Structure Elucidation of phase I Metabolites of the Microtubule

Article pubs.acs.org/jnp

Structure Elucidation of phase I Metabolites of the Microtubule Perturbagens: Ceratamines A and B Sara E. Smith,*,† Matthew C. Dello Buono,† Daniel J. Carper,‡ Robert S. Coleman,‡ and Billy W. Day†,§ †

Department of Chemistry, University of Pittsburgh, Pittsburgh, Pennsylvania 15260, United States Department of Chemistry, The Ohio State University, Columbus, Ohio 43210, United States § Department of Pharmaceutical Sciences, University of Pittsburgh, Pittsburgh, Pennsylvania 15261, United States ‡

S Supporting Information *

ABSTRACT: The heterocyclic alkaloids, ceratamines A and B, are isolates from a marine Pseudoceratina sp. sponge. They behave as antimitotic agents, with IC50 values in the low micromolar range. The mechanism of this activity involves the disruption of microtubule dynamics; therefore, the ceratamines are of great interest in cancer drug discovery. Studies of in vitro metabolism were performed using rat liver microsomes to begin to understand the pharmacokinetics of these unique natural products. A total of eight metabolites were identified using UV and LC− MS/MS techniques. The majority of metabolites were formed as a result of various demethylation reactions. The formation of two metabolites, M1 and M3, involved monooxygenation, most likely on the aromatic ring, however the exact structure has not been determined. UV absorbance revealed a hypsochromic shift as a result of monooxygenation, an observation that may suggest the loss of aromaticity; however, further investigation is required. The structures of two major metabolites of ceratamine B, M4 and M6, were confirmed by 1H NMR spectroscopy. These metabolites formed as a result of demethylation at the methoxy and aminoimidazole, respectively.

T

overexpression of P-glycoprotein and mutations in the tubulin genes.11,12 The identification of additional microtubule perturbers, especially those with unique structures and novel binding sites, may provide additional options for treating resistant cancers.10,12 To aid in this identification, highthroughput screens have been developed.10 Ceratamines A and B are heterocyclic alkaloids isolated from an extract of a marine Pseudoceratina sp. sponge, which was found to arrest MCF-7 breast carcinoma cells in an assay designed by Roberge et al.10,13 These natural products displayed antimitotic activity with IC50 values of 15 μM and 25 μM, respectively. Their ability to disrupt microtubule dynamics has also been reported.13,14 The ceratamines are most likely not inhibitors of microtubule formation, as purified bovine tubulin polymerized spontaneously in the presence of both compounds. Polymerization did, however, occur in the presence of either compound at a low tubulin concentration, suggesting that the ceratamines are able to disrupt microtubule dynamics through stabilization of tubulin polymerization. The ceratamine binding site has not been identified to date, but a competitive binding assay revealed that they do not bind at the paclitaxel site.13 Ceratamines A and B are not as potent as many traditional microtubule stabilizers, such as paclitaxel or discodermolide; however, their structural differences have led to great excitement.13 Although the ceratamines contain an atypical

he search for natural products with the ability to arrest cells in mitosis has been of interest to the field of cancer drug discovery for several decades.1 Much of this excitement can be attributed to the clinical success of known antimitotic agents such as paclitaxel (Taxol) and docetaxel (Taxotere), both of which, target microtubules.2−4 Microtubules are biological polymers, consisting of subunits of α and β tubulin heterodimers.5 They are a key component in providing structural support to the cell and also behave as a “railway”, allowing the transport of cellular components by motor proteins.6,7 Most importantly, microtubules play a critical role in mitosis, as they are necessary for the formation and function of the mitotic spindle, a structure consisting of astral, kinetochore, and polar microtubules.5 This complex is responsible for the movement of duplicated chromosomes to opposite ends of the cell prior to mitosis.7 Microtubules are formed through a very unique process, with rapid changes occurring between growth and disassembly phases.8 This dynamic instability is essential for mitosis, as the intricate structure of the mitotic spindle requires a vast reorganization of microtubules.5 The disruption of microtubule dynamics, either by preventing tubulin polymerization, or by blocking microtubule disassembly, could result in antimitotic activity. This activity could have the potential to selectively affect highly proliferative cells; therefore, microtubules have been considered a valuable target in cancer research.9 Although effective, a negative aspect of currently used microtubule perturbers is the development of drug resistance.10 This can be ascribed to a number of mechanisms, including © 2014 American Chemical Society and American Society of Pharmacognosy

Received: December 26, 2013 Published: June 25, 2014 1572

dx.doi.org/10.1021/np4010882 | J. Nat. Prod. 2014, 77, 1572−1578

Journal of Natural Products

Article

The goal of this work was to perform a qualitative study that would allow for the identification of phase I metabolites of the natural products, ceratamines A and B, formed in rat liver microsomes. Metabolic incubations were performed at various time points (data not shown). Metabolites were formed in an active enzyme system in the presence of NADPH within 15 min. Several controls were performed, including incubations with heat-inactivated and boiled microsomes to ensure the chemical stability of each compound. Species that were only detected within the active enzyme system with NADPH were considered for structure elucidation. Reversed-phase HPLC was used to separate the newly formed metabolites, with UV being the initial detection method. The chromatograms suggested the presence of at least seven metabolites of ceratamine A and five metabolites of ceratamine B after 60 min of incubation with rat liver microsomes (Figure 2).

imidazo [4,5-d] azepine component, they are structurally simple in comparison to other known microtubule stabilizers. Total syntheses of both compounds have been reported by Coleman et al.15 There has also been a recent suggestion that microtubule stabilizers may be useful in the treatment of tauopathies, such as Alzheimer’s disease. This would require the ability to cross the blood-brain barrier, which is a challenge for most drugs. The simple structures of the ceratamines may aid in the possibility of achieving brain penetration.16 Due to the great potential of these alkaloids in the treatment of cancer and select neurological diseases, early preclinical studies will be useful in lead compound optimization. Therefore, the fate of the ceratamines within rat liver microsomes was examined. This study was performed in a qualitative manner, with the focus on the structure elucidation of phase I metabolites of each compound.

■

RESULTS AND DISCUSSION MS Spectra of Ceratamines A and B. The protonated molecule, [M + H]+, of each of the ceratamines appeared as three ion signals with a 1:3:1 intensity ratio (Figure 1). This Figure 2. RP HPLC-UV chromatograms of 60 min rat microsomal incubations of (A) ceratamine A and (B) ceratamine B.

When comparing the chromatograms obtained for each compound, it became apparent that at least two metabolites appear to have similar retention times. A coinjection experiment was performed in order to confirm this observation (Figure 3). Although the presence of 14 metabolites was

Figure 1. [M + H] ion clusters of (A) ceratamine A and (B) ceratamine B; CID product ions of (C) ceratamine A and (D) ceratamine B.

Figure 3. RP-HPLC-UV chromatogram of a coinjection of 60 min rat microsomal incubations of ceratamine A and B. The coinjection experiment suggests that two metabolites, M4 and M6, are formed for both compounds, and that CB, a metabolite of ceratamine A, has a retention time identical to that of ceratamine B.

ratio corresponds to the presence of two bromine atoms and also served as a useful marker for metabolite identification. Each compound was fragmented into several product ions via collision-induced dissociation (CID). The major CID product ions of interest were m/z 203 and 189 for ceratamine A and B, respectively. These ions were the result of a neutral loss of 264, indicating dissociation at the benzylic position (Figure S1 in the Supporting Information [SI]). This was also supported by accurate mass determinations and fragmentation prediction via PeakView software from ABSciex (Table S1, Figures S2−S3 in the SI). Identification of these major product ions was essential for preliminary structure elucidation of potential metabolites.

suggested by the individual injections, the coinjection experiment resulted in the identification of eight significant metabolites. The metabolites of interest were selected on the basis of the relative abundance, which was calculated by comparing the peak area of the metabolite to the peak area of the parent drug (Table 1). M1−M3 were unique to ceratamine B, while M4 and M6 were common to both compounds. M5 and M7 were unique to ceratamine A. An additional metabolite of ceratamine A, labeled as CB, was found to have a retention time consistent with that of ceratamine B. As the biological activity of ceratamine B has already been established, this observation suggests that the

+

1573



Article

of demethylation.12 The ceratamines also contain an aromatic ring, which is a probable site for hydroxylation.12 The section(s) of the molecule that had been changed due to cytochromes P450-mediated metabolism was determined by comparison of the major product ion with that of the parent drug and also by the consideration of the effects of such changes on the relative polarity of the metabolite. M1. M1, which was found to be unique to ceratamine B, had the shortest retention time, indicating it to be the most polar metabolite. The protonated molecules differed from the parent by +2 Da and had m/z values of 455, 457, and 459 (Figure 4).

Table 1. Relative Abundances of Phase I Metabolites of the Ceratamines after 60 min Incubations with Rat Liver Microsomes relative abundance (%) analyte

ceratamine A

ceratamine B

M1 M2 M3 M4 M5 M6 M7 CB

− − − 0.4 12.4 1.9 24.3 5.8

0.9 3.2 3.3 15.6 − 23.6 − −

overall effects of ceratamine A would result from a combination of effects of both ceratamine A and an active metabolite, CB. Preliminary structure elucidation was performed for the eight metabolites using tandem mass spectrometry. Although the structures of the metabolites were unknown, the types of metabolic reactions were limited to phase I due to the enzyme source and cofactor. Phase I metabolism typically involves simple reactions, leaving the majority of the compound unchanged.17 Therefore, potential metabolites should share similar characteristics with the parent drugs. The presence of ion signals with a 1:3:1 isotopic intensity ratio was used as a marker for metabolites of the ceratamines. The difference in the m/z values of the protonated molecules was then utilized to determine the type of structural change. The proposed location of the structural change(s) was determined by comparing the major product ions of each metabolite to the parent compound. Quadrupole-time-of-flight (QqTOF) mass spectrometry was used as the second method of detection to identify metabolites. QqTOF instruments offer an increased sensitivity in full-scan mode due to an improved duty cycle. QqTOF-MS was used to identify the precursor and major product ion of each metabolite. Although eight phase I metabolites are of interest, there were only four differences in the m/z values of the protonated molecules (Table 2). Therefore, many isobaric species were present which led to challenges in structure elucidation. When examining the structures of the ceratamines, there are several sites that would be expected to be metabolically labile. Dealkylation reactions occur adjacent to a heteroatom; therefore, the methylated aminoimidazole, tertiary amide (for ceratamine A), and methoxy groups are all expected to be sites

Figure 4. (A) [M + H]+ ion cluster of M1, (B) major CID product ion of M1.

This metabolite was most likely formed through two metabolic reactions; monooxygenation and demethylation. The major CID product ion of M1 was m/z 175, which differed by −14 Da from the major product ion of ceratamine B. This indicated that the demethylation occurred within the imidazoazepinone. The only likely site of dealkylation is at the aminoimidazole. On the basis of the comparison of the major product ions, it was suggested that the monooxygenation occurred on the aromatic ring. However, the exact structure cannot be elucidated on the basis of mass spectrometry techniques alone. M2. Unique to ceratamine B, M2 resulted from a −28 Da loss. The protonated molecules had m/z values of 425, 427, and 429, which is consistent with a double demethylation. The major CID product ion had an m/z value of 175, indicating that one demethylation occurred within the imidazoazepinone (Figure 5). M2 is most likely the result of demethylations at the aminoimidazole and the methoxy. M3. The third phase I metabolite was also found to be unique to ceratamine B. The protonated molecules had m/z values of 469, 471, and 473, a difference of +16 Da from the parent. This is consistent with a monooxygenation. The major

Table 2. Chromatographic and Mass Spectrometric Information of Metabolites of the Ceratamines Used for Preliminary Structure Elucidation analyte

tr (min)

M1 M2 M3 M4 M5 M6 M7 CBa

7.2 7.9 13.7 20.1 28.1 29.6 33.6 35.9

a

[M + H]+ (m/z) 455, 425, 469, 439, 453, 439, 453, 453,

457, 427, 471, 441, 455, 441, 455, 455,

459 429 473 443 457 443 457 457

mass difference from ceratamine A [M + H]+ (m/z)

mass difference from ceratamine B [M + H]+ (m/z)

major CID product ion (m/z)

N/A N/A N/A −28 −14 −28 −14 −14

+2 −28 +16 −14 N/A −14 N/A N/A

175 175 189 189 203 175 189 189

Source of CB was from an incubation of ceratamine A with rat liver microsomes. 1574



Article


CID product ion of m/z 189 indicated that this structural change most likely occurred on the aromatic ring, although the specific structure cannot be identified by tandem mass spectrometry (Figure 6).

Figure 7. Major CID product ions of (A) M4 from a rat microsomal incubation of ceratamine A, (B) M4 from a rat microsomal incubation of ceratamine B, (C) M6 from a rat microsomal incubation of ceratamine A, and (D) M6 from a rat microsomal incubation of ceratamine B.

(Figure S5). This difference of −14 Da is indicative of a demethylation reaction. The major product ions were essential in determining the location of dealkylation. M5 fragmented to form m/z 203, which is the same product ion as ceratamine A (Figure 8). Therefore, M5 is the result of a demethylation at the methoxy. M7 and CB shared the same major CID product ion of m/z 189, a loss of 14 Da from the major CID product ion of ceratamine A. This suggested that the demethylation must occur on the imidazoazepinone. The final eluting metabolite of ceratamine A was observed to coelute with ceratamine B during the coinjection experiment. This revealed that a demethylation at the tertiary amide results in a longer retention time and, therefore, a less polar metabolite. Although M7 and CB are isobaric and have identical major product ions, this observation allowed for their structures to be distinguished. It was concluded that M7 was the result of demethylation at the aminoimidazole and CB was formed due to demethylation at the tertiary amide. Confirmatory Structure Elucidation. Further structure elucidation was performed for all eight metabolites using accurate mass determination. Large-scale microsomal incubations were performed for ceratamine B allowing for the isolation of the minor metabolite, M3, and the major metabolites: M4 and M6. Accurate Mass Determinations. Accurate mass determinations were performed using QqTOF-MS in order to support the proposed molecular formulas of the metabolites of the ceratamines (Table 3). The parent compounds were used as internal accurate mass calibrants (ceratamine A + H, 466.97127 Da; ceratamine B + H: 452.95562 Da). With errors typically below 5 ppm, the proposed molecular formulas were supported by accurate mass.


M4 and M6. Two isobaric metabolites were identified in rat liver microsomal incubations of each of the ceratamines. The protonated molecules of these metabolites had m/z values of 439, 441, and 443 (Figure S4). This is a −28 Da difference from ceratamine A and a −14 Da difference from ceratamine B. The molecular weight differences suggest that M4 and M6 are a result of a double demethylation of ceratamine A and a single demethylation of ceratamine B. Although M4 and M6 were found to be isobaric, they differed by their product ions (Figure 7). The major CID product ion of M4 was m/z 189, a −14 Da difference from the major product ion of ceratamine A and the same as the major product ion of ceratamine B. This indicates that only one demethylation reaction occurred within the imidazoazepinone. Since there is only one available site for demethylation within ceratamine B, M4 most likely resulted from a demethylation at the aminoimidazole. The second demethylation reaction must have occurred at the methoxy group, which would not result in a mass change of the major product ion. M6 had a major CID product ion of m/z 175, a loss of 28 Da from the major product ion of ceratamine A and a loss of 14 Da from the major product ion of ceratamine B. This suggests that M6 is the result of a double demethylation from ceratamine A, both of which occur within the imidazoazepinone. A single demethylation of ceratamine B at the aminoimidazole would also form M6. M5, M7, and CB. The final three metabolites were unique to ceratamine A. All three metabolites were isobaric, having protonated molecules with m/z values of 453, 455, and 457 1575



Article

Figure 8. Major CID product ions of (A) M5, (B) M7, and (C) CB.

form as a result of a demethylation at the aminoimidazole. This would not result in any change to the aromatic ring, which is consistent with the observation of no significant shift in the absorbance of the B band. Tandem mass spectrometry data supported the addition of an oxygen atom on the aromatic ring in order to form M3; however, the exact structure of this monooxygenation could not be elucidated. When examining the absorbance data of M3, a hypsochromic shift was noticed at the B band. The addition of a hydroxy would be expected to increase delocalization and result in a bathochromic shift. Therefore, the observed hypsochromic shift does not support an aromatic hydroxylation. The loss of delocalization suggests that the metabolic reaction possibly resulted in a loss of aromaticity. An additional absorbance band was noticed within the UV/ vis region for ceratamine B (λmax: 357 nm). This band was attributed to a π → π* transition of the highly conjugated imidazoazepinone portion of ceratamine B. No change was observed in this absorbance band for M3 or M4. Tandem mass spectrometry suggested no structural changes within the imidazoazepinone portion for either metabolite, therefore no shift in the visible band was expected. A hypsochromic shift (357 to 335 nm) was observed for M6, suggesting that a demethylation at the aminoimidazole would result in a loss of delocalization. 1 H NMR Spectroscopy. 1H NMR was utilized to further confirm the proposed structures of the two major isolated metabolites, M4 and M6. Table 4 displays the 1H NMR

Table 3. Accurate Mass Determination Results for Proposed Molecular Formulas of Metabolites of the Ceratamines

analyte

proposed molecular formula

experimentally determined monoisotopic ion signal (m/z)

theoretical monoisotopic mass (Da)

error (ppm)

M1 M2 M3 M4 M5 M6 M7 CBa

C15H12Br2N4O3 C15H12Br2N4O3 C16H14Br2N4O3 C15H12Br2N4O2 C16H14Br2N4O2 C15H12Br2N4O2 C16H14Br2N4O2 C16H14Br2N4O2

454.9357 424.9235 468.9486 438.9395 452.9570 438.9391 452.9533 452.9577

454.9349 434.9243 468.9505 438.9400 452.9556 438.9400 452.9556 452.9556

1.8 1.9 4.1 1.1 3.1 2.0 5.1 4.6

a

Source of ceratamine B was from an incubation of ceratamine A with rat liver microsomes.

UV/Vis Spectroscopy. Absorbance data was collected for the three isolated metabolites of ceratamine B: M3, M4, and M6 (Figure 9). Two peaks observed within the UV range for all

Table 4. 1H NMR Spectroscopic Data of Ceratamine B, M4, and M6

Figure 9. (A) UV/vis absorbance spectra of ceratamine B, M3, M4, and M6 showing E band, B band, and imidazoazepinone band. (B) UV/vis absorbance spectra of ceratamine B, M3, M4, and M6, focusing on the B band and imidazoazepinone band.

three metabolites and ceratamine B were considered to be the E and B bands, consistent with π → π* transitions of the dibrominated benzene moiety. The high substitution of the benzene ring caused a bathochromic shift of the E band, allowing it to be observed within the UV region (λmax ≈ 203 nm). The B band of ceratamine B has a λmax of 268 nm. The presence of an electron-donating group caused a red shift of the maximum absorbance of the B band.18 When comparing the E and B bands of the metabolites to ceratamine B, no major differences were noticed for M4 or M6. Preliminary structure elucidation suggested that M4 was formed as a result of a demethylation of the methoxy. Although this structural change occurred on the aromatic ring, M4 still contains an electron-donating group; thus, a significant change in the B band was not expected.18 M6 has been proposed to

position

ceratamine B δh, (J in Hz)

M4 δh, (J in Hz)

M6 δh, (J in Hz)

1 2 3 4 5 6 7 8

3.07, d (4.8) 8.67, br q (4.8) 6.42, d (9) 7.15, br t (7.8) 11.40, br d (6) 4.17, s 7.70, s 3.70, s

3.06, d (5.4) 8.67, br d 6.40, d (9) 7.10, br d (9) 11.30, br s 4.10, s 7.90, s 6.57, s

n.d.a 8.53, br s 6.48, d (9) 7.21, br t (8.4) n.d.a 4.15, s 7.95, s 3.73, s

a

n.d.: not detected.

spectroscopic data for ceratamine B, M4, and M6. Ceratamine B contains a total of 15 protons at eight different positions (Figure 10 and S6 in the SI). Preliminary structure elucidation suggested that M4 formed as a result of a demethylation at the methoxy. The 1H NMR spectrum of M4 showed an absence of a singlet at δ 3.70, consistent with the loss of the methyl of the 1576



Article

min, 11 min hold at 35% CH3CN, decrease to 15% CH3CN over 2 min, 3 min hold at 15% CH3CN. Ideal MS parameters were determined by infusion of each of the ceratamines. For metabolites of ceratamine A: IonSpray voltage was 5300 V, source temperature was 450 °C, declustering potential was 70 V, and focusing potential was 225 V. A collision energy of 35 (arbitrary units) was used for metabolites of ceratamine A. For metabolites of ceratamine B: IonSpray voltage was 5300 V, source temperature was 425 °C, declustering potential was 70 V, and focusing potential was 300 V. A collision energy of 35 (arbitrary units) was used for metabolites of ceratamine B. Isolation of metabolites from large-scale incubations was accomplished by coupling a Shimadzu FRC-10A fraction collector with the HPLC. A semipreparative ThermoFisher C18 column (150 mm × 4.6 mm, 5 μm particle diameter) was utilized for metabolite isolation. Mobile phase A was H2O containing 0.1% acetic acid. Mobile phase B was CH3CN. A total flow rate of 1.0 mL/min was used with a 1000 μL injection volume. UV analysis was performed using a wavelength of 350 nm. The mobile phase gradient used for the isolation of ceratamine B metabolites was as follows: 3 min hold at 15% CH3CN, increase to 19% CH3CN over 1 min, 5 min hold at 19% CH3CN, increase to 25% CH3CN over 1 min, 3 min hold at 30% CH3CN, increase to 38% CH3CN over 1 min, 3 min hold at 38% CH3CN, decrease to 15% CH3CN over 2 min, 3 min hold at 15% CH3CN. Time-based fraction collection was used to isolate the metabolites of interest: M3:10.80−13.30 min; M4:14.50−16.60 min, and M6:18.90−21.30 min. A delay volume of 164 μL was utilized. Ceratamines A and B were prepared by total synthesis. Sprague−Dawley rat liver microsomes were purchased from Invitrogen. β-Nicotinamide adenine dinucleotide phosphate, reduced tetra(cyclohexyl ammonium) salt was purchased from Sigma-Aldrich and Santa Cruz Biotechnology. Ammonium acetate was purchased from Sigma-Aldrich. All other chemicals were purchased from Fisher Scientific. Microsomal Incubations. Incubation mixtures contained a final drug concentration of 10 μM, a final microsomal protein concentration of 0.3 mg/mL, and a final concentration of NADPH of 1 mM in 100 mM ammonium acetate buffer, pH 7.4. Reactions were initiated by the addition of NADPH and were incubated in a water bath at 37 °C. Reactions were stopped by the addition of one volume of ice cold MeOH. Samples and controls were placed on ice for approximately 5 min and were then centrifuged for 10 min at 4500 rpm. The supernatant was collected and stored at −20 °C until LC−MS/ MS analysis. Two types of control incubations were also performed. These incubations contained microsomal protein previously inactivated by heating at 45 °C for 30 min or microsomal protein that had been boiled for 6 min. Accurate Mass. Accurate mass determinations were performed using a QSTAR Elite QqTOF mass spectrometer. The monoisotopic [M + H]+ ion signal from the corresponding parent compound was used as an accurate mass calibrant (ceratamine A: 466.9713 Da; ceratamine B: 452.9556 Da). Large-Scale Microsomal Incubations. Incubation mixtures contained a final drug concentration of 50 μM and a final microsomal protein concentration of 2 mg/mL in 100 mM ammonium acetate buffer, pH 7.4. Total incubation volume was 5 mL. Reactions were initiated by the addition of NADPH. Cofactor was added every 15 min, maintaining a final concentration of 1 mM. Reactions were incubated in a water bath at 37 °C for 1 h. Following incubation, reactions were

Figure 10. Proposed structures of two major phase I metabolites of ceratamine B formed by rat liver microsomes. Structure elucidation was accomplished via tandem mass spectrometry and UV/vis and 1H NMR spectroscopy.

methoxy. The addition of a broad singlet at δ 6.57 was attributed to the resulting phenol (Figure S7 in the SI).18 M6 was proposed to result from a demethylation of the aminoimidazole of ceratamine B. The 1H NMR spectrum revealed an absence of a doublet at δ 3.07 ppm, consistent with the loss of the methyl at the aminoimidazole. Also, the multiplicity of the signal at δ 8.67 has changed from a broad quartet to a broad singlet (Figure S8 in the SI). This change is due to the loss of any neighboring protons to the aminoimidazole, consistent with a demethylation. In conclusion, a total of eight metabolites were identified for the ceratamines. Preliminary structure elucidation, performed using tandem mass spectrometry, revealed that the majority of ceratamine metabolites are the result of various demethylation reactions. All proposed molecular formulas were supported by accurate mass. Monooxygenation was projected to occur in the formation of two metabolites, M1 and M3. MS/MS data suggested that the location of monooxygenation was along the aromatic ring. A hypsochromic shift in the B absorbance band of M3 revealed that the monooxygenation may result in a loss of aromaticity; however, this observation needs to be further investigated before a structure can be proposed. Finally, the proposed structures of two major metabolites, M4 and M6, were confirmed by 1H NMR spectroscopy (Figure 10).

■

EXPERIMENTAL SECTION General Experimental Procedures. UV/vis spectra were collected using a PerkinElmer Lambda 35 dual beam UV/vis spectrometer. Absorbance values were measured over a range of 200−700 nm. A slit width of 1 nm and a scan speed of 480 nm/ min were utilized. Water was used as the background reference. 1 H NMR spectra were collected using a Bruker 600 MHz NMR spectrometer equipped with a 5 mm cryoprobe. DMSO-d6 was used as a solvent for all NMR analysis. A Shimadzu Prominence LC equipped with a UV detector and coupled to a QSTAR Elite QqTOF mass spectrometer with a TurboIon spray source (Applied Biosystems/MDS Sciex) was used to identify all metabolites and obtain accurate mass data. A ThermoFisher C18 column (150 mm × 2.1 mm, 5 μm particle diameter) was used for analytical scale separation of metabolites. Mobile phase A was H2O containing 0.1% acetic acid. Mobile phase B was CH3CN. A total flow rate of 0.2 mL/min was used with a 25 μL injection volume. UV analysis was performed using a wavelength of 350 nm. The mobile phase gradient used for the separation of ceratamine metabolites was as follows: 3 min hold at 15% CH3CN, increase to 17% CH3CN over 1 min, 18 min hold at 17% CH3CN, increase to 27% CH3CN over 1 min, 4 min hold at 27% CH3CN, increase to 35% CH3CN over 2 1577



Article

stopped by the addition of one volume of ice cold MeOH. Samples were placed on ice for approximately 5 min and were then centrifuged for 20 min at 4500 rpm. The supernatant was collected and stored at −20 °C until isolation. A total of eight large-scale reactions were performed.

■

ASSOCIATED CONTENT

S Supporting Information *

This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Tel: 1-724-575-0139, E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the University of Pittsburgh, Department of Pharmaceutical Sciences. The authors would like to acknowledge T. Miller and P. Oberly for their technical assistance with the QStar Elite QqTOF mass spectrometer.

■

REFERENCES

(1) Molinski, T.; Dalisay, D.; Lievens, S.; Saludes, J. Nat. Rev. 2009, 8, 69−85. (2) Cragg, G. Med. Res. Rev. 1998, 18, 315−331. (3) Georg, G.; Chen, T.; Ojima, I.; Vyas, D. ACS Symp. Ser. 1994, 583, 1−3. (4) Rowinsky, E. Annu. Rev. Med. 1997, 48, 353−374. (5) Cooper, G. The Cell: A Molecular Approach, 2nd ed., Sinauer Associates: Sunderland, MA, 2000. (6) Jordan, M.; Wilson, L. Nat. Rev. Cancer 2004, 4, 253−265. (7) Wittman, T.; Hyman, A.; Desai, A. Nat. Cell Biol. 2001, 3, 28−34. (8) Kirschner, M.; Mitchison, T. Cell 1986, 45, 329−342. (9) Stanton, R.; Gernert, K.; Nettles, J.; Aneja, R. Med. Res. Rev. 2011, 31, 443−481. (10) Roberge, M.; Cinel, B.; Andersen, H.; Lim, L.; Jiang, X.; Xu, L.; Bigg, C.; Kelly, M.; Andersen, R. Cancer Res. 2000, 60, 5052−5058. (11) Galletti, M.; Maghani, M.; Renzulli, M.; Botta, M. ChemMedChem. 2007, 2, 920−942. (12) Martello, L.; McDaid, H.; Regl, D.; Yang, C.; Meng, D.; Pettus, T.; Kaufman, M.; Arimoto, H.; Danishefsky, S.; Smith, A., III; Horwitz, S. Clin. Cancer Res. 2000, 6, 1978−1987. (13) Karjala, G.; Can, Q.; Manzo, E.; Andersen, R.; Roberge, M. Cancer Res. 2005, 65, 3040−3043. (14) ter Haar, E.; Kowalski, R.; Hamel, E.; Lin, C.; Longley, R.; Gunasekera, S.; Rosenkranz, H.; Day, B. Biochemistry 1996, 35, 243− 250. (15) Coleman, R.; Campbell, E.; Carper, D. Org. Lett. 2009, 11, 2133−2136. (16) Ballatore, C.; Brunden, K.; Huryn, D.; Trojanowski, J.; Lee, V.; Smith, A. J. Med. Chem. 2012, 55, 8979−8996. (17) Coleman, M. Human Drug Metabolism: An Introduction; WileyBlackwell: Hoboken, NJ, 2010. (18) Pasto, D.; Johnson, C. Organic Structure Determination; PrenticeHall, Inc.: Englewood Cliffs, NJ, 1969; pp 83−108.

1578


Structure Elucidation of phase I Metabolites of the Microtubule

Recommend Documents