Stereospecificity of retinol saturase: Absolute configuration, synthesis, and biological evaluation of dihydroretinoids

Article abstract of DOI:10.1021/ja710487q

Retinol saturase carries out a stereospecific saturation of the C13-C14 double bond of all-trans-retinol to generate (13R)-all-trans-13,14-dihydroretinol. This compound is found in cells expressing mouse or zebrafish retinol saturase and in the livers of mice fed retinyl palmitate. All-trans-13,14-dihydroretinol is oxidized in vivo to all-trans-13,14-dihydroretinoic acid, a highly selective agonist of the retinoic acid receptor. The naturally occurring (13R)-all-trans-13,14-dihydroretinoic acid is a weaker agonist than the (13S) enantiomer, indicating enantioselective recognition by the ligand-binding pocket of this receptor. Consequently the (13S) enantiomer, acting through the retinoic acid receptor, also inhibits adipose differentiation more potently than the (13R) enantiomer. Copyright

Full text of DOI:10.1021/ja710487q

Published on Web 01/08/2008
Stereospecificity of Retinol Saturase: Absolute Configuration, Synthesis, and
Biological Evaluation of Dihydroretinoids
Alexander R. Moise,*,† Marta Dom´ınguez,‡ Susana Alvarez,‡ Rosana Alvarez,‡ Michael Schupp,§
Ana G. Cristancho,‡ Philip D. Kiser,† Angel R. de Lera,‡ Mitchell A. Lazar,§ and
Krzysztof Palczewski†
Department of Pharmacology, Case School of Medicine, Case Western ReserVe UniVersity,
CleVeland, Ohio 44106-4965, Departamento de Qu´ımica Orga´nica, Facultad de Qu´ımica, UniVersidade de Vigo,
36310 Vigo, Spain, and DiVision of Endocrinology, Diabetes, and Metabolism, Department of Medicine and
Department of Genetics, and The Institute for Diabetes, Obesity, and Metabolism, UniVersity of PennsylVania
School of Medicine, Philadelphia, PennsylVania 19104
Received November 20, 2007; E-mail: ram50@case.edu
Mouse RetSat catalyzes the saturation of the C13-C14 double
bond of all-trans-retinol to produce all-trans-13,14-dihydroretinol
81 (Figure 1). A related enzyme in zebrafish catalyzes the saturation
of the C7-C8 double bond in addition to the C13-C14 double
bond of all-trans-retinol to produce both 8 and all-trans-7,8-
dihydroretinol.2 Further oxidation of 8 and of all-trans-7,8-
dihydroretinol by retinol dehydrogenases and then by retinaldehyde
dehydrogenase enzymes leads to formation of all-trans-13,14-
dihydroretinoic acid 9 and all-trans-7,8-dihydroretinoic acid, com-
pounds whose levels are exquisitely controlled in vivo by the
enzymes that catalyze their synthesis and breakdown.2,3 Both 9 and
all-trans-7,8-dihydroretinoic acid are highly selective agonists in
activating the retinoic acid receptor (RAR) but not the retinoid X
receptor (RXR).3,4 Because 13,14-dihydroretinoids are chiral com-
pounds it is important to determine their absolute configuration and
evaluate how the different enantiomers interact with binding
proteins, receptors, and enzymes. In this study we investigated the
absolute configuration of biologically derived 8 and consequently
of 9 and evaluated the activation of RAR by the enantiomers of 9.
To establish the absolute configuration of 8 we examined the
products of mouse and zebrafish RetSat by chiral HPLC. We also
examined the endogenous form of compound 8 purified from livers
of mice gavaged with all-trans-retinyl palmitate. For our analyses
by chiral HPLC we established authentic standards by stereospecific
syntheses of the two enantiomers of 8 (Scheme 1 and Supporting
Information Scheme S-2). The synthetic scheme leading to the
biological material (R)-all-trans-13,14-dihydroretinol (R)-8 is de-
picted in Scheme 1. The chirality was transferred from that of the
γ-alkoxy group in (4S)-4,5-(O-isopropylidene)pent-2-enoate enan-
tiomer (4S)-1,5 and Suzuki coupling was selected as the connective
method to construct the polyene skeleton.6 An analogous sequence
produced (S)-all-trans-13,14-dihydroretinol (S)-8 from (4R)-4,5-(O-
isopropylidene)pent-2-enoate (4R)-1. Determination of the enan-
tiomeric excess of the target compounds was based on HPLC
separation by a Chiracel OD-H 0.46 cm × 15 cm column that
afforded an enantiomeric excess (ee) value of >96% for each
enantiomer of 8.
Figure 1. Chiral HPLC analysis of compound 8 product of mouse RetSat
and of compound 8 purified from mouse liver in comparison with synthetic
(R) and (S)-8 standards. Compounds were analyzed separately or in
combination as indicated. Blue line chromatograms depict the chiral HPLC
analysis of the synthetic standards.
Scheme 1. Preparation of (R)-All-trans-13,14-dihydroretinol,
(R)-8a
a Reagents and reaction conditions: (a) MeLi, THF, -78 °C. (b) H5IO6,
1:1 THF/Et2O, 25 °C, 14 h, 68%. (c) CrCl2, THF, 2-(dichloromethyl)-
4,4,5,5-tetramethyl-1,3,2-dioxaborolane 5, LiI, 25 °C, 14 h, 68%. (d) Iodide
6, Pd(PPh3)4, THF, 10% aq TlOH, 25 °C, 2 h, 55%. (e) DIBAL-H, THF,
-78 °C, 2 h, 73%.
(Figure 1 and Figure S-1). Therefore, both mouse and zebrafish
RetSat have the same stereospecificity at the C13-C14 double
bond. We also examined the endogenous form of compound 8
purified from the livers of mice gavaged with all-trans-retinyl
palmitate. Chiral HPLC analysis of compound 8 found in vivo
indicates it is predominantly (R)-8 (Figure 1). A smaller peak that
co-migrates with (S)-8 may represent a racemization product of
(R)-8 or a saturation product of all-trans-retinol that occurs by a
pathway independent of RetSat (Figure 1, marked by asterisk). We
employed a second chromatographic chiral separation method by
We analyzed compound 8 purified from previously described
human embryonic kidney cells (HEK) 293 cells that express mouse
or zebrafish RetSat (HEK-mRetSat and HEK-zRetSat, respec-
tively).1,2 This analysis showed that both mouse and zebrafish RetSat
produce the (R)-8 enantiomer as it had chromatographic properties
on chiral HPLC identical to synthetic (R)-8 but distinct from (S)-8
† Case Western Reserve University.
‡ Universidade de Vigo.
§ University of Pennsylvania School of Medicine.
9
1154
J. AM. CHEM. SOC. 2008, 130, 1154-1155
10.1021/ja710487q CCC: $40.75 © 2008 American Chemical Society

C O M M U N I C A T I O N S
stained with Oil Red-O as well as by the expression of adipocytic
markers FABP4 and PPARγ (Figure 2, parts C and D). The
inhibition of adipogenesis by a racemic mixture of 8 can be relieved
by the addition of AGN193618, a specific antagonist of RAR.9
Therefore, 8 inhibits adipogenesis through activation of RAR. The
human RARγ ligand binding pocket can discriminate between the
active agonist BMS270394 and its inactive enantiomer BMS270395.10
On the basis of the crystal structure of RARγ bound to each
enantiomer, this selectivity was proposed to result from the
unfavorable contacts between the ligand and the ligand binding
pocket. In the case of the (R) and (S) enantiomers of 9 we believe
a similar mechanism may occur where (R)-9 interacts unfavorably
with several residues, that may include those that interact with
carboxylate (Figure S-5). Further crystallographic studies could shed
light on the interactions of the (R) and (S) enantiomers of 9 with
RAR.
We present data that establishes the stereospecificity of RetSat
and the absolute configuration of its product as (R)-8. The (R)
configuration renders dihydroretinoids weaker agonists of RAR than
the (S) enantiomers. These studies should help clarify the physi-
ological function of RetSat as its expression in adipose tissue would
result in conversion of an inhibitor of adipose differentiation, all-
trans-retinol, into a much weaker inhibitor of differentiation, that
is, (R)-8. Our findings are also relevant to understanding the
interaction of RAR with chiral agonists.
Figure 2. (A,B) Stereospecificity of RAR activation by (R)- and (S)-13,-
14-dihydroretinoids. Activation of RAR by enantiomers of 8 or 9 was
evaluated by assaying the level of â-galactosidase whose expression is
controlled by a retinoic acid response element. (C,D) Effect of enantiomers
of 8 (1 µM final concentration) on adipocyte differentiation (left panels).
Inhibition of adipocyte differentiation by a racemic mixture of 8 (500 nM)
was relieved by the RARR antagonist AGN193618 (2 µM) (right panels).
Accumulation of neutral lipids was evaluated by Oil Red-O staining and
differentiated cells were observed by phase contrast (C, bottom panels).
Expression of adipocyte specific markers (FABP4 PPARγ) was evaluated
by immunoblotting and compared to the loading control protein RAN (D).
Acknowledgment. This work was supported by NIH Grant R01
R0109339 (to K.P.), SAF Grant 2007-63880-FEDER (to A.R.de
L.), NIH Grant R01 DK49780 (to M.A.L.) M.S. was supported by
a Mentored Fellowship award from the American Diabetes As-
sociation. We thank B. Norman and W. Chin (Lilly) for providing
AGN193618 and for valuable comments. We also thank Leslie
Webster for valuable comments and Nathan Cobb and Witold
Surewicz for their help in recording the CD spectra of dihydro-
retinoids.
derivatizing the synthetic compounds (R)-8 and (S)-8 and the
compound 8 purified from HEK-mRetSat with (S)-(+)-R-methoxy-
R-trifluoromethylphenylacetyl (MTPA) chloride to generate the
MTPA esters. Following separation of the MTPA esters of 8 by
normal phase HPLC we found that the MTPA ester of purified
endogenous 8 co-migrates with the MTPA ester of (R)-8 but not
with the MTPA ester of (S)-8 (Figure S-2).
Supporting Information Available: Experimental procedures,
analysis of the stereospecificity of zebrafish RetSat, chiral separation
of (R)- and (S)-8 after derivatization with MTPA, preparation of (R)-
and (S)-8, chromatographic separation and CD analysis of (R)- and
(S)-7, CD analysis of (R)- and (S)-8, structural modeling of (R)- and
(S)-9 inside RARγ. This material is available free of charge via the
Internet at http://pubs.acs.org.
Previous studies have shown that a racemic mixture of compound
9 can activate RAR but not RXR.3 In this study we investigated
the activation of RAR by the enantiomers of 9 with a cell-based
RAR transactivation assay.7 We found that (S)-9 is more potent in
activating RAR than (R)-9, yet both activate RAR less than all-
trans-retinoic acid at physiological (nanomolar) concentrations
(Figure 2A). Surprisingly, at supraphysiological concentrations (S)-9
is a more potent activator of RAR than all-trans-retinoic acid.
Similar results were obtained with enantiomers of 8 which are
converted by cells to the corresponding acids. In these experiments,
the RAR activation potency of compounds is (S)-8 > all-trans-
retinol > (R)-8 (Figure 2B). Therefore, endogenous dihydroretinoids
in the (R) configuration are the least active RAR agonists compared
to the (S) enantiomers or all-trans-retinoic acid. All-trans-retinoic
acid plays an important role in adipose differentiation acting through
the RAR to prevent differentiation.8 RetSat is an enzyme important
in adipose differentiation (Schupp et al., unpublished) so we
investigated the role of dihydroretinoids in adipose differentiation.
We found that while all retinoids inhibit differentiation, (R)-8
inhibits adipose differentiation much less than (S)-8 or all-trans-
retinol. This was evidenced by the accumulation of neutral lipids
References
(1) Moise, A. R.; Kuksa, V.; Imanishi, Y.; Palczewski, K. J. Biol. Chem.
2004, 279, 50230-50242.
(2) Moise, A. R.; Isken, A.; Dominguez, M.; de Lera, A. R.; von Lintig, J.;
Palczewski, K. Biochemistry 2007, 46, 1811-1820.
(3) Moise, A. R.; Kuksa, V.; Blaner, W. S.; Baehr, W.; Palczewski, K. J.
Biol. Chem. 2005, 280, 27815-27825.
(4) LeMotte, P. K.; Keidel, S.; Apfel, C. M. Biochim. Biophys. Acta 1996,
1289, 298-304.
(5) Leonard, J.; Mohialdin, S.; Reed, D.; Ryan, G.; Swain, P. A. Tetrahedron
1995, 51, 12843-12858.
(6) Torrado, A.; Iglesias, B.; Lo´pez, S.; de Lera, A. R. Tetrahedron 1995,
51, 2435-2454.
(7) Wagner, M.; Han, B.; Jessell, T. M. DeVelopment 1992, 116, 55-66.
(8) Kuriharcuch, W. Differentiation 1982, 23, 164-169. Chawla, A.; Lazar,
M. A. J. Biol. Chem. 1993, 268, 16265-16269. Xue, J. C.; Schwarz, E.
J.; Chawla, A.; Lazar, M. A. Mol. Cell. Biol. 1996, 16, 1567-1575. Bonet,
M. L.; Ribot, J.; Felipe, F.; Palou, A. Cell. Mol. Life Sci. 2003, 60, 1311-
1321.
(9) Teng, M.; Duong, T. T.; Johnson, A. T.; Klein, E. S.; Wang, L.; Khalifa,
B.; Chandraratna, R. A. J. Med. Chem. 1997, 40, 2445-2451.
(10) Klaholz, B. P.; Mitschler, A.; Belema, M.; Zusi, C.; Moras, D. Proc. Natl.
Acad. Sci. U.S.A. 2000, 97, 6322-6327.
JA710487Q
9
J. AM. CHEM. SOC. VOL. 130, NO. 4, 2008 1155