Potent inhibition of retinoic acid metabolism enzyme(s) by novel azolyl retinoids

Article abstract of DOI:10.1016/S0960-894X(00)00391-7

Novel (±)-4-azolyl retinoic acid analogues 4, 5, 7 and 8 have been designed and synthesized and have been shown to be powerful inhibitors of hamster microsomal all-trans-retinoic acid 4-hydroxylase enzyme(s). (±)-4-(1H-Imidazol-1-yl)retinoic acid (4) is t

Full text of DOI:10.1016/S0960-894X(00)00391-7

Bioorganic & Medicinal Chemistry Letters 10 (2000) 1905±1908
Potent Inhibition of Retinoic Acid Metabolism Enzyme(s)
by Novel Azolyl Retinoids
Vincent C. O. Njar,a,* Ivo P. Nnaneb and Angela M. H. Brodiea
aDepartment of Pharmacology and Experimental Therapeutics, School of Medicine, University of Maryland,
Baltimore, MD 21021-1559, USA
bDepartment of Pharmaceutical Sciences, Temple University School of Pharmacy, Philadelphia, PA 19140, USA
Received 5 May 2000; accepted 13 June 2000
AbstractÐNovel (Æ)-4-azolyl retinoic acid analogues 4, 5, 7 and 8 have been designed and synthesized and have been shown to be
powerful inhibitors of hamster microsomal all-trans-retinoic acid 4-hydroxylase enzyme(s). (Æ)-4-(1H-Imidazol-1-yl)retinoic acid
(4) is the most potent inhibitor of this enzyme reported to date. # 2000 Elsevier Science Ltd. All rights reserved.
Introduction
elucidated, a novel P450 enzyme (designated CPY26)
with speci®c ATRA 4-hydroxylase activity, which is
also rapidly induced by ATRA has recently been cloned
from zebra ®sh,6a mouse,6b and man.6c
Retinoids, natural and synthetic analogues of all-trans
retinoic acid (ATRA) play key roles in many biological
functions, including induction of cellular proliferation,
di€erentiation and apoptosis as well as developmental
changes.1 ATRA is being used in di€erentiation therapy
of cancer, in cancer chemoprevention, and for the
treatment of acne. In spite of these encouraging results, the
e€ects of ATRA therapy on human prostate cancer in the
clinic has been scarce and disappointing.2 It has been sug-
gested that the therapeutic e€ects of ATRA are under-
mined by its rapid in vivo metabolism and catabolism by
cytochrome P450-dependent enzyme(s).3
ATRA 4-hydroxylase activity was previously thought to
mainly reside in the liver, but its presence has now been
unequivocally demonstrated in a number of organs,
skin, and tumor cells.7 In principle, inhibitors of ATRA
4-hydroxylase (also referred to as retinoic acid metabo-
lism blocking agents [RAMBAs]) should delay in vivo
ATRA catabolism resulting in increased endogenous
levels. This e€ect should improve control of neoplastic
di€erentiation and growth and possibly exhibit antitumor
activity. This rationale has been extensively tested with
liarozole fumarate (LiazalTM) in vitro, in vivo as well as
in the clinic by researchers of Janssen Pharmaceutica
NV and collaborators.4,8 Laizal may soon become avail-
able for the treatment of a number of cancers and also of
dermatological diseases.9 However, this compound is
not speci®c and inhibits ATRA 4-hydroxylase only at
micromolar concentration.
One of the strategies for preventing in vivo catabolism
of ATRA is to inhibit the P450 enzyme(s) responsible
for this process. Indeed, this seems to be an emerging
approach that may yield e€ective agents for the chemo-
prevention and/or treatment of prostate and other kinds
of cancer.4 The major pathway of metabolic deactivation
of ATRA starts with hydroxylation at C-4 to form 4-
hydroxy-ATRA that is oxidized into 4-oxo-ATRA, which
is further transformed into more polar metabolites.5 The
®rst and rate-limiting step in the process is catalyzed by
a cytochrome P450-dependent 4-hydroxylase enzyme.
Although the exact nature of this enzyme remains to be
Because of the potential of RAMBAs in cancer therapy
and also the need to design more potent and speci®c
inhibitors of the enzyme, we recently became interested
in this area. Given the signi®cance of azole groupings of
many drugs,10 which are inhibitors of P450 enzymes, we
reasoned that introducing azole grouping at C-4 of
ATRA should yield speci®c and potent inhibitors of this
enzyme. With this design, it may be possible to produce
substrate-based compounds that will not only interact with
*Corresponding author. Fax: +1-410-706-0032; e-mail: vnjar001@
umaryland.edu
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00391-7

1906
V. C. O. Njar et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1905±1908
the retinoid-binding site of the enzyme, thus introducing
high speci®city, but also provide a sixth ligand to the
enzyme's heme iron resulting in tight binding. We have
successfully used this approach in preparing inhibitors
with Ki values in the nanomolar range against 17a-
hydroxy/17,20-lyase (CYP17), another cytochrome
P450-dependent enzyme that catalyzes the conversion of
pregnane steroids into androgens.11 In this communi-
cation, we wish to report the synthesis and biological
activity of novel and highly potent substrate-based
inhibitors of ATRA metabolism enzyme(s). One of our
compounds, (Æ)-4-(1H-imidazol-1-yl)retinoic acid (4) is
the most potent RAMBA reported to date.12
the reaction of 2 with appropriate N,N0-carbonyldiazole
reagents in dry CH3CN at rt to give the corresponding
4-azolyl retinoids in almost quantitative yields. Indeed,
this smooth allylic nucleophilic substitution reaction is
remarkable given that such reactions are usually plagued
with rearrangement reactions.14
The synthesis of our most potent inhibitor, (Æ)-4-(1H-
imidazol-1-yl)retinoic acid 4, is described in detail below.
Compound 2 was synthesized in three steps from the
commercially available ATRA (1) following established
procedure.15 The hydroxy compound 2 in dry CH3CN
was treated with CDI at rt to give the 4-imidazole deri-
vative 3 (95% yield). Alkaline hydrolysis of 3 in re¯uxing
methanol under N2 gave the desired (Æ)-4-(1H-imida-
zol-1-yl)retinoic acid (4, 80% yield).
Chemistry
Although a number of azoles have been synthesized
directly from alcohols or from their corresponding halides
or sulfonates, we expected diculties in the present case
because of the allylic nature of the 4-hydroxy group of
4-hydroxy methyl retinoate (4-HMR, 2), a key inter-
mediate in our projected synthetic scheme. Literature
precedent for the conversion of benzylic alcohols into
the corresponding imidazole involves treatment of the
alcohol with N,N0-carbonyldiimidazole, CDI) in re¯ux-
ing CHCl3 for 1 h.13 Only a trace (ꢀ3%) of the desired
product together with many intractable by-products
were obtained on applying these conditions to 2. Fur-
thermore, attempts to transform 2 to the corresponding
sulfonate or halide failed. We eventually synthesized the
desired 4-azolyl retinoids as outlined in Scheme 1. A key
step in the preparation of our new inhibitors involves
Reaction of 2 with N,N0-carbonyldi(1,2,4-triazole) (CDT,
Scheme 1) as described above gave a mixture of two
regioisomers, which were readily separated by ¯ash col-
umn chromatography (FCC, silica gel) into the less
polar 1H-1,2,4-triazole (5) and the more polar 4H-1,2,4-
triazole (6) in the ratio of 1.8:1, respectively. These iso-
mers were easily identi®ed by their proton NMR spectra
as the two nonequivalent protons of the triazole moiety
in 5 appeared as two singlets at d 7.98 and 8.02 while the
two equivalent protons in 6 appeared as a singlet at d
8.15. These assignments are in agreement with data of
1,2,4-triazole regioisomers of established structure.11
These triazoles were each hydrolyzed to give the desired
(Æ)-4-(1H-1,2,4-triazol-1-yl)retinoic acid (7) and (Æ)-4-
(4H-1,2,4-triazol-1-yl)retinoic acid (8). All compounds
gave satisfactory analytical and spectroscopic data.16
Scheme 1. Reagents and conditions: (i) TMSCHN2/benzene:MeOH, N2, rt; (ii) Activated MnO2/CH2Cl2, rt; (iii) NaBH4/MeOH, rt; (iv) CDI/
CH3CN, rt; (v) 10% KOH/MeOH, N2, re¯ux; (vi) CDT/CH3CN, rt.

V. C. O. Njar et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1905±1908
1907
Proton NMR chemical shifts for these compounds were
assigned by comparison with reported values of closely
related compounds.11,15
Table 1. Inhibition of ATRA catabolism
Compound
IC50, mMa
4
5
7
0.10 (Æ0.001)
0.68 (Æ0.003)
0.88 (Æ0.005)
1.62 (Æ0.008)
6.00 (Æ0.008)
34.00 (Æ0.008)
Biological Results and Discussions
8
Liazal
Ketoconazole
The prospective inhibitors were evaluated using micro-
somal preparations of male hamster liver forti®ed with
NADPH. Using [11,12-3H]retinoic acid (7.72pmol) as
substrate, the activity of ATRA 4-hydroxylase corre-
sponds to production of polar metabolites, including 4-
hydroxy- and 4-oxo-ATRA.17 The products were analyzed
and quanti®ed by reverse phase HPLC, the radioactivity
being measured by an on-line radio-detector. Figure 1A
shows a representative chromatogram of the radioactivity
pro®le obtained by HPLC analysis of metabolites of
ATRA metabolism.
aValues are means of three experiments, standard deviation is given in
parentheses. IC50 values for inhibition of ATRA catabolism were
measured using 7.27 pmol [11,12-3H]-ATRA as substrate and were
determined from dose±response curves.
inhibitor of this enzyme previously described. The inhibi-
tory potencies of these compounds versus the novel
cytochrome P450 enzyme (CYP26) with speci®c ATRA
4-hydroxylase activity, are warranted and should be of
utmost interest. Furthermore, it would be expected that
these compounds will inhibit ATRA metabolism in vivo
resulting in enhanced basal ATRA levels, given that their
inhibitory potencies are in the range that could potentially
be useful for physiological studies.
Thus, ATRA was converted into at least four highly
polar metabolites (HPM, retention time, Rt=6±10 min),
two prominent metabolites of medium polarity (MMP,
Rt=14 and 16min, respectively), which eluted at the same
positions as authentic 4-oxo-and (Æ)-4-hydroxy-ATRA,
respectively. In the presence of 1.0 mM 4 (Fig. 1B), the
HPMs were completely abolished and the MMPs were
signi®cantly suppressed. This result indicates that the
two prominent MMP, most probably, 4-hydroxy- and
4-oxo-ATRA are the obligatory precursors of the
HPMs, in agreement with current consensus.17
In summary, we have developed a new method that
enabled us to synthesize novel 4-azolyl retinoids that
proved to be powerful inhibitors of hamster microsomal
ATRA metabolism enzyme(s). Compound 4 is the most
potent RAMBA known to date. These compounds may
have therapeutic potential for the treatment of cancers
and dermatological diseases. Further work in this direc-
tion is being pursued.19
IC50 values for inhibition of ATRA catabolism by our
novel compounds, liarozole fumarate and ketoconazole
were determined using 7.72 pmol [11,12-3H]-ATRA as
the substrate. The results are presented in Table 1.
Acknowledgements
All the novel azolyl compounds were more potent than
either ketoconazole or liarozole. The results suggest that
the nature of azole is important in determining the a-
nity for the enzyme. The most active compound, 4 with
an IC50 value of 0.1 mM is 60-fold more potent than
liarozole (IC50=6.0 mM) the only RAMBA to undergo
phase III clinical studies in prostate cancer and in
psoriasis.9,13 Compound 4 is by far more potent than any
We thank Dr. Marcel Janssen, Janssen Pharmaceutica,
Beerse, Belgium for the generous gift of liarozole and
Professor H. L. Holland, Department of Chemistry,
Brock University, St. Catharines, Canada for HRMS.
Note Added in Proof
We recently became aware of a report by L. Nadin and
M. Murray which suggests that CYP2C8 is a major
contributor to ATRA 4-hydroxylation in human liver
(Biochem. Pharmacol. 1999, 58, 1201).
References and Notes
1. Mangelsdorf, D. A.; Umesono, K; Evans, R. M. In The
Retinoids; Sporn, M. B., Roberts, A. B., Goodman, D. S., Eds;
Raven: New York, 1994; pp 319±349.
2. Hong, W. K.; Itri, L. In The Retinoids; Sporn, M. B.,
Roberts, A. B., Goodman, D. S., Eds; Raven: New York,
1994; pp 923±930
3. Napoli, J. L.; McCormick, A. M. Biochem. Biophys. Acta
1981, 660, 165.
4. Miller, Jr., W. H. Cancer 1998, 83, 1471.
5. Frolik, C. A.; Roberts, A. B.; Tavala, T. E.; Roller, P.;
Newton, D. L.; Sporn, M. B. Biochemistry 1979, 18, 2092.
Figure 1. Reverse-phase HPLC chromatograms showing the metabo-
lism of [11,12-3H] in the absence (A) or presence of 1 mM 4 (B).18

1908
V. C. O. Njar et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1905±1908
6. (a) White J. A.; Guo, Y-D.; Baetz, K.; Beckett-Jones, B.;
Bonasoro, J.; Hsu, K. E.; Dilworth, F. J.; Jones. G.; Petko-
vich, M. J. Biol. Chem. 1996, 271, 29922. (b) Abu-Abed, S. S.;
Beckett, B. R.; Chiba, H.; Chithalen, J. V.; Jones, G.; Metzger,
D.; Chambon, P.; Petkovich, M. J. Biol. Chem. 1998, 273,
2409. (c) White J. A.; Beckett-Jones, B.; Guo, Y-D.; Dilworth,
F. J.; Bonasoro, J.; Jones. G.; Petkovich, M. J. Biol. Chem.
1997, 272, 18538.
7. (a) Vanden Bossche, H.; Willemsens, G. In Retinoids: 10
Years On; Staurat J. H., Ed.; Karger: Basel, 1990; pp 78±88.
(b) Wouters, W.; Van Dun, J.; Dillen, A.; Coene, M.-C.;
Cools, W.; De Coster, R. Cancer Res. 1992, 52, 2841. (c)
Krekels, M. D. W. G.; Zimmerman, J.; Janssens, B.; Van
Ginckel, R.; Van Hove, C.; Coene, M.-C.; Wouters, W.
Prostate 1995, 29, 35-41.
8. (a) Van Wauwe, J. P.; Coene, M-C.; Goossens, J.; Van
Nijen, G.; Cools, W.; Lauwers, W. J. Pharmacol. Exp. Ther.
1988, 245, 718. (b) De Coster, R.; Wouters, W.; Van Ginckel,
R.; End, D.; Krekels, M.; Coene, M.-C.; Bowden, C. J.
Steroid Biochem. Molec. Biol. 1992, 43, 197.
9. Waxman, J.; Roylance, R. Eur. J. Cancer 1998, 34, 437.
10. Vanden Bossche, H.; Koymans, L.; Mooreels, H. Phar-
mac. Ther. 1995, 57, 79.
11. Njar, V. C. O.; Kato, K.; Nnane, I. P.; Grigoryev, D. N.;
Long, B. J.; Brodie, A. M. H. J. Med. Chem. 1998, 41, 902.
12. A portion of this work was recently presented as an oral
paper: Njar, V. C. O.; Nnane, I. P.; Brodie, A. M. H. Abstract
of Papers, 218th National Meeting of the American Chemical
Society, New Orleans, LA, August 22±26 1999; American
Chemical Society: Washington, DC, 1999; MEDI 166.
13. Freyne, E.; Raeymaekers, A.; Venet, M.; Sanz, G.; Wou-
ters, W.; De Coster, R.; Van Wauwe, J. Bioorg. Med. Chem.
Lett. 1998, 8, 267.
3H, 18-CH3), 1.87 (m, 2H, 2-Hs), 2.02 (s, 3H, 19-CH3), 2.32 (s,
3H, 20-CH3), 4.84 (s, 1H, 4-H), 5.85 (s, 1H, 14-H), 6.21 (s, 3H,
7-, 10- and 12-Hs), 6.33 (d, J=15.0 Hz, 1H, 8-H), 7.00 (t,
J=14.0 Hz, 1H, 11-H), 7.16 (s, 1H, 40-H), 7.26 (s, 1H, 50-H),
7.46, (s, 1H, 20-H), 8.75 (brs, 1H,-COOH). Anal. calcd for
C23H30O2N2: C, 75.38; H, 8.25; N. 7.64. Found: C, 75.72; H,
8.65; N, 7.67. HRMS calcd for C23H30O2N2 366.3061, found
366.305. Data for 7: mp 95±97 ꢁC. 1H NMR (300 MHz) d 1.10
and 1.13 (2s, 6H, 16- and 17-CH3s), 1.50 (brs, 2H, 3-Hs), 1.65
(s, 3H, 18-CH3), 1.97 (m, 2H, 2-Hs), 2.03 (s, 3H, 19-CH3), 2.36
(s, 3H, 20-CH3), 4.86 (s, 1H, 4-H), 5.84 (s, 1H, 14-H), 6.32 (m,
4H, 7-, 8-, 10- and 12-Hs), 7.01 (t, J=14.5 Hz, 1H, 11-H), 8.10
(s, 1H, 30-H), 8.31 (s, 1H, 50-H), 8.46 (brs, 1H,-COOH). Anal.
calcd for C22H29O2N3: C, 71.90; H, 7.95; N, 11.43. Found: C,
71.70; H, 8.11; N, 11.53. HRMS calcd for C22H29O2N3
367.4938, found 367.4935. Data for 8: mp 105±108 ꢁC. 1H
NMR (300 MHz) d 1.11 and 1.14 (2s, 6H, 16- and 17-CH3s),
1.52 (m, 2H, 3-Hs), 1.65 (s, 3H, 18-CH3), 1.96 (m, 2H, 2-Hs),
2.03 (s, 3H, 19-CH3), 2.37 (s, 3H, 20-CH3), 4.78 (s, 1H, 4-H),
5.85 (s, 1H, 14-H), 6.19 (m, 4H, 7-, 8-, 10- and 12-Hs), 7.01 (t,
J=14.2 Hz, 1H, 11-H), 8.46 (s, 2H, 30- and 50-Hs), 8.65 (brs,
1H,-COOH). Anal. calcd for C22H29O2N3: C, 71.90; H, 7.95;
N, 11.43. Found: C, 71.90; H, 7.79; N, 11.30. HRMS calcd for
C22H29O2N3 367.4938, found 367.4939.
17. Van Weuwe, J.; Van Nyen, G.; Coene, M.-C.; Stoppie, P.;
Cols, W.; Goossens, J.; Borghgraef, P.; Janssen, P. A. J. J.
Pharmacol. Expt. Ther. 1992, 261, 773.
18. Hamster liver microsomes (500 mg of protein) were incu-
bated for 30 min at 37 ꢁC with [11,12-3H]ATRA (0.4 mCi,
7.72 pmol) and NADPH. The extracts were analyzed with C18
Bondapak column eluted with a multi-linear gradient solvent
system: I, MeOH:H2O:HCOOH (60:40:0.05) containing
10 mM NH4OAc (100!0%) and ii, MeOH (0!100%) at
2 mL/min. The Rts of ATRA, 4-hydroxy-ATRA and 4-oxo-
ATRA were determined by UV absorbance at 350 nm. Typi-
cally, 80Æ5% of [11,12-3H]ATRA was converted into the
metabolites.
14. Njar, V. C. O. Synthesis, in press.
15. (a) Rosenberger, M. J. Org. Chem. 1982, 47, 1698. (b)
Samokyszyn, V. M.; Gall, W. E.; Zawada, G.; Freyalden-
hoven, M. A.; Chen, G.; Mackenzie, P. I.; Tephly, T. R.;
Radominska-Pandya, A. J. Med. Chem. 2000, 275, 6908.
19. Detailed experiments with these novel RAMBAs involving
enzyme speci®city, retinoid receptor binding, and in vivo anti-
tumor studies are in progress.
16. Data for 4: mp 128±130 ꢁC. H NMR (300 MHz, CDCl3)
1
d 1.13 (s, 6H, 16- and 17-CH3s), 1.56 (m, 2H, 3-Hs), 1.67 (s,