JOURNAL OF MASS SPECTROMETRY J. Mass Spectrom. 2008; 43: 865–876 Published online 2 June 2008 in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/jms.1438 Review Mass spectrometry of selective receptor modulators

Mario Thevis∗ and Wilhelm Schanzer¨

Institute of Biochemistry, Center for Preventive Doping Research, German Sport University Cologne, Carl-Diem Weg 6, 50933 Cologne, Germany

Received 28 March 2008; Accepted 1 May 2008

Nonsteroidal selective modulators (SARMs) are an emerging class of drugs for treatment of various diseases including and muscle wasting as well as the correction of age-related functional decline such as muscle strength and power. Several SARMs, which have advanced to preclinical and clinical trials, are composed of diverse chemical structures including arylpropionamide-, bicyclic hydantoin-, quinoline-, and tetrahydroquinoline-derived nuclei. Since January 2008, SARMs have been categorized as anabolic agents and prohibited by the World Anti-Doping Agency (WADA). Suitable detection methods for these low-molecular weight drugs were based on mass spectrometric approaches, which necessitated the elucidation of dissociation pathways in order to characterize and identify the target analytes in doping control samples as well as potential metabolic products and synthetic analogs. Fragmentation patterns of representatives of each category of SARMs after electrospray ionization (ESI) and collision-induced dissociation (CID) as well as electron ionization (EI) are summarized. The complexity and structural heterogeneity of these drugs is a daunting challenge for detection methods. Copyright  2008 John Wiley & Sons, Ltd.

KEYWORDS: sport; doping; mass spectrometry; SARMs; orbitrap; anabolics

INTRODUCTION flutamide (Fig. 1, 2),9 both of which include an arylpro- pionamide nucleus. The first selective androgen receptor The desire to reverse or slow age-related maladies in men has modulators (SARMs) resulting from these studies were S-1 been stimulus for scientific research for more than 140 years. and S-4 (Fig. 1, 3 and 4, respectively), which were later Reports from 1869, about suggestions to inject semen into the termed Andarine and Ostarine, respectively. In addition blood of elderly men to improve mental and physical pow- to these SARMs, numerous other chemical structures were ers, and subsequent experiments with saline extracts of dog found to possess SARM-like activities,14 and currently at least testicles demonstrated the overwhelming desire to discover four categories of AR agonists have entered 1 a chemical fountain of youth. Although these attempts were preclinical or clinical trials. These groups of compounds not successful for a variety of endocrinological and chemi- are classified, on the basis of their chemical core struc- 2 cal reasons, the hormone quest had started. Following the tures, into (1) arylpropionamides, (2) bicyclic hydantoins, discovery of the anabolic androgenic principal (3) quinolines, and (4) tetrahydroquinolines; however, new 3,4 5,6 7,8 in 1935, a major goal of early and recent research substances with SARM properties are constantly reported, in biochemistry was the separation of anabolic and and there is great medicinal interest in this novel class of androgenic effects, as well as probing for tissue selectivity therapeutics.12 Major advantages of SARMs over in of potential therapeutics that would enable the treatment replacement therapies are the considerably reduced unde- or prevention of debilitating diseases, e.g. muscular dys- sirable effects such as hepatic toxicity, decreased levels of 9 10–12 trophy, benign prostate hyperplasia or osteoporosis. HDL cholesterol, gynecomastia, and negative influences on A scientific breakthrough in this regard was accomplished prostate and cardiovascular systems.12 Moreover, SARMs in 1998 with the determination of anabolic properties of have demonstrated full anabolic activity in target tissues nonsteroidal agents that were derived from androgen recep- such as muscles and bones, as well as a considerable gain in 13 tor (AR) antagonists such as bicalutamine (Fig. 1, 1) and lean body mass concomitant with a dose-dependent increase in functional performance.15 On the basis of these facts, the World Anti-Doping Agency (WADA) added SARMs to Ł Correspondence to: Mario Thevis, Institute of Biochemistry, the Prohibited List in January 2008.16 Subsequently, dop- Center for Preventive Doping Research, German Sport University Cologne, Carl-Diem Weg 6, 50933 Cologne, Germany. ing control laboratories were urged to establish screening E-mail: [email protected] and confirmation procedures and/or implement the newly

Copyright  2008 John Wiley & Sons, Ltd. 866 M. Thevis and W. Schanzer¨

NC O N O O 2 O S F OH CF3 N F C N H O 3 H OH H C CH3 H3C 3 1 2

R1 3: R = NO , R = F, R = H O 1 2 2 3 4: R = NO , R = NHCOCH , R = H A O B R2 1 2 2 3 3 5: R = NO , R = Cl, R = F CF3 N 1 2 2 3 H OH 6: R1 = CN, R2 = F, R3 = H H3C R3 7: R1 = CN, R2 = CN, R3 = H

O HO O HO O

H3C H C N NO2 N N N N 3 N N N H O CF3 O H3C Cl O 8 9 10

CF3 CH3 CF3 CF3 CF3

N CF3 CH3

O N N CH3 H H O N N O N H H H 11 12 13

H C CH 3 N 3 CH3 CF3 O2N O2N

N CH3 OH OH N N H CH H CH O N H3C 3 H3C 3 H 14 15 16 Figure 1. Chemical structures of arylpropionamide-derived selective androgen receptor modulators with antagonistic [1 () and 2 ()] and agonistic activity (3–7) – advanced representatives of the latter are Andarine (3)and Ostarine (4); hydantoin-derived androgen receptor antagonists [ (8)] and agonists [BMS-564 929 (9)and10]; 2-quinolinone-derived SARMs with antagonistic (11) and agonistic [LG 121 071 (12), LGD 2226 (13), and 14] activity; and tetrahydroquinoline-derived SARMs with bicyclic (S-40 503, 15) and tricyclic (16) nuclei.

261.0488 100 m/z 150 90 m/z 205 80

O2N 70 O A H 60 O B N CF N 3 CH3 H H3C OH 50 O 40 30 150.0560 Relative abundance (%) 289.0436 20 [M-H]- 10 440.1067 205.0228

120 160 200 240 280 320 360 400 440 m/z

Figure 2. ESI product spectrum of [M H] D 440 of Ostarine recorded on an LTQ Orbitrap.

Copyright  2008 John Wiley & Sons, Ltd. J. Mass Spectrom. 2008; 43: 865–876 DOI: 10.1002/jms Mass spectrometry of SARMs 867

O2N O O2N H O O N F3C N O F3C N H3C OH H C H3C 3 O m/z 289 m/z 440

O2N

O2N F3C N O F3C NH CH2

H3C m/z 205 m/z 261 Scheme 1. Principal dissociation pathways of arylpropionamide-based SARMs demonstrated with Ostarine (4). added target compounds into existing assays.17 As most 278.0693 (a) 100 detection methods applied to doping control samples are m/z 193 90 based on mass spectrometric approaches,18–20 information HO O 80 10 about the behavior of these new analytes under different ion- 5 11 6 4 9 ization and dissociation conditions are of utmost importance. 70 7 3 1 N 12 N In this review, we present the mass spectra of selected SARMs 8 N 2 16 60 14 13 [M+H]+ with arylpropionamide-, bicyclic hydantoin-, quinoline-, and O H3C Cl 260.0589 50 306.0645 tetrahydroquinoline-based structures using electrospray ion- m/z 96 40 (- H O) ization (ESI) with collision-induced dissociation (CID) or 2 288.0537 262.0379 electron ionization (EI). 30

Relative abundance (%) 20 86.0598 ELECTROSPRAY IONIZATION – MASS 10 193.0163 96.0442 SPECTROMETRY 244.0637 80 100 120 140 160 180 200 220 240 260 280 300 Using soft ionization techniques such as ESI, protonated m/z or deprotonated molecules of the selected SARMs were generated. The CID of these precursor in MS/MS HO O 10 11 n 5 286.0384 and MS experiments provided comprehensive structural (b) 100 6 4 9 7 1 3N N information, which will facilitate the identification of such 8 2 12 90 N 16 compounds, related substances and potential metabolic 14 13 80 O H3C Cl products in doping control samples. 242.0487 - 70 MS3 [M-H] Arylpropionamide-derived SARMs 60 304.0485 260.0593 Several arylpropionamide-derived SARMs, described as 50 191.0017 promising drug candidates,21 have structural differences 40 limited to the number and nature of ring substituents as 165.0226 outlined with selected examples in Fig. 1 (3–7). All of these 30 260.0592 substances are efficiently deprotonated, using ESI, yielding Relative abundance (%) 20 109.2983 22 m/z abundant [M H] ions. Employing CID, diagnostic prod- 10 uct ions were generated that allowed the characterization of 112.0409 193.0174 248.0229 both aromatic ring systems (A and B) as illustrated with 80 100 120 140 160 180 200 220 240 260 280 300 the product ion mass spectrum of Ostarine (Fig. 2). Depro- m/z tonation of SARMs bearing propionanilide-derived nuclei Figure 3. ESI product ion spectra of (a) [M C H]C D 306 and is suggested to occur at the amide nitrogen due to the (b) [M H] D 304 of BMS-564 929 recorded on an LTQ acidity of the respective hydrogen, which results from sig- Orbitrap. The inset of (b) contains the MS3 spectrum generated nificant electron-withdrawing inductive effects (–I-effects) from m/z 260. exerted by substituents such as the trifluoromethyl- and nitro-functions. The deprotonated molecule of 4 eliminates N-(4-hydroxyphenyl)-acetamide yielding the product ion at the B-ring is followed by the release of carbon monoxide m/z 289,23 which corresponds to m/z 269 in case of ana- giving rise to the most abundant product ion at m/z 261 24 lytes with a cyano residue located at R1,andthelossof (corresponding to m/z 241 with R1 D CN). The A-ring

Copyright  2008 John Wiley & Sons, Ltd. J. Mass Spectrom. 2008; 43: 865–876 DOI: 10.1002/jms 868 M. Thevis and W. Schanzer¨

H+ H+

HO O 10 11 + N (m/z 114) 5 6 4 9 7 1 3N N O C N N OH 8 N 2 12 16 14 13 m/z 96 H C Cl O 15 Cl 3 m/z 306 m/z 193 H+ - H O - CH CHO 2 3 H+ O O - CO N N N N N N N O H C Cl + 3 H O H3C Cl m/z 262 HO m/z 288 Cl N CH N 3 - CO - H O O 2 N m/z 278 H+

Cl N CH N 3 O m/z 260 Scheme 2. Fragmentation pattern of the protonated hydroxybicyclic hydantoin-derived SARM BMS-564 929 (9).

H O O O - - N N O N N N N H O O Cl Cl

m/z 304

- CO2 - H2O

O O 6 5 4 7 3 N N N N H N - C4H7N 8 NH H Cl O H3C Cl m/z 260 m/z 286 O C N - N

H2C Cl

- CH2CHCCH m/z 191

H O 6 O 5 4 7 3 - HN=CO N CN N CN 8 CH2 N H2C H H3C Cl Cl O

m/z 243 Scheme 3. Dissociation pathways of deprotonated hydroxybicyclic hydantoin-derived SARMs BMS-564 929 (9).

Copyright  2008 John Wiley & Sons, Ltd. J. Mass Spectrom. 2008; 43: 865–876 DOI: 10.1002/jms Mass spectrometry of SARMs 869 is further characterized by the formation of product ions and high muscle-tissue selectivity.27–29 Drug candidates with corresponding to bisubstituted and deprotonated anilines, bicyclic hydantoin core (Fig. 1, 9 and 10) are readily proto- e.g. 4-nitro-3-trifluoromethyl-aniline at m/z 205 in case of 4 nated as well as deprotonated using ESI, and product ion (Fig. 2). A corresponding product ion at m/z 185 was found mass spectra are diagnostic (Fig. 3). in CID-spectra of 6. The elucidation of the B-ring structures Hydantoins are likely protonated at either of the nitro- resulted from the generation of product ions representing gens or the carbonyl oxygens30 with a thermodynamically the deprotonated and substituted hydroxyphenyl residues, favored initial O-protonation.31 The complex structure of which yielded a product ion at m/z 150 in case of 4 (Fig. 2, respective SARMs, however, modifies the proton affinity of D D 23,24 R2 NHCOCH3,R3 H). The principal dissociation N-3 resulting in a preferred protonation at N-1 and carbonyl route of arylpropionamide-derived SARMs is summarized residues as substantiated by density functional theory (DFT) in Scheme 1. Using a few but diagnostic product ions, modi- calculations.32 Still, the mobile nature of protons (mobile fications of either ring system can be detected, allowing the proton model33,34), particularly after the excitation of ion- identification of metabolic products23,25,26 as well as modified ized molecules under CID conditions, allows dissociation designer analogs. pathways starting from both options. BMS-564 929 (9) yields Bicyclic hydantoin-derived SARMs a protonated molecule at m/z 306, which eliminated water (18 u) and carbon monoxide (28 u) in either sequence to Hydroxybicyclohydantoin-derived SARMs are structurally related to hydantoin-based AR antagonists such as Nilu- generate product ions at m/z 288, 278 and 260 (Fig. 3(a)). tamide (Fig. 1, 8), but the presence of a hydroxylated Moreover, the loss of acetaldehyde is characteristic of a five-membered ring, e.g. in BMS-564 929 (Fig. 1, 9), enables 6-hydroxylated bicyclic hydantoin such as 9, while other excellent AR binding affinities with activating properties analogs with 7-hydroxylation, for instance, are lacking this particular fragment in CID spectra.32 Additional product ions found at m/z 193 and 96 are complementary fragments CH3 CF3 originating from a cleavage of the hydantoin core follow- 100 256.0819 N CH3 ing the fission of the linkages between N-1 and C-2 as well 90 as N-3 and C-4.30 Evidence for this dissociation route was 80 O N H obtained by the analysis of stable isotope-labeled analogs to 32 70 257.0897 9 (Scheme 2). 60 257.0897 Using negative ionization, deprotonation of 9 is sug- MS3 285-257 50 gested to result from the abstraction of the protons located either at C-5 or the hydroxyl function at C-6. Under CID 40 229.0583 conditions, the MS2 product ion spectrum contains only few 30 188.0944 but informative ions, specifically characterizing the struc- Relative Abundance (%) 20 267.1104 ture of BMS-564 929 (Fig. 3(b)), which was substantiated by 241.0585 [M+H]+ 10 285.1211 MS3 experiments. The loss of carbon dioxide (44 u) from 229.0586 [M H] (m/z 304) yielding the product ion at m/z 260 80 100 120 140 160 180 200 220 240 260 280 requires a complex rearrangement involving an intermediate m/z six-membered ring structure. The hydroxyl function located Figure 4. ESI product ion spectrum of [M C H]C D 285 of a at C-6 plays a key role as analogs lacking this residue do bisalkylated 2-quinolinone-derived SARM recorded on an LTQ not show the loss of CO2.Subsequently,m/z 260 releases Orbitrap. the newly formed 2,3-dihydro-1H-pyrrole (69 u) residue

CH CF 3 CF 3 - 29 Da 3 . + . + N CH3 ( C2H5) N CH3 H H O N O N H H m/z 285 m/z 256

- 15 Da (. CH3)

CF3 . CF H + - 28 Da 3 . + ( CH2N) N CH2

O N H O N m/z 213 H m/z 241 Scheme 4. Principal dissociation routes of protonated 2-quinolinone-derived SARMs demonstrated with compound 14.

Copyright  2008 John Wiley & Sons, Ltd. J. Mass Spectrom. 2008; 43: 865–876 DOI: 10.1002/jms 870 M. Thevis and W. Schanzer¨

yielding the product ion at m/z 191 (Fig. 3(b), inset). Comple- 100 272.1520 mentarily, the deprotonated molecule of 9 eliminates water 90 (18 u), and consecutive losses of imino-methanone (43 u) and but-1-en-3-yne (52 u) also give rise to m/z 191, as 80 O N summarized in Scheme 3. 70 2 [M+H]+ 60 OH N 289.1547 Quinolinone-derived SARMs H 50 H C CH3 Several bi- and tricyclic quinoline derivatives have agonistic 3 35–40 40 SARM-like activity, two of which (LG 121 071 and LGD m/z 216 2226) are depicted in Fig. 1 (12 and 13, respectively). In 30 216.0892 271.1442

analogy to AR antagonists such as LG 120 907 (Fig. 1, 11), Relative abundance (%) 20 199.0866 259.1441 203.0814 these include a 4-trifluoromethyl-2-quinolinone nucleus, but 10 175.0500 223.1077 different C-ring substituents enable the activation of AR. 151.0500 243.1617 The keto-function of the A-ring and the ethyl residue at the 80 100 120 140 160 180 200 220 240 260 280 C-ring presumably mimic the 3-keto- and 17-OH-functions of m/z testosterone.21 Instead of a C-ring, LGD 2226 (Scheme 6, 13) Figure 5. ESI product ion spectrum of [M C H]C D 289 of a includes a 6-located bis(trifluoroethyl)amine residue at the tricyclic tetrahydroquinolinone-derived SARM recorded on an 4-trifluoromethyl-2-quinolinone nucleus and demonstrated LTQ Orbitrap. considerable tissue selectivity and AR binding affinities.41,42 A comprehensive series with alternative alkylations such as bisethylation of the amino function (Fig. 1, 14) has been system of 2-quinolinones that promotes the generation of tested.40 radical cations under CID conditions. The resulting odd- Quinolinone-derived SARMs are efficiently ionized using electron ion at m/z 256 further dissociates by eliminating positive ESI, and common as well as unique dissociation a methyl radical (15 u) to yield a core product ion at pathways were observed for these compounds (e.g. sub- m/z 241 that represents the common nucleus of bisalky- stance 14,Fig.1).43,44 The protonated molecule (m/z 285), lated 4-trifluoro-2-quinolinones.46 In subsequent MS3 exper- preferably eliminated ethylene (28 u) and an ethyl rad- iments, the even-electron product ion at m/z 241 released a ical (29 u), yielding the product ions at m/z 256 and methyleneamine radical (28 u) giving rise to the product 257 (Fig. 4). While the first mentioned pathway follows ion at m/z 213 (Scheme 4). The alternation between even- the commonly accepted even-electron rule,43,45 the loss of and odd-electron ions may be due to the unusual properties the ethyl radical is attributed to the conjugated -electron of 2-quinolinones to form stable radical cations. In contrast

10 O 9 O - CH O 11 O + -.OH 2 N+ 4 N (-17 u) N+ (- 30 u) HO 3 HO 14 . OH OH 1 N 2 12 N N H 13 H H m/z 272 m/z 259

- . CH(CH3)2CH2OH O (-73 u) O N+ N+ HO HO O + N N N H O H m/z 215 N+ m/z 217 N HO . m/z 199 N O H 4 N+ m/z 216 HO 9 3 2 10 14 1N 11 12 H 13

- C6H12 (-84 u) O N+ HO

N

m/z 175

Scheme 5. Principal dissociation pathways of protonated tricyclic tetrahydroquinoline-based SARMs as shown for compound 16.

Copyright  2008 John Wiley & Sons, Ltd. J. Mass Spectrom. 2008; 43: 865–876 DOI: 10.1002/jms Mass spectrometry of SARMs 871

73 protonation is not clear as the proton affinity of 1,2,3,4- 100 tetrahydroquinoline of 225 kcal/mol50 is higher than the 90 corresponding affinity of nitrobenzene (164 kcal/mol),51 but 80 241 the – I-effect caused by the NO2 residue is not accounted for. Moreover, based on the mobile proton model,33 CID can 70 O2N O induce proton migration and trigger dissociation processes 60 O F at various sites of the molecule. Hence, the loss of the OH- 50 CF3 N radical might result from immediate protonation at the nitro H O H C function or after proton transfer. In addition to the loss of 40 3 ž TMS OH, losses of water ( 18 u), formaldehyde ( 30 u), and 30 151 the two-linked side chain with homolytic or heterolytic Relative abundance (%) 459 20 187 cleavages are observed yielding the product ions at m/z +. 115 M 271, 259 and 217, 216 and 215, respectively (Fig. 4). The 10 95 349 474 elimination of the hydroxyl radical (m/z 272) is followed by the loss of a 2-methyl propanol radical (73 u) yielding m/z 50 100 150 200 250 300 350 400 450 199, and the release of formaldehyde from the precursor ion m/z resulting in the fragment at m/z 259 is followed by the loss of Figure 6. EI mass spectrum of the TMS-derivative of Andarine 4-methylpent-2-ene (84 u) yielding 6-nitroquinoline (m/z recorded on an Agilent 6890/5973 GC–MSD. 175), which necessitates an intramolecular rearrangement (Scheme 5). Dissociation pathways were supported by high- to bisalkylated 2-quinolinones, monoalkylated analogs were resolution/high-accuracy mass spectrometry and analysis of reported to yield a common product ion at m/z 228 (instead of chemically synthesized standards. m/z 241), which represents the 6-amino-4-trifluoromethyl- 46 1H-quinolin-2-one core. Precursor ion scanning utilizing ELECTRON IONIZATION – MASS diagnostic product ions such as m/z 241 and 228 should SPECTROMETRY enable the detection of these known compounds as well as unknown, structurally related substances and metabolites. GC–MS still plays an important role in sports drug testing. Most of the commonly employed doping control analytical Tricyclic tetrahydroquinoline-derived SARMs methods using GC–MS require conversion of target analytes , , In addition to quinoline-based SARMs, tetrahydroquinoline- to trimethylsilylated (TMS) derivatives.18 20 52–55 Elucidation derived drug candidates were reported to possess tissue- of EI mass spectra of target analytes after trimethylsilyla- selective AR agonist activity.47,48 Two representatives are tion is of great interest since the existing GC–MS-based depicted in Fig. 1 (compounds 15 and 16), and mass spectral procedures can be adapted for the analysis of many SARMs. data are available for the tricyclic derivative 16 (Fig. 5).49 The protonated molecule at m/z 289 dissociates under Arylpropionamide-derived SARMs CID conditions by the loss of a hydroxyl radical (17 u) Andarine and Ostarine (Fig. 1, 3 and 4), which repre- originating from the nitro function.44 Thesiteofinitial sent advanced arylpropionamide-derived SARMs, have a

O2N O -233 u O F O F + O F CF N + 3 H H C O O O 3 + TMS TMS m/z 474 TMS m/z 241 m/z 241 -TMSOH (-90 u)

O N O2N 2 O O -125 u O F + O F CF3 N O CF N + + 3 H H O H2C H3C TMS H TMS m/z 349 m/z 151 m/z 474 -NO2 (-46 u)

. -CH2CHOTMS (-116 u) + CF N C O 3 H m/z 187 Scheme 6. Principal fragmentation routes of the TMS-derivative of compound 3 after EI.

Copyright  2008 John Wiley & Sons, Ltd. J. Mass Spectrom. 2008; 43: 865–876 DOI: 10.1002/jms 872 M. Thevis and W. Schanzer¨

free hydroxyl function that is readily derivatized using consist of the carbons C-6–C-8 including the O-TMS residue. trimethylsilylating agents. Consequently, the molecular This ion derived from the hydroxylated and condensed weight of the analytes is incremented by 72 u, and dissoci- five-member ring structure was also described for various ation pathways are considerably influenced by the presence steroids bearing a 17-hydroxylated and TMS D-ring18,57–59 of TMS residues (e.g. Andarine, Fig. 6). The molecular ion and is characteristic of EI mass spectra derived from 6- observed at m/z 474 eliminated a methyl radical (15 u), hydroxylated bicyclic hydantoins. yielding the ion at m/z 459, but cleavages of C–C bonds comprising the central chiral carbon atom predominated. Quinolinone-derived SARMs The loss of a 1-fluoro-4-methoxy-benzene radical yielded the The dissociation route of TMS-derivatives of 2-quinolinone- fragment ion at m/z 349, while the release of a N-(4-nitro-3- based SARMs after EI is primarily characterized by the loss trifluoromethyl-phenyl)formamide radical gave rise to m/z 241.49 Subsequent dissociations of m/z 349 or 241 yielded 129 the ions at m/z 187 or 151 by losses of NO2 (46 u) and 100 CH2CHOTMS (116 u) or TMSOH (90 u), respectively, as 90 m/z 129 demonstrated in MS3 experiments (Scheme 6). 80 TMS Bicyclic hydantoin-derived SARMs 70 O O Trimethylsilylation of the hydroxybicyclohydantoin-derived 60 73 N N SARM BMS-564 929 (Fig. 1, 9) yields a molecule with a 50 N monoisotopic mass of 377 u, which decomposes using EI O H C Cl to product ions at m/z 362, 349, 321, and 129 (Fig. 7). 40 3 The first two ions are attributed to the losses of a methyl 30 . Relative abundance (%) + radical (15 u) and a molecule of carbon monoxide (28 u), M 20 321 377 respectively. The latter was reported in earlier studies on 362 100 the fragmentation of hydantoins,56 but the elimination of 10 185 207 265287 349 56 u yielding the ion at m/z 321 is distinctive for bicyclic 60 100 140 180 220 260 300 340 380 hydantoins. This may result from the loss of propenal that m/z necessitates the migration of the TMS residue from the hydroxyl function to the C-3-linked oxygen (Scheme 7). Figure 7. EI mass spectrum of the TMS-derivative of The most abundant fragment at m/z 129 is proposed to BMS-564 929 recorded on an Agilent 6890/5973 GC–MSD.

H3C CH3 + Si O. O H C CH H3C 3 3 + - 248 u Si O H N N H C N 3

O H3C Cl CH2 m/z 377 m/z 129

H C H3C 3 H C CH3 H3C CH3 3 Si Si + .O O +O O . N N H N N N N O H C Cl O H3C Cl 3

m/z 377 m/z 377

CH3 H3C Si - propenal O+ (- 56 u) H3C . N N HN

O H3C Cl

m/z 321 Scheme 7. Principal fragmentation routes of the TMS-derivative of compound 9 after EI.

Copyright  2008 John Wiley & Sons, Ltd. J. Mass Spectrom. 2008; 43: 865–876 DOI: 10.1002/jms Mass spectrometry of SARMs 873

CF3 CF3

N CF3 H C CH 3 3. + Si O N H C 3 m/z 464

- 15 u

CF3 CF CF3 3 N CH N CF3 2 - 152 u H C H3C 3 + + Si O N Si O N H3C H3C m/z 449 m/z 297

- 84 u - 28 u

CF3 CF CF 3 N 3 . - 96 u H C H C 3 + 3 + Si O N Si O N

H3C H3C m/z 365 m/z 269 Scheme 8. Principal fragmentation routes of the TMS-derivative of compound 13 after EI.

215 100 449 100

90 CF3 90 CF3 80 80 N CF3 70 +. 70 M O2N 60 TMS O N 464 60 297 O TMS 50 50 N

40 40 395 30 190 30 169 Relative abundance (%) Relative abundance (%) 20 20 . 445 73 M+ 73 148 311 365 10 10 360 240269 154 199 345 59 89 115 185 241255 60 100 140 180 220 260 300 340 380 420 460 60 100 140 180 220 260 300 340 m/z m/z Figure 8. EI mass spectrum of the TMS-derivative of Figure 9. EI mass spectrum of the TMS-derivative of LGD-2226 recorded on an Agilent 6890/5973 GC–MSD. compound 16 recorded on an Agilent 6890/5973 GC–MSD. of a methyl radical (15 u) and subsequent fragmentation oftheN-linkedalkylsidechains,e.g.ionsareobservedfor compound 13 (Fig. 1) at m/z 365, 297 and 269 (Fig. 8).60 The Tricyclic tetrahydroquinoline-derived SARMs release of a methyl group from the TMS functionality is fol- The EI mass spectra derived from tricyclic lowed by the eliminations of 1,1,1,3,3,3-hexafluoropropane tetrahydroquinoline-based SARMs such as compound 16 (152 u) or trifluoroethane (84 u) yielding m/z 297 and (Fig. 9) produced product ions due to the tricyclic nucleus 365, respectively. Both fragment ions were shown to pro- after elimination of the side chain. Such ions are found at m/z duce m/z 269 by the losses of methyleneamine- or 2,2,2- 215 and 169, which are proposed to consist of the tetrahydro- trifluoroethylideneamine-radicals, respectively (Scheme 8). cyclopenta-quinoline core with and without nitro function, In addition to these dissociation routes, the loss of a tri- respectively. The molecular ion (Fig. 9, m/z 360) was com- fluoromethyl radical (69 u) was observed leading to the monlyobservedaswellasanM-15ion,whichwasfoundto fragment ion at m/z 395. be independent from trimethylsilylation.49

Copyright  2008 John Wiley & Sons, Ltd. J. Mass Spectrom. 2008; 43: 865–876 DOI: 10.1002/jms 874 M. Thevis and W. Schanzer¨

ANALYTICAL CHALLENGE

The enormous structural heterogeneity of SARMs, the limited knowledge of their metabolism and the fact that drugs, which have been misused in sports are not necessarily pharmaceutically and/or clinically approved,17,61–64 presents a challenge for doping control authorities. Comprehensive screening for representatives of each category of emerging SARMs requires concerted analytical activities including ) Analyzer References z

/ LC–MS/MS and GC–MS and, thus, detailed information on m the mass spectrometric behavior under a variety of ionization and dissociation conditions summarized in Table 1. Arylpropionamide-based SARMs (e.g. compounds 3 and 4) were selectively and sensitively analyzed using LC–MS/MS-based assays24 with limits of detection (LODs) below 1 ng/ml of urine. Their poor gas-chromatographic properties with or without derivatization did not allow for adequate LODs using GC–MS procedures. In contrast, ) Major fragment ions (

z the rather low-ionization efficiency of bicyclic hydantoins /

m such as compound 9 using positive or negative ESI neces- sitated the use of adduct ion formation to screen for this ion ( Molecular drug candidate and related compounds at LODs of at least 20 ng/ml.65 Here, GC–MS yielded better results improving the detection limit to 10 ng/ml.49 2-Quinolinone- and tri- cyclic tetrahydroquinolinone-based SARMs were efficiently 60 46 Analyzer/ analyzed with established GC–MS and LC–MS/MS Dissociation

) methods. GC–MS-based procedures were slightly better z / due to the considerable volatility of 2-quinolinone-derived m SARMs, particularly of compound 13, due to the presence of nine fluorine atoms. LODs below 1 ng/ml in spiked urine specimens were obtained by both analytical approaches. ESI EI (TMS-derivative)

CONCLUSION

289 261289 261288 278 205286 260 205375 310 260272 Ion trap/CID 271 248 Ion trap/CID 241 Ion 474 trap/CID 259 Ion 585 trap/CID 459 Ion 377 trap/CID 241 570 Ion trap/CID 480 362 464Mass 321 151 360 449 222 spectrometry 395 345 129 Quadrupole 215 Quadrupole 297 24,49 Quadrupole 169 24,49 32,49 is Quadrupole an Quadrupoleindispensable 46,60 49 tool for 32,49 sports drug C C C

) Major product ions ( testing. Details on the dissociation behavior of new, emerging z H] H] H] H] H] H]

/ drugs under various ionization and fragmentation condi- C C C m tions is needed to comprehensively screen for these com- ion ( Precursor pounds in doping control samples. Rapid implementation of new analytes into detection assays will minimize their use by amateur and professional athletes. Structural character- istics of classes of compounds that enable the identification of conserved features or molecular nuclei, providing a mass 305 304 [M

Mol wt spectrometric ‘signature’ of substances and their metabolic

(monoisotopic) products in complex biological matrices, will allow their sensitive and specific detection.

Acknowledgements The authors thank the Federal Office of Sports, Magglingen, Switzerland, and the Manfred-Donike-Institute for Doping Analysis, Cologne, Germany, for supporting the presented work.

REFERENCES

1. de Kruif P. TheMaleHormone. Harcourt, Brace and Company: ArylpropionamideBicyclic hydantoin2-quinolinoneTricyclic tetrahydroquinoline 441 288 305 440 [M 289 [M 306 [M 392 393 [M Arylpropionamide 402 401 [M New York, 1945. 2. Maisel AQ. The Hormone Quest. Random House: New York, 1965. Summary of typical product ions derived from representatives of four classes of SARMs using ESI–MS/MS or EI–MS 3. Butenandt A, Hanisch G. Uber¨ testosteron. Umwandlung des dehydroandrosterons in androstendiol und testosteron, ein weg zur darstellung des tesotosterons aus cholesterin. Hoppe-Seyler’s 4 9 13 16 Table 1. Compound3 Class Zeitschrift fur Physiologische Chemie 1935; 237: 89.

Copyright  2008 John Wiley & Sons, Ltd. J. Mass Spectrom. 2008; 43: 865–876 DOI: 10.1002/jms Mass spectrometry of SARMs 875

4. Ruzicka L, Wettstein A. Sexualhormone VII. Uber¨ die kunstliche¨ receptor modulator (SARM) in rats–implication of molecular Herstellung des testikelhormons testosteron (Androsten-3-on- properties and intensive metabolic profile to investigate ideal 17-ol). Helvetica Chimica Acta 1935; 18: 1264. pharmacokinetic characteristics of a propanamide in preclinical 5. Kochakian CD. Metabolic effects of anabolic-androgenic steroids study. Drug Metabolism and Disposition 2006; 34: 483. in experimental animals. In Anabolic-Androgenic Steroids, vol. 43, 26. Kuuranne T, Leinonen A, Schanzer¨ W, Kamber M, Kostiainen R, Kochakian CD (ed). Springer: Berlin, 1976; 5. Thevis M. Aryl-propionamide-derived selective androgen 6. Ruzicka L, Goldberg MW, Rosenberg HR. Sexualhormone X. receptor modulators: LC-MS/MS characterization of the in vitro Herstellung des 17-Methyltestosteron und anderer androsten- synthesized metabolites for doping control purposes. Drug und androstanderivate. Zusammenhange¨ zwischen chemischer Metabolism and Disposition 2008; 36: 571. konstitution und mannlicher¨ hormonwirkung. Helvetica Chimica 27. Ostrowski J, Kuhns JE, Lupisella JA, Manfredi MC, Beehler BC, Acta 1935; 18: 1487. Krystek SR, Bi Y, Sun C, Seethala R, Golla R, Sleph PG, 7. Rommerts FFG. Testosterone: an overview of biosynthesis, Fura A, An Y, Kish KF, Sack JS, Mookhtiar KA, Grover GJ, transport, metabolism and non-genomic actions. In Hamann LG Jr. Pharmacological and X-ray structural Testosterone – Action, Deficiency, Substitution,NieschlagE, characterization of a novel selective androgen receptor Behre HM (eds). Cambridge University Press: Cambridge, 2004; modulator: potent hyperanabolic stimulation of skeletal muscle 1. with hypostimulation of prostate in rats. Endocrinology 2006; 8. Gooren LJ, Bunck MC. Androgen replacement therapy: present 48:4. and future. Drugs 2004; 64: 1861. 28. Sun C, Robl JA, Wang TC, Huang Y, Kuhns JE, Lupisella JA, 9. Chen J, Kim J, Dalton JT. Discovery and therapeutic promise of Beehler BC, Golla R, Sleph PG, Seethala R, Fura A, Krystek SR, selective androgen receptor modulators. Molecular Interventions An Y, Malley MF, Sack JS, Salvati ME, Grover GJ, Ostrowski J, 2005; 5: 173. Hamann LG Jr. Discovery of potent, orally-active, and muscle- 10. Bagatell CJ, Bremner WJ. in men–uses and abuses. selective androgen receptor modulators based on an N-aryl- New England Journal of Medicine 1996; 334: 707. hydroxybicyclohydantoin scaffold. Journal of Medicinal Chemistry 11. Negro-Vilar A. Selective androgen receptor modulators 2006; 49: 7596. (SARMs): a novel approach to androgen therapy for the new 29. Hamann LG, Manfredi MC, Sun C, Krystek SR, Huang Y, Bi Y, millennium. Journal of Clinical Endocrinology and Metabolism 1999; Augeri DJ, Wang T, Zou Y, Betebenner DA, Fura A, Seethala R, 84: 3459. Golla R, Kuhns JE, Lupisella JA, Darienzo CJ, Custer LL, 12. Cadilla R, Turnbull P. Selective androgen receptor modulators in Price JL, Johnson JM, Biller SA, Zahler R, Ostrowski J Jr. Tandem drug discovery: medicinal chemistry and therapeutic potential. optimization of target activity and elimination of mutagenic Current Topics in Medicinal Chemistry 2006; 6: 245. potential in a potent series of N-aryl bicyclic hydantoin-based 13. Dalton JT, Mukherjee A, Zhu Z, Kirkovsky L, Miller DD. selective androgen receptor modulators. Bioorganic and Medicinal Discovery of nonsteroidal androgens. Biochemical and Biophysical Chemistry Letters 2007; 17: 1860. Research Communications 1998; 244:1. 30. Shen JX, Brodbelt J. Reactions of hydantoin and succinimide 14. Mohler ML, Nair VA, Hwang DJ, Rakov IM, Patil R, Miller DD. anticonvulsants with dimethyl ether ions in a quadrupole ion Nonsteroidal tissue selective androgen receptor modulators: trap mass spectrometer. Journal of Mass Spectrometry 1996; 31: a promising class of clinical candidates. Expert Opinion on 1389. Therapeutic Patents 2005; 15: 1565. 31. Wang F, Ma S, Zhang D, Cooks RG. Proton affinity and gas- 15. GTx, Incorporated. Ostarine achieved the primary endpoint phase basicity of urea. Journal of Physical Chemistry 1998; 102: of increasing lean body mass and a secondary endpoint 2988. of improving functional performance, 2006; http://www. 32. Thevis M, Kohler M, Schlorer¨ N, Kamber M, Kuhn A, salesandmarketingnetwork.com/news release.php?ID D 2 015 Linscheid MW, Schanzer¨ W. Mass spectrometry of hydantoin- 328 (08-12-2006). derived selective androgen receptor modulators. Journal of Mass 16. World Anti-Doping Agency. The 2008 prohibited list, 2008; Spectrometry 2008; 43: 639. http://www.wada-ama.org/rtecontent/document/2008 List 33. Wysocki VH, Tsaprailis G, Smith LL, Breci LA. Mobile and En.pdf (28-11-2007). localized protons: a framework for understanding peptide 17. Thevis M, Schanzer¨ W. Emerging drugs – potential for misuse 35 in sport and doping control detection strategies. Mini-Reviews in dissociation. Journal of Mass Spectrometry 2000; : 1399. Medicinal Chemistry 2007; 7: 533. 34. Paizs B, Suhai S. Fragmentation pathways of protonated 18. Thevis M, Schanzer¨ W. Mass spectrometry in doping control peptides. Mass Spectrometry Reviews 2005; 24: 508. analysis. Current Organic Chemistry 2005; 9: 825. 35. Edwards JP, West SJ, Pooley CL, Marschke KB, Farmer LJ, 19. Thevis M, Schanzer¨ W. Current role of LC-MS(/MS) in doping Jones TK. New nonsteroidal androgen receptor modulators control. Analytical and Bioanalytical Chemistry 2007; 388: 1351. based on 4-(trifluoromethyl)-2(1H)-pyrrolidino[3,2-g] quinoli- 20. Thevis M, Schanzer¨ W. Mass spectrometry in sports drug none. Bioorganic and Medicinal Chemistry Letters 1998; 8: 745. testing: structure characterization and analytical assays. Mass 36. Edwards JP, Higuchi RI, Winn DT, Pooley CL, Caferro TR, Spectrometry Reviews 2007; 26: 79. Hamann LG, Zhi L, Marschke KB, Goldman ME, Jones TK. 21. Gao W, Kim J, Dalton JT. Pharmacokinetics and pharmacody- Nonsteroidal androgen receptor agonists based on namics of nonsteroidal androgen receptor ligands. Pharmaceuti- 4-(trifluoromethyl)-2H-pyrano[3,2-g]quinolin-2-one. Bioorganic cal Research 2006; 23: 1641. and Medicinal Chemistry Letters 1999; 9: 1003. 22. Wu Z, Gao W, Phelps MA, Wu D, Miller DD, Dalton JT. 37. Hamann LG, Mani NS, Davis RL, Wang XN, Marschke KB, Favorable effects of weak acids on negative-ion electrospray Jones TK. Discovery of a potent, orally active, nonsteroidal ionization mass spectrometry. Analytical Chemistry 2004; 76: 839. androgen receptor agonist: 4-ethyl-1,2,3,4-tetrahydro-6- 23. Gao W, Wu Z, Bohl CE, Yang J, Miller DD, Dalton JT. (trifluoromethyl)-8-pyridono[5,6-g]- quinoline (LG121071). Characterization of the in vitro metabolism of selective androgen Journal of Medicinal Chemistry 1999; 42: 210. receptor modulator using human, rat, and dog liver enzyme 38. van Oeveren A, Motamedi M, Mani NS, Marschke KB, Lopez FJ, preparations. Drug Metabolism and Disposition 2006; 34: 243. Schrader WT, Negro-Vilar A, Zhi L. Discovery of 6-N,N- 24. Thevis M, Kamber M, Schanzer¨ W. Screening for metabolically bis(2,2,2-trifluoroethyl)amino-4-trifluoromethylquinolin-2(1H)- stable aryl-propionamide-derived selective androgen receptor one as a novel selective androgen receptor modulator. Journal of modulators for doping control purposes. Rapid Communications Medicinal Chemistry 2006; 49: 6143. in Mass Spectrometry 2006; 20: 870. 39. van Oeveren A, Motamedi M, Martinborough E, Zhao S, Shen Y, 25. Wu D, Wu Z, Yang J, Nair VA, Miller DD, Dalton JT. West S, Chang W, Kallel A, Marschke KB, Lopez FJ, Negro- Pharmacokinetics and metabolism of a selective androgen Vilar A, Zhi L. Novel selective androgen receptor modulators:

Copyright  2008 John Wiley & Sons, Ltd. J. Mass Spectrom. 2008; 43: 865–876 DOI: 10.1002/jms 876 M. Thevis and W. Schanzer¨

SAR studies on 6-bisalkylamino-2-quinolinones. Bioorganic and 53. Donike M, Zimmermann J, Barwald¨ KR, Schanzer¨ W, Christ V, Medicinal Chemistry Letters 2007; 17: 1527. Klostermann K, Opfermann G. Routinebestimmung von 40. van Oeveren A, Pio BA, Tegley CM, Higuchi RI, Wu M, anabolika in harn. Deutsche Zeitschrift fur Sportmedizin 1984; Jones TK, Marschke KB, Negro-Vilar A, Zhi L. Discovery of 35: 14. an androgen receptor modulator pharmacophore based on 2- 54. Donike M. Acylierung mit Bis(Acylamiden); N-Methyl- quinolinones. Bioorganic and Medicinal Chemistry Letters 2007; 17: Bis(Trifluoracetamid) und Bis(Trifluoracetamid), zwei neue 1523. reagenzien zur trifluoracetylierung. Journal of Chromatography 41. Wang F, Liu XQ, Li H, Liang KN, Miner JN, Hong M, Kallel EA, 1973; 78: 273. van Oeveren A, Zhi L, Jiang T. Structure of the ligand-binding 55. Hemmersbach P, de la Torre R. Stimulants, narcotics and ˇ- domain (LBD) of human androgen receptor in complex with blockers: 25 years of development in analytical techniques for a selective modulator LGD2226. Acta Crystallographica Section F doping control. Journal of Chromatography B 1996; 687: 221. Structural and Biological Crystallography Communication 2006; 62: 56. Locock RA, Coutts RT. The mass spectra of succinimides, 1067. hydantoins, oxazolidinediones and other medicinal anti- 42. Rosen J, Negro-Vilar A. Novel, non-steroidal, selective androgen epileptic agents. Organic Mass Spectrometry 1970; 3: 735. receptor modulators (SARMs) with anabolic activity in bone and 57. Vouros P, Harvey DJ. Method for Selective Introduction of muscle and improved safety profile. Journal of Musculoskeletal and Trimethylsilyl and perdeuterotrimethylsilyl groups in hydroxy Neuronal Interactions 2002; 2: 222. steroids and its utility in mass spectrometric interpretations. 43. Karni M, Mandelbaum A. The ‘‘Even-Electron-Rule’’. Organic Analytical Chemistry 1973; 45:7. Mass Spectrometry 1980; 15: 53. 58. Durbeck¨ HW, Buker¨ I. Studies on anabolic steroids. The 44. Levsen K, Schiebel HM, Terlouw JK, Jobst KJ, Elend M, Preiss A, mass spectra of 17˛-Methyl-17ˇ-hydroxy-1,4-androstadien-3- Thiele H, Ingendoh A. Even-electron ions: a systematic study of one (Dianabol) and its Metabolites. Biomedical Mass Spectrometry the neutral species lost in the dissociation of quasi-molecular 1980; 7: 437. ions. Journal of Mass Spectrometry 2007; 42: 1024. 59. Schanzer¨ W, Horning S, Opfermann G, Donike M. Gas 45. McLafferty FW, Turecek F. Interpretation of Mass Spectra. chromatography/mass spectometry identification of long- University Science Books: Mill Valley, 1993. term excreted metabolites of the 4-chloro- 46. Thevis M, Kohler M, Maurer J, Schlorer¨ N, Kamber M, 1,2-dehydro-17alpha- in humans. Journal of Schanzer¨ W. Screening for 2-quinolinone-derived selective Steroid Biochemistry and Molecular Biology 1996; 57: 363. androgen receptor agonists in doping control analysis. Rapid 60. Thevis M, Kohler M, Schlorer¨ N, Fussholler¨ G, Schanzer¨ W. Communications in Mass Spectrometry 2007; 21: 3477. Detection of 2-quinolinone-derived SARMs using GC-MS and LC- 47. Hanada K, Furuya K, Yamamoto N, Nejishima H, Ichikawa K, MS/MS. Presented at the 26th Cologne Workshop on Doping Analysis, Nakamura T, Miyakawa M, Amano S, Sumita Y, Oguro N. Bone February 24–29, Cologne, 2008. anabolic effects of S-40 503, a novel nonsteroidal selective 61. Catlin DH, Sekera MH, Ahrens BD, Starcevic B, Chang Y-C, androgen receptor modulator (SARM), in rat models of Hatton CK. : discovery, synthesis, and osteoporosis. Biological and Pharmaceutical Bulletin 2003; 26: 1563. detection in urine. Rapid Communications in Mass Spectometry 48. Hanada K, Furuya K, Inoguchi K, Miyakawa M, Nagata K. 2004; 18: 1245. Kaken Pharmaceutical. Tetrahydroquinoline Derivatives, EP 1 62. Catlin DH, Ahrens BD, Kucherova Y. Detection of norbolethone, 221 439 B1, 2007. an anabolic steroid never marketed, in athletes’ urine. Rapid 49. Thevis M, Kohler M, Schlorer¨ N, Schanzer¨ W. EI- and ESI-MS of Communications in Mass Spectometry 2002; 16: 1273. selective androgen receptor modulators, 2008. 63. Sekera MH, Ahrens BD, Chang YC, Starcevic B, Georgakopou- 50. Murena F, Gioia F. Catalytic hydroprocessing of chlorobenzene- los C, Catlin DH. Another designer steroid: discovery, synthesis, pyridine mixtures. Journal of Hazardous Materials 1998; 60: 271. and detection of ‘madol’ in urine. Rapid Communications in Mass 51. Eckert-Maksic M, Hodoscek M, Kovacek D, Maksic ZB, Spectometry 2005; 19: 781. Primorac M. Theoretical model calculations of the absolute 64. Handelsman DJ. Designer androgens in sport: when too much proton affinities of benzonitrile, nitroso- and nitrobenzene. is never enough. Science Signalling 2004; 244: pe41. Journal of Molecular Structure (Theochem) 1997; 417: 131. 65. Thevis M, Kohler M, Thomas A, Maurer J, Schlorer¨ N, 52. Donike M, Zimmermann J. Zur Darstellung von Trimethylsilyl-, Kamber M, Schanzer¨ W. Determination of benzimidazole- Triethylsilyl- und tert.- Butyldimethylsilyl- enolathern¨ von and bicyclic hydantoin-derived selective androgen receptor ketosteroiden fur¨ gas- chromatographische und massenspek- antagonists and agonists in human urine using LC-MS/MS. trometrische Untersuchungen. Journal of Chromatography 1980; Analytical and Bioanalytical Chemistry 2008; 391: 251. 202: 483.

Copyright  2008 John Wiley & Sons, Ltd. J. Mass Spectrom. 2008; 43: 865–876 DOI: 10.1002/jms