interference.
Neither of the blanks showed any transitions at 4.15 minutes at either the first blank run or the second(fig. 21 A, C) and nor did the transitions (fig. 21B, 21D). Based on this there was no proof for carry over. The third run, the blank matrix showed an increased intensity at 4.15 minutes (fig 22). Transitions 97 and 109 showed peaks for TA but the other transitions except for 253.2, which has an increased intensity for the system peak seemed to be working.
Transitions 97 and 109 for TA needed therefore to be replaced.
A B
C D
30
A B
Figure 22. (A) Blank serum extracted with diethyl ether/ethyl acetate. (B) Transition 97 and 109 for TA. There is not any difference in retention time between the reference mixture and the extraction methods (table 5). Intensities for the extraction methods are fairly equal (fig. 24) and for further analysis both of them can be used.
Table 5. Retention time based on extraction method
Analyte Referens mix
Serum mix diethylether
Serum mix tert-butylmetylether TA 2.07 2.04 2.04 TP 2.51 2.49 2.49 TB 3.26 3.22 3.22 TPh 3.40 3.36 3.36 TI 3.65 3.62 3.62 TE 4.08 4.04 4.04 TC 4.17 4.13 4.13 TD 4.95 4.95 4.94 TU 5.06 5.06 5.05 Range 2.99 3.02 3.01
31
4.7. Linearity
In order to see if quantification analysis was possible it was important that the intensity peak area was linear dependent on the ester concentration. Linearity was tested based on the results from the matrix test (4.6). Each signal for every transition was integrated and based on the total intensity for each ester graphs were created. Concentrations tested ranges from 0.001 µg/mL to 1 µg/mL.
Intensity was linear dependent on the ester concentration. A representative calibration for TI showed a straight line close to the intensity/concentration-ratio without any outliers (fig. 23). The linearity has an r-factor of 0.999 showing that quantification is possible (fig. 23A). The lowest concentration tested, 0.001 µg/mL, here showed for TI in figure 23B shows that quantification is possible even for low sample concentrations. The signals did not show any interference and the peaks are close to symmetric. Linearity for the other testosterone esters is attached in appendix 3 and table 6 shows the raw data.
Table 6. Intensity (peak area) based on testosterone ester concentration for C18 column with mobile phase B MeOH and A pH 5. Analyte 1 µg/mL 0.5 µg/mL 0.1 µg/mL 0.05 µg/mL 0.01 µg/mL 0.005 µg/mL 0.001 µg/mL
TA 4.01e7 2.27e7 4.55e6 2.50e6 4.00e5 1.88e5 4.30e4
TB 7.76e6 4.41e6 8.93e5 4.66e5 7.98e4 3.21e4 7.76e3
TC 1.30e7 6.96e6 1.43e6 7.86e5 1.28e5 6.86e4 2.30e4
TD 7.03e6 3.70e6 7.51e5 4.34e5 7.84e4 2.21e4 8.98e3
TE 9.93e6 5.43e6 1.07e6 6.24e5 1.55e5 7.54e4 3.19e4
TPh 1.43e7 8.06e6 1.57e6 8.59e5 1.49e5 6.31e4 1.64e4
TP 1.91e7 1.03e7 1.96e6 1.09e6 1.97e5 8.44e4 1.66e4
TI 1.49e7 8.02e6 1.61e6 8.56e5 1.67e5 6.04e4 1.24e4
TU 1.07e7 6.06e6 9.07e5 4.87e5 1.18e5 3.07e4 1.54e4
A B
Figure 23. Test of linearity for TI from concentrations of 0.001 µg/mL to 1 µg/mL. (A) Linearity based on transition 97 and (B) integration of peak areas for all transitions at 0.001 µg/mL.
32
The linearity study makes it possible to estimate the limit of detection from the lowest concentration at 0.001 µg/mL (table 7). Calculations backward based on varying amount of serum and reconstitution solution (3.4.3.) shows that the factor divided by the lowest
concentration is 2.5 whichever concentration for the testosterone esters is used. The factor affects how low the concentration in the serum could be. Varying amount of serum and reconstitution solution changes the factor and an increase in the factor makes it possible to detect even lower quantities.
Table 7. Estimation of detection limits at changed parameters (3.4.3.).
Parameters Initial conc. and amounts Alternative A Alternative B Alternative C testosterone esters 0.001 µg/mL 0.001 µg/mL 0.001 µg/mL 0.001 µg/mL serum 0.5 mL 1 mL 1 mL reconstitution solution (ACN) 0.2 mL 0.1 mL 0.1 mL Calc. factor 2.5 5 5 10 100% recovery 0.4 ng/mL 0.02 ng/mL 0.02 ng/mL 0.01 ng/mL 80% recovery 0.5 ng/mL 0.025 ng/mL 0.025 ng/mL 0.0125 ng/mL 4.7. Final method
This study has ended up in a suggestion for a method setup based on the best results of selection of gradient, column, mobile phases, method and serum preparations. The MS-method was set after the serum runs in favour to receive a cleaner chromatogram. Unique transitions are found for each ester. The peaks were separated and seven of nine esters are base-line separated using reference mixture on a C18 column with MeOH and ammonium acetate 5 mM pH 5 as mobile phases (fig. 24). The chromatography for the spiked serum (fig. 25) provides all nine testosterone esters visible and separated. The selectivity was not as good as for the reference mixture (fig. 24), due to the decreased concentration from the extraction and the spiked serum is not as pure as the references showing a less clean chromatogram. TA was no longer the testosterone ester with the strongest sensitivity for the method after the change in transitions. TU still proves to be the one least sensitive to MeOH. The sensitivity for the testosterone esters depends on the amount of transitions and how large the signal for each one of them was. Signal intensity was high enough for quantification in serum based on the intensity for the lowest concentration tested (4.7.).
33
Figure 24. Reference mixture of nine testosterone esters at a concentration of 1µg/mL run on a C18 column with mobile phase A at pH 5 and MeOH for mobile phase B.
Figure 25. Serum spiked with nine testosterone esters at a concentration of 1 µg/mL extracted with diethyl ether/ethyl acetate run on a C18 column with mobile phase A at pH 5 and MeOH for mobile phase B.
Figure 26. Blank serum extracted with diethyl ether/ethyl acetate run on a C18 column with mobile phase A at pH 5 and MeOH for mobile phase B.
TA TP TB TPh TI TE TC TD TU TA TP TB TPh TI TE TC TD TU
34
5.
Conclusion
A method that could be used for determination of testosterone esters in serum has been found. A suggestion for a method set up of testosterone esters has been made in reference solutions as a primer for quantification in serum.
The MS-method was set and at least three possible transitions for each testosterone ester were found. The best choice of column proved to be the C18 column where all the esters were separated and seven of them were base-line separated. The C18 column along with methanol and ammonium acetate buffer, 5 mM, pH 5 showed the highest sensitivity for Multiple Reaction Monitoring-detection. A gradient profile for a total runtime of 5.6 minutes was established. Two alternative extraction procedures were tested, with tert-butylmethylether or diethyl ether/ethyl acetate and both seemed to work satisfactory. Analysis of serum proved to work well and no severe interferences occurred. Results from the linearity tests indicate that future quantification method in serum will be possible.
The suggested method has been proven to work well. Further development should be focused on validation such as determinations of; limit of detection, precision, calibration and
35
6. Acknowledgement
I am grateful for having had the opportunity to write this report. It was of big interest and I have been exited from the very beginning.
I would like to thank my supervisors Yvonne Lood and Martin Josefsson for all the support and help. I also would like to thank all the people at the National Board of Forensic Medicine that I have been in contact with during this final thesis.
36
References
[1] A. Eriksson, Strength training and anabolic steroids – A comparative study of the vastus lateralis, a thigh muscle and the trapezius, a shoulder muscle, of strength-trained athletes, PhD thesis, Umeå University, 2006
[2] F. Sjöqvist, M. Garle, A. Rane, Use of doping agents, particularly anabolic steroids, in sports and society, Lancet 371 (2008) 1872-1879
[3] Svensk författningssamling, Lag (1991:1969) om förbud mot visa dopingsmedel, hämtat 2010-05-20, www.riksdagen.se
[4] J. Jakobsson Schultze, Genetics of androgen disposition – Implications for doping tests, PhD thesis, Karolinska Institutet, 2007
[5] T. Rosén, Hormondoping in, Endokrinologi, Ed. S. Werner, Liber (2006) p262 [6] H.M. Behre, E. Nieschlag, Testosterone buciclate (20 Aet-1) in hypogonadal men: pharmacokinetics and pharmacodynamics of the new long-acting androgen ester, J Clin Endocrinol Metab. 75 (1992) 1204-1210
[7] X. De la Torre, J. Segura, A. Polettini, Detection of testosterone esters in human plasma by GC/MS and GC/MS/MS in, Recent advances in doping analysis, Ed. M. Donike, H. Geyer, U. Mareck-Engelke, Sport und Buch Strauß (1995) 59-80
[8] U. Turpeinen, S. Linko, O. Itkonen and E. Hämäläinen, Determination of testosterone in serum by liquid chromatography-tandem mass spectrometry, Scand J Clin Lab. 68 (2008) 50-57
[9] S-H. Peng, J. Segura, M. Farré, J.C. Gonzáles. X. De la Torre, Plasma and urinary markers of oral testosterone undecanoate misuse, Steroids 67 (2002) 39-40
[10] D.C. Harris, Quantitative chemical analysis, sixth edition, W.H. freeman and Company (2003) chapter 23, 25
[11] V.R. Meyer, Practical High-performance Liquid Chromatography, fourth edition, John Wiley and sons (2004) chapter 1, 10
[12] D. Guillarme. J. Ruta, S. Rudaz and J-L. Veuthey, New trends in fast and high-resolution liquid chromatography: a critical comparison of existing approaches, Anal Bioanal Chem. 3 (2009) 1069-1082
[13] P.J. Taylor, Method development and optimization of LC-MS in, Applications of LC-MS in toxicology, Ed. A. Polettini, Pharmaceutical Press (2006) p23
[14] P. Marquet, Systematic toxicological analysis with LC-MS in, Applications of LC-MS in toxicology, Ed. A. Polettini, Pharmaceutical Press (2006) p111
37
Appendix A. Names and Structures of Testosterone Compounds
Table I. Names and structures of compounds sorted by retention time from the final results (*not tested) Transitions chosen are presented in the m/z-column, were the precursor is marked with M.
Structure Analyte Rt (min) MW (g/mol) m/z
O C H3 CH3 OH Testosterone (4-Androsten-17β-ol-3-one) *na 288.414 289.3M [8] CH3 CH3O O CH3 O Testosterone acetate (4-Androsten-17β-ol-3-one Acetate) 2.04 330.46 135.1 253.2 289.3 331.2M CH3 CH3 O O O C H3 Testosterone propionate (4-Androsten-17β-ol-3-one Propionate) 2.49 344.49 97.1 109.0 123.1 187.2 345.1M O C H3 CH3O O Testosterone benzoate (4-Androsten-17β-ol-3-one Benzoate) 3.22 392.53 97.1 105.0 109.1 393.2M CH3 CH3 O O O Testosterone phenylpropionate (4-Androsten-17β-ol-3-one Phenylpropionate) 3.36 420.591 97.1 109.0 163.2 173.1 421.3M CH3 CH3 O O O C H3 C H3 Testosterone isocaproate (4-Androsten-17β-ol-3-one Isocaproate) 3.62 386.58 97.1 109.1 175.2 271.2 387.2M CH3 CH3 O O O C H3 Testosterone enanthate (4-Androsten-17β-ol-3-one Enanthate) 4.04 400.59 97.0 109.0 113.1 401.3M CH3 CH3 O O O Testosterone cypionate (4-Androsten-17β-ol-3-one Cypionate) 4.13 412.6 97.1 109.1 125.1 163.2 413.3M CH3 CH3 O O O C H3 Testosterone decanoate (4-Androsten-17β-ol-3-one Decanoate) 4.96 442.67 97.1 109.0 119.2 443.3M CH3 CH3 O O O CH3 Testosterone undecanoate (4-Androsten-17β-ol-3-one Undecanoate) 5.06 456.7 97.1 109.1 169.2 457.3M
38
Appendix B. Transitions in the MS-method Tested with Reference Solutions
Figure I. Blank of ammonium formiate pH 3 showing a peak at 2.01 minutes.
Figure II. Transitions for TA tested on a phenyl column of mobile phase A pH 3 and mobile phase B ACN.
Figure III. Transitions for TB tested on a phenyl column of mobile phase A pH 3 and mobile phase B ACN. 331.2 > 135.1 331.2 > 109 331.2 > 97 TIC 393.2 > 109 393.2 > 105 393.2 > 97.1 TIC
39
Figure IV. Transitions for TC tested on a phenyl column of mobile phase A pH 3 and mobile phase B ACN.
Figure V. Transitions for TD tested on a phenyl column of mobile phase A pH 3 and mobile phase B ACN.
Figure VI. Transitions for TE tested on a phenyl column of mobile phase A pH 3 and mobile phase B ACN. 413.3 > 301.2 413.3 > 163.2 413.3 > 125.1 413.3 > 97.1 TIC 443.3 > 97.1 443.3 > 119.2 443.3 > 109 443.3 > 123.1 TIC 401.3 > 113.1 401.3 > 109 401.3 > 97 TIC
40
Figure VII. Transitions for TP tested on a phenyl column of mobile phase A pH 3 and mobile phase B ACN.
Figure VIII. Transitions for TPh tested on a phenyl column of mobile phase A pH 3 and mobile phase B ACN.
Figure IX. Transitions for TI tested on a phenyl column of mobile phase A pH 3 and mobile phase B ACN. 345.1 > 187.2 345.1 > 123.1 345.1 > 109 TIC 421.3 > 173.1 421.3 > 163.2 421.3 > 109 421.3 > 97.1 TIC 387.2 > 175.2 387.2 > 149.1 387.2 > 109.1 387.2 > 97.1 TIC
41
Figure X. Transitions for TU tested on a phenyl column of mobile phase A pH 3 and mobile phase B ACN. 457.3 > 169.2
457.3 > 109.1 457.3 > 97.1 TIC
42
Appendix C. Transitions Based on the Serum Analysis
Transitions based on the analysis for blank serum and spiked serum.
A
B C
D E
Figure XI. A) Blank serum extracted with diethyl ether/ethyl acetate. B-E) Transitions chosen for testosterone esters. The testosterone esters do not show any interferences except for TA which shows an increase in signal at 4.15 for transition 97 and 109.
43 A
B C
D E
Figure XII. A) Blank serum extracted with tert-butylmethylether. B-E) Transitions chosen for testosterone esters. The testosterone esters do not show any interferences except for TA which shows an increase in signal at 4.14 for transition 97 and 109.
44 A
B C
D E
Figure XIII. (A) Serum spiked with testosterone esters of 1 µg/mL extracted with diethyl ether/ethyl acetate. (B-E) Transitions chosen for testosterone esters.
TU TD TPh TE TB TI TPh TC TE TP TA TA TP TB TPh TI TE TC TD TU
45
Appendix D. Linearity Study
D.1. Chromatograms from the Linearity Study
A B
C D
Figure XIV. (A) Blank of mobile phase A pH 5. (C) Reference mixture for the testosterone esters of 1 µg/mL and transition 97 for the testosterone esters in the (B) blank and (D) the reference mixture.
TU TD TPh TC TE TB TI TP TA TU TD TPh TC TE TB TI TP TA
46
A B
C D
Figure XV. Reference mixture for the testosterone esters of (A) 0.005 µg/mL and (C) 0.001 µg/mL. Transition for 97 for the testosterone esters at the concentration of (B) 0.005 µg/mL and (D) 0.001 µg/mL.
TU TD TPh TC TE TB TI TP TA TU TD TPh TC TE TB TI TP TA
47
D.2. Linearity Evaluated by Using Transition 97
A B
C C
Figure XVI. Test of linearity for the reference mixtures for concentrations of 0.001 µg/mL to 1 µg/mL. Linearity based on transition 97 for (A) TA, (B) TB, (C) TP and (D) TU.
48
A B
C D
Figure XVII. Test of linearity for the reference mixtures from concentrations of 0.001 µg/mL to 1 µg/mL. Linearity based on transition 97 for A) TC, B) TD, C) TE and (D) TPh.
49
Appendix E. Transitions Used in the Final Study
E.1. Final Transition MethodTable II. Final transition method.
Precursor Product ion Cone (V) Coll (eV) Rt Compound
331.20 135.10 30.00 25.00 2.04 Testosterone acetate 331.20 253.20 30.00 15.00 2.04 Testosterone acetate 331.20 289.30 30.00 15.00 2.04 Testosterone acetate 345.10 97.10 35.00 25.00 2.49 Testosterone propionate 345.10 109.00 35.00 25.00 2.49 Testosterone propionate 345.10 123.10 35.00 25.00 2.49 Testosterone propionate 345.10 187.20 35.00 15.00 2.49 Testosterone propionate 387.20 97.10 35.00 25.00 3.62 Testosterone isocaproate 387.20 109.10 35.00 30.00 3.62 Testosterone isocaproate 387.20 175.20 35.00 25.00 3.62 Testosterone isocaproate 387.20 271.20 35.00 15.00 3.62 Testosterone isocaproate 393.20 97.10 35.00 25.00 3.22 Testosterone benzoate 393.20 105.00 35.00 25.00 3.22 Testosterone benzoate 393.20 109.10 35.00 25.00 3.22 Testosterone benzoate 401.30 97.00 35.00 20.00 4.04 Testosterone enanthate 401.30 109.00 35.00 25.00 4.04 Testosterone enanthate 401.30 113.10 35.00 20.00 4.04 Testosterone enanthate 413.30 97.10 40.00 25.00 4.13 Testosterone cypionate 413.30 109.10 40.00 30.00 4.13 Testosterone cypionate 413.30 125.10 40.00 20.00 4.13 Testosterone cypionate 413.30 163.20 40.00 20.00 4.13 Testosterone cypionate 421.30 97.10 30.00 25.00 3.36 Testosterone phenylpropionate 421.30 109.00 30.00 30.00 3.36 Testosterone phenylpropionate 421.30 163.20 30.00 20.00 3.36 Testosterone phenylpropionate 421.30 173.10 30.00 20.00 3.36 Testosterone phenylpropionate 443.30 97.10 25.00 25.00 4.96 Testosterone decanoate 443.30 109.10 25.00 30.00 4.96 Testosterone decanoate 443.30 169.20 25.00 35.00 4.96 Testosterone decanoate 457.30 97.10 35.00 25.00 5.06 Testosterone undecanoate 457.30 109.10 35.00 25.00 5.06 Testosterone undecanoate 457.30 169.20 35.00 20.00 5.06 Testosterone undecanoate
50
E.2. Tests of Final Transitions
Tests of transitions used in the final study based on the reference mix, spiked serum sample and blank serum.
A B
C D
51
A B
C D
52
A B
C D
