QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: dmd.108.020651v1 36/8/1457    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Zollinger, M. Articles by Sedrani, R. Search for Related Content PubMed PubMed Citation Articles by Zollinger, M. Articles by Sedrani, R.

SHORT COMMUNICATION

The Macrolide Everolimus Forms an Unusual Metabolite in Animals and Humans: Identification of a Phosphocholine Ester

Novartis Pharma AG, Basel, Switzerland (M.Z., C.S., R.D.); and Novartis Institutes for BioMedical Research, Basel, Switzerland (W.S., R.S.)

(Received January 28, 2008; Accepted April 24, 2008)

	Abstract

Top Abstract Materials and Methods Results and Discussion References

 
The immunosuppressant macrolide everolimus was found to be metabolized in animals and humans to a phosphocholine ester (ATG181), a hitherto unknown type of conjugate in xenobiotic metabolism. The structure of ATG181 was elucidated by mass spectrometry and confirmed by synthesis. ATG181 was among the most prominent metabolites of everolimus in rat, monkey, and human blood and was found also in various tissues of the rat, whereas no ATG181 was identified in the urine and feces of the species investigated. The metabolite showed binding to FK506 binding protein with a 2- to 3-fold higher affinity than everolimus. However, ATG181 exhibited only marginal in vitro immunosuppressive activity and is therefore very unlikely to contribute in a relevant manner to the immunosuppressive effect of everolimus.

	Materials and Methods

Top Abstract Materials and Methods Results and Discussion References

 
Test Compound and Chemicals. Unlabeled, ³H-labeled, and ¹⁴C-labeled everolimus (labeling positions given in Fig. 1A) were synthesized by Novartis (Basel, Switzerland). Reagents and solvents were of analytical grade and obtained from commercial sources.

View larger version (11K): [in this window] [in a new window]   FIG. 1. Chemical structure of everolimus and its metabolite ATG181. Positions of 3H- and 14C-labeling indicated by number signs and asterisks, respectively.

Isolation of Metabolite ATG181. For structure elucidation, the ¹⁴C-labeled metabolite ATG181 was isolated from rat kidney homogenate (pool across three animals). The homogenate was buffered with phosphate, pH 6.5, and extracted with methanol. The extract was subjected to high-performance liquid chromatography on a reversed-phase column with gradient elution. Fractions of the column effluent containing [¹⁴C]ATG181 were combined, providing approximately 900 pmol of [¹⁴C]ATG181 for analysis by liquid chromatography/mass spectrometry (LC/MS) and liquid chromatography/tandem mass spectrometry (LC/MS/MS). A smaller amount of [¹⁴C]ATG181 was isolated in a similar manner from rat blood (pool across three animals). Moreover, [¹⁴C]ATG181 was isolated from the blood (2 h) of a male stable renal transplant patient treated with a single p.o. dose of 3 mg of [¹⁴C]everolimus. This aspect of the study was performed in accordance with Good Clinical Practice guidelines and the Declaration of Helsinki (1964 and subsequent revisions). [¹⁴C]ATG181 from rat and human blood was analyzed by LC/MS to show its identity with the metabolite in rat kidney.

Mass Spectrometry. The LC/MS and LC/MS/MS analyses were performed on a Micromass (Wythenshawe, UK) Q-Tof 2 instrument, equipped with an electrospray interface. For the chromatography, a 150 x 1.0-mm Symmetry C18 column (3.5-µm particles; Waters, Milford, MA) was used. The column was thermostated at 60°C. The components were eluted with a gradient of 10 mM ammonium acetate in water, pH 6.5, versus methanol at a total flow rate of 60 µl/min. For hydrogen-deuterium exchange experiments, the water in the mobile phase was replaced by D₂O (Cambridge Isotope Laboratories, Andover, MA; 99.9 atom-% D), and the methanol was replaced by CH₃OD (Cambridge Isotope Laboratories; 99 atom-% D). After the column, the effluent was split in a ratio of approximately 1:5. The smaller part was directed into the electrospray interface. The rest was passed through a diode-array detector and then either collected in fractions for off-line radioactivity monitoring, or discarded. The electrospray interface was operated with nitrogen as nebulizer gas (7 bar), as desolvation gas (350 l/h), and as cone gas (40 l/h). The desolvation temperature was 150°C; the source block temperature was 80°C; and the spray capillary was set to 3.0 kV. Both mass spectra with single-stage mass separation and tandem mass spectra were recorded in the positive ion mode. Collisional activation was performed with argon (6–7 · 10^-5 mbar, measured in the housing outside the collision cell) at a collision offset of 50 V (everolimus) or 60 V (ATG181). The quadrupole mass analyzer was set to unit mass resolution, and the time-of-flight mass analyzer was operated at a resolution of approximately 8000. The [M+H]⁺ ion of reserpine at m/z 609.2812 was used as lock mass for recalibrating the spectra to obtain exact mass data.

Synthesis of ATG181. ATG181 was prepared by a three-step, one-pot synthesis adapted from a literature precedent (Ishihara and Sano, 1996). Briefly, everolimus was converted to its dichlorophosphate using phosphorus oxychloride and triethylamine as base catalyst, further converted to the chlorophosphocholine tosylate with choline tosylate in pyridine, and finally hydrolyzed with water to the phosphocholine ester ATG181. The product was purified by solid-phase extraction, crystallization, and semipreparative high-performance liquid chromatography and characterized by mass spectrometry and NMR spectroscopy.

Assay for Binding to FK506 Binding Protein. The measurements were performed using a microtiter plate-based competitive binding assay as described previously (Schuler et al., 1997). Briefly, binding of biotinylated recombinant human FK506 binding protein (FKBP-12) to immobilized FK506 was measured in the absence (as a control) or presence of serial dilutions of the test compound, and the concentration resulting in 50% inhibition of FKBP-12 binding to the immobilized FK506 (IC₅₀ value) was calculated.

Mixed Lymphocyte Reaction. The mixed lymphocyte reaction (MLR) experiments (two-way MLR) were set up as described (Schuler et al., 1997), mixing equal numbers of spleen cells (10⁵) from two genetically different mouse strains, i.e., CBA (H-2^k) and BALB/c mice (H-2^d) (BRL, Füllinsdorf, Switzerland). The cells were cultivated for 4 days in the absence or presence of appropriate serial dilutions of the test compound; cell proliferation was then determined by measuring [³H]thymidine incorporation into DNA following an additional 16-h incubation period with [³H]thymidine.

	Results and Discussion

Top Abstract Materials and Methods Results and Discussion References

 
Structure Elucidation of Metabolite ATG181 by Mass Spectrometry. For the elucidation of its structure, ATG181 was isolated from the kidney of rats treated with a single p.o. dose of 15 mg/kg [¹⁴C]everolimus (structure and labeling positions shown in Fig. 1A). The isolated metabolite was investigated by mass spectrometry. Kidney was chosen as the source of the metabolite because a previous study in the rat had shown highest concentrations of ATG181 in this organ among a number of tissues, including blood (unpublished results). From the mass spectral data, the structure of ATG181 shown in Fig. 1B was derived in the following way.

An ion of the intact molecule of ATG181 was observed at m/z 1123 (nominal mass) showing the same isotope pattern (as a result of the ¹⁴C-labeling) as the [M+Na]⁺ ion of the parent compound everolimus (insets in Fig. 2). Even though electrospray ionization of everolimus in the positive ion mode produced almost exclusively the [M+Na]⁺ ion (m/z 980), the m/z 1123 ion of ATG181 proved to be the protonated molecule ([M+H]⁺, M referring to the neutral zwitterion). This was shown by adding potassium (50 mM) to the mobile phase, which resulted in the formation of an ion at m/z 1161 (m/z 1123-H+K; data not shown), whereas no ions at m/z 1139 (m/z 1123-Na+K) or m/z 1144 (m/z 1123-NH₄+K) appeared. Hence, the ion at m/z 1123 was neither of the [M+Na]⁺ nor of the [M+NH₄]⁺ type. The odd-numbered difference in nominal mass between the neutral molecules of everolimus and ATG181 (165 mass units) indicated that ATG181 was a conjugate of everolimus with an entity containing an odd number of nitrogens. Conjugation with an amino acid could be excluded based on mass spectrometric H/D exchange experiments, showing only three exchangeable hydrogens in protonated ATG181, as found for sodiated everolimus (data not shown). These findings and the observation of mass differences of 59 (trimethylamine) in the product ion spectrum of Fragment a of ATG181 (Fig. 2C), compatible with the presence of a trimethylammonium group, suggested that ATG181 was a conjugate with phosphocholine. Exact mass measurements were in agreement with this proposal (cf. legend of Fig. 2). The fragment ions of protonated ATG181 at m/z 532 and 829 (Fig. 2B) appeared to correspond to the Fragments a and C of everolimus at m/z 389 and 686 (Boernsen et al., 2007), differing from their precursor ions by 591 and 294 mass units, respectively. These two fragment ions and a few ions in the low mass region, all containing the trimethylammonium function, were the only ones observed in the product ion spectrum of protonated ATG181, in contrast to the product ion spectrum of sodiated everolimus, which showed fragment ions containing all parts of the molecule (Fig. 2A). This difference between the two compounds is likely because of a strong charge-localizing effect by the quaternary ammonium function in ATG181. Based on the Fragments a and C and the low-mass fragments of protonated ATG181, especially the one at m/z 228, the phosphocholine group could be localized at the terminal hydroxy group of the side chain of everolimus. In addition, Fragment C of ATG181 indicated that the lactone macrocycle was still intact. The UV spectra of ATG181 and everolimus, which were monitored in parallel with the mass spectra, were essentially identical, showing partially resolved maxima at 268, 277, and 288 nm. This confirmed that the chromophore (the triene group) of everolimus was not affected by the biotransformation to ATG181. Hence, ATG181 was identified as a phosphocholine ester of everolimus with the 2-hydroxyethoxy side chain as the site of modification.

View larger version (22K): [in this window] [in a new window]   FIG. 2. Product ion mass spectra of sodiated everolimus (A), protonated ATG181 (B), and Fragment a of ATG181 (C), obtained by LC/MS/MS. Collisional activation with argon at collision offset voltages of 50 V (sodiated everolimus), 60 V (protonated ATG181), or 35 V (Fragment a of ATG181). Differences between measured and calculated masses <5 mDa for all the signals labeled with four digits after the decimal point. Insets: intact molecule regions of the single-stage mass spectra of [14C]everolimus and [14C]ATG181, respectively, showing characteristic isotope patterns as a result of the 14C-labeling.

Occurrence and Abundance of Metabolite ATG181. [¹⁴C]-ATG181 isolated from rat and human blood after treatment with a single p.o. dose of [¹⁴C]everolimus (15 mg/kg and 3 mg, respectively) showed the same retention time and electrospray mass spectrum as ATG181 isolated from rat kidney. This showed the identity of the metabolite in the three matrices.

In addition to rat blood and kidney and human blood, ATG181 was identified (based on retention time) also in the blood of mice and cynomolgus monkeys after a single p.o. dose of [³H]everolimus (0.9 and 5 mg/kg, respectively) and in various tissues of rats after a single and 21 daily p.o. doses of 0.5 mg/kg [³H]everolimus. In human blood, ATG181 was among the most prominent metabolites, accounting for 6% of the area under the curve (AUC)_{0–24 h} of total radiolabeled material. Other important metabolites were the seco acids, formed by opening of the macrocycle at the lactone function either by hydrolysis or β-elimination (4 and 7% of AUC, respectively), and several hydroxylated metabolites (most abundant one, 13% of AUC). Parent drug accounted for 40% of the AUC. In rat and monkey blood, ATG181 was also among the most abundant metabolites (6–7% of the AUC_{0–24 h} of total radiolabeled material). Only in mouse blood, ATG181 was less prominent (3% of AUC_{0–24 h} of total radiolabeled material). In rats, at 24 h after the 21st daily dose of [³H]everolimus, ATG181 was detected in all the tissues analyzed, accounting for 2.5% of radioactivity in the liver, 18% in the kidney, 6% in the stomach, 5% in the duodenum, 8% in the heart, and 7% in the lung. However, no ATG181 was identified in the urine or feces of the species investigated.

Metabolic Pathway to ATG181. To our knowledge, phosphocholine conjugates of xenobiotics have not been described so far, except for very close analogs of endogenous compounds like fluorescence-labeled endogenous lipids (e.g., Kok et al., 1997) or an unnatural stereoisomer of dihydroceramide (Dragusin et al., 2003). In contrast, phosphocholine esters are very common as endogenous lipids, e.g., in the form of sphingomyelins or phosphatidylcholines (lecithins). The latter are formed, among other pathways, by the coupling of diacylglycerol with cytidine diphosphate choline, an activated form of phosphocholine. In humans, this reaction is catalyzed by cholinephosphotransferase, found in the Golgi apparatus, as well as by choline/ethanolaminephosphotransferase, found in both the endoplasmic reticulum and nuclear membranes (Henneberry et al., 2002). It is conceivable that the same enzymes are involved in the conversion of everolimus to the metabolite ATG181. However, this was not investigated. Although the metabolism of endogenous compounds is normally catalyzed by enzymes with a relatively narrow substrate specificity, in contrast to enzymes typically involved in drug metabolism, occasionally xenobiotics are sufficiently similar to endogenous compounds to undergo the same pathways as their endogenous counterparts. One of the best known examples is the β-oxidation of xenobiotic carboxylic acids with long alkyl chains, resembling endogenous fatty acids. In the case of everolimus, however, hardly any structural similarity to 1,2-diacylglycerols or ceramides, the precursors of sphingomyelins, seems to exist, rendering the formation of ATG181 quite unexpected.

Pharmacological Activity Data. The macrolide rapamycin and rapamycin derivatives, like everolimus, are not immunosuppressive per se. To be active they need to form a complex with an abundant intracellular binding protein, FKBP-12; this complex in turn binds to and inhibits the actual target protein, mammalian target of rapamycin (reviewed, for example, in Abraham and Wiederrecht, 1996). Therefore, binding to FKBP-12 is a necessary albeit not sufficient prerequisite for immunosuppressive activity of these macrolides (Dumont et al., 1990). ATG181 was found to bind 2- to 3-fold stronger to FKBP-12 than everolimus, with IC₅₀ values of 2.0 ± 0.7 and 5.3 ± 1.2 nM, respectively (mean ± S.D. of n = 3). However, the ability of ATG181 to inhibit a T-cell immune response, as measured in vitro using the mouse MLR assay, was at least 100-fold lower than that of everolimus; for ATG181 an IC₅₀ value of 108 ± 22 nM (mean ± S.D. of n = 6) was determined, whereas everolimus showed IC₅₀ values of 0.8 and 1.0 nM in two independent experiments where ATG181 and everolimus were tested side-by-side. The loss of activity of ATG181 may be because of its inability to efficiently permeate into cells as the compound is zwitterionic and very polar. Alternatively, the phosphocholine group, although not interfering with FKBP-12 binding, may prevent efficient binding of the ATG181/FKBP-12 complex to the mammalian target of rapamycin by steric hindrance or electronic repulsion as a result of the charged phosphate and/or ammonium groups. One of these effects, or the combination of both, could explain the very low immunosuppressive activity of ATG181. In any event, binding of a compound to FKBP-12 (a protein with peptidyl-prolyl isomerase activity) per se has apparently no biological consequences except for antagonizing the immunosuppressive activity of FK506, rapamycin, and everolimus; no other biological effects have been found so far for the known nonimmunosuppressive FKBP-12–binding molecules (Dumont et al., 1992; Fehr et al., 1996).

In conclusion, although ATG181 is one of the major metabolites of everolimus in human blood, its low activity, compared with everolimus, makes it very unlikely that ATG181 contributes in a relevant manner to immunosuppression of patients treated with everolimus.

	Acknowledgments

 
We thank Jean-Pierre Baldeck for the metabolite profiling and isolation work and Julien France for recording and interpreting the NMR spectra, Catherine Wioland for performing the pharmacological assessments, and Dr. Alain Schweitzer for his support of the animal experiments.

	Footnotes

 
Part of this work has been presented at the 53rd meeting of the American Society for Mass Spectrometry Conference on Mass Spectrometry and Allied Topics, San Antonio, TX, June 5–9, 2005.

Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.

doi:10.1124/dmd.108.020651.

ABBREVIATIONS: LC/MS, liquid chromatography/mass spectrometry; LC/MS/MS, liquid chromatography/tandem mass spectrometry; FKBP-12, FK506 binding protein; MLR, mixed lymphocyte reaction; AUC, area under the curve.

The online version of this article (available at http://dmd.aspetjournals.org) contains supplemental material.

¹ Current affiliation: Theragenomics Associates, Grellingen, Switzerland.

Address correspondence to: Markus Zollinger, Novartis Pharma AG, WKL-135.2.21 P.O. Box, CH-4002 Basel, Switzerland. E-mail: markus.zollinger{at}novartis.com

	References

Top Abstract Materials and Methods Results and Discussion References

 


Abraham RT and Wiederrecht GJ (1996) Immunopharmacology of rapamycin. Annu Rev Immunol 14: 483-510.[CrossRef][Medline]

Boernsen KO, Egge-Jacobsen W, Inverardi B, Strom T, Streit F, Schiebel H-M, Benet LZ, and Christians U (2007) Assessment and validation of the MS/MS fragmentation patterns of the macrolide immunosuppressant everolimus. J Mass Spectrom 42: 793-802.[CrossRef][Medline]

Dragusin M, Gurgui C, Schwarzmann G, Hoernschemeyer J, and van Echten-Deckert G (2003) Metabolism of the unnatural anticancer lipid safingol, L-threo-dihydrosphingosine, in cultured cells. J Lipid Res 44: 1772-1779.[Abstract/Free Full Text]

Dumont FJ, Melino MR, Staruch MJ, Koprak SL, Fischer PA, and Sigal NH (1990) The immunosuppressive macrolides FK-506 and rapamycin act as reciprocal antagonists in murine T cells. J Immunol 144: 1418-1424.[Abstract]

Dumont FJ, Staruch MJ, Koprak SL, Siekierka JJ, Lin CS, Harrison R, Sewell T, Kindt VM, Beattie TR, Wyvratt M, et al. (1992) The immunosuppressive and toxic effects of FK-506 are mechanistically related: pharmacology of a novel antagonist of FK-506 and rapamycin. J Exp Med 176: 751-760.[Abstract/Free Full Text]

Dunn C and Croom KF (2006) Everolimus: a review of its use in renal and cardiac transplantation. Drugs 66: 547-570.[CrossRef][Medline]

Fehr T, Sanglier JJ, Schuler W, Gschwind L, Ponelle M, Schilling W, and Wioland C (1996) Antascomicins A, B, C, D and E. Novel FKBP12 binding compounds from a Micromonospora strain. J Antibiot (Tokyo) 49: 230-233.[Medline]

Henneberry AL, Wright MM, and McMaster CR (2002) The major sites of cellular phospholipid synthesis and molecular determinants of fatty acid and lipid head group specificity. Mol Biol Cell 13: 3148-3161.[Abstract/Free Full Text]

Ishihara M and Sano A (1996) Facile and useful synthesis of enantiomeric phosphatidylcholines. Chem Pharm Bull 44: 1096-1098.

Kirchner GI, Winkler M, Mueller L, Vidal C, Jacobsen W, Franzke A, Wagner S, Blick S, Manns MP, and Sewing KF (2000) Pharmacokinetics of SDZ RAD and cyclosporin including their metabolites in seven kidney graft patients after the first dose of SDZ RAD. Br J Clin Pharmacol 50: 449-454.[CrossRef][Medline]

Kok JW, Nikolova-Karakashian M, Klappe K, Alexander C, and Merrill AH Jr (1997) Dihydroceramide biology. Structure-specific metabolism and intracellular localization. J Biol Chem 272: 21128-21136.[Abstract/Free Full Text]

Schuler W, Sedrani R, Cottens S, Häberlin B, Schulz M, Schuurman HJ, Zenke G, Zerwes HG, and Schreier MH (1997) SDZ RAD, a new rapamycin derivative: pharmacological properties in vitro and in vivo. Transplantation 64: 36-42.[CrossRef][Medline]

Vidal C, Kirchner GI, and Sewing KF (1998) Structural elucidation by electrospray mass spectrometry: an approach to the in vitro metabolism of the macrolide immunosuppressant SDZ RAD. J Am Soc Mass Spectrom 9: 1267-1274.[CrossRef][Medline]

This Article Abstract Full Text (PDF) All Versions of this Article: dmd.108.020651v1 36/8/1457    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Zollinger, M. Articles by Sedrani, R. Search for Related Content PubMed PubMed Citation Articles by Zollinger, M. Articles by Sedrani, R.