Conformationally defined retinoic acid analogues. 4. Potential new agents for acute promyelocytic and juvenile myelomonocytic leukemias

Article abstract of DOI:10.1021/jm970635h

We recently synthesized several conformationally constrained retinoic acid (RA) analogues [8-(2'cyclohexen-1'-ylidene)-3,7-dimethyl-2,4,6- octatrienoic acids with different alkyl substituents at 2' (R₁) and 3' (R₂) positions on the cyclohexene ring] (Muccio et al. J. Med. Chem. 1996, 39, 3625) as cancer chemopreventive agents. UAB8 (R₁ = Et; R₂ = (i)Pr), which contains sufficient steric bulk at the terminal end of the polyene chain to mimic the trimethylcyclohexenyl ring of RA, displayed biological properties similar to those of RA. To explore the efficacy of this retinoid in acute promyelocytic leukemia (APL) and juvenile myelomonocytic leukemia (JMML), we evaluated UAB8 isomers in in vitro assays which measure the capacity of retinoids to inhibit aberrant myeloid colony growth from blood or bone marrow cells obtained from human JMML patients and in assays measuring the potential of retinoids to differentiate NB4 cells (an APL cell line). Both (all-E)- and (13Z)-UAB8 were 2-fold more active than RA in the NB4 cell differentiation assay; however, only (all-E)-UAB8 had comparable activity to the natural retinoids in the JMML cell assays. These results were compared to the biological effectiveness of a new retinoid, UAB30 [8-(3',4'-dihydro-1'(2'H)- naphthalen-1'-ylidene)-3,7-dimethyl-2,4,6octatrienoic acid], which had different nuclear receptor binding and transactivational properties than UAB8. Relative to (all-E)-RA and (all-E)-UAB8, (all-E)-UAB30 bound well to RARα but did not activate transcription-mediated RARα homodimers, even though it was effective in RARβ- and RARγ-mediated transactivational assays. In APL assays, this retinoid had much reduced activity and was only moderately effective in JMML assays and in cancer chemoprevention assays.

Full text of DOI:10.1021/jm970635h

J . Med. Chem. 1998, 41, 1679-1687
1679
Con for m a tion a lly Defin ed Retin oic Acid An a logu es. 4. P oten tia l New Agen ts
for Acu te P r om yelocytic a n d J u ven ile Myelom on ocytic Leu k em ia s
Donald D. Muccio,*,† Wayne J . Brouillette,*,† Theodore R. Breitman,| Mohammed Taimi,| Peter D. Emanuel,§
Xiao-kun Zhang, Guo-quan Chen, Brahma P. Sani,‡ Pratap Venepally,‡ Lakshmi Reddy,‡ Muzaffar Alam,†
Linda Simpson-Herren,‡ and Donald L. Hill‡
Departments of Chemistry and Medicine, University of Alabama at Birmingham, Birmingham, Alabama 35294,
Laboratory of Biological Chemistry, Division of Basic Sciences, National Cancer Institute, National Institutes of Health,
Bethesda, Maryland 20892, The Burnham Institute, La J olla Cancer Research Center, 10901 North Torrey Pines Road,
La J olla, California 92037, and Departments of Biochemistry and Toxicology, Southern Research Institute,
2000 Ninth Avenue South, Birmingham, Alabama 35205
Received September 22, 1997
We recently synthesized several conformationally constrained retinoic acid (RA) analogues [8-(2′-
cyclohexen-1′-ylidene)-3,7-dimethyl-2,4,6-octatrienoic acids with different alkyl substituents
at 2′ (R1) and 3′ (R2) positions on the cyclohexene ring] (Muccio et al. J . Med. Chem. 1996, 39,
3625) as cancer chemopreventive agents. UAB8 (R1 ) Et; R2 ) iPr), which contains sufficient
steric bulk at the terminal end of the polyene chain to mimic the trimethylcyclohexenyl ring
of RA, displayed biological properties similar to those of RA. To explore the efficacy of this
retinoid in acute promyelocytic leukemia (APL) and juvenile myelomonocytic leukemia (J MML),
we evaluated UAB8 isomers in in vitro assays which measure the capacity of retinoids to inhibit
aberrant myeloid colony growth from blood or bone marrow cells obtained from human J MML
patients and in assays measuring the potential of retinoids to differentiate NB4 cells (an APL
cell line). Both (all-E)- and (13Z)-UAB8 were 2-fold more active than RA in the NB4 cell
differentiation assay; however, only (all-E)-UAB8 had comparable activity to the natural
retinoids in the J MML cell assays. These results were compared to the biological effectiveness
of a new retinoid, UAB30 [8-(3′,4′-dihydro-1′(2′H)-naphthalen-1′-ylidene)-3,7-dimethyl-2,4,6-
octatrienoic acid], which had different nuclear receptor binding and transactivational properties
than UAB8. Relative to (all-E)-RA and (all-E)-UAB8, (all-E)-UAB30 bound well to RARR but
did not activate transcription-mediated RARR homodimers, even though it was effective in
RARâ- and RARγ-mediated transactivational assays. In APL assays, this retinoid had much
reduced activity and was only moderately effective in J MML assays and in cancer chemopre-
vention assays.
In tr od u ction
Retinoid receptors and other members of this super-
family of nuclear receptors (that include the steroid,
thyroid, and vitamin D hormone receptors and other
‘orphan’ receptors without known ligands) are new
organ tumors, and major therapeutic successes have
been demonstrated with retinoids in certain leukemias.3
(all-E)-RA treatment of patients with acute promyelo-
cytic leukemia (APL) leads to a 90% complete remission
rate in these patients by inducing normal maturation
and apoptosis of APL myeloblasts to neutrophils, but
this differentiation therapy is transient and is commonly
followed by relapse within 3-15 months, probably due
to the development of resistance to RA.4 (13Z)-RA
effectively controls the excessive myeloproliferation in
up to 50% of children with juvenile myelomonocytic
leukemia (J MML).5 However, this treatment is not
curative and at best can lead to a period of prolonged
stabilization of disease, but ultimately patients need to
undergo allogenic bone marrow transplantation.4,6
To provide new, more effective therapeutic agents
with the beneficial effects of RA but with reduced side
effects, we designed7 conformationally locked retinoids.
We previously demonstrated7,8 that (all-E)- and (9Z)-
UAB8 retinoids (containing large R1 and R2 substituents
in the ring region of the UAB retinoid general structure)
targets for drug development.1 It is thought that
retinoic acid (RA) and synthetic retinoids act as ligand-
dependent transcription factors with different members
of nuclear retinoid receptors to control gene transcrip-
tion responsible for cellular proliferation, differentiation,
development, and cell death.2 Two classes of nuclear
retinoid receptors (RARs and RXRs) have been identi-
fied so far, and each has several different subtypes (R,
â, γ). Both (all-E)- and (9Z)-RA bind to RARs and
activate transcription mediated by RAR/RXR het-
erodimers, but (9Z)-RA is the only known natural ligand
for the RXRs which mediate transcription by forming
homodimers or heterodimers.
Recent advances in chemoprevention have heightened
interest in the use of retinoids in several types of solid
* To whom correspondence should be addressed.
† Department of Chemistry, University of Alabama at Birmingham.
§ Department of Medicine, University of Alabama at Birmingham.
| National Institutes of Health.
La J olla Cancer Research Center.
‡ Southern Research Institute.
S0022-2623(97)00635-3 CCC: $15.00 © 1998 American Chemical Society
Published on Web 04/11/1998

1680 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 10
Muccio et al.
Sch em e 1
were selective RAR and RXR agonists and further
showed a correlation with enhanced biological activity
in skin. In this study we examine three UAB8 isomers
in in vitro differentiation and antiproliferation assays
that predict for efficacy in APL and J MML therapy,
respectively, and show equal or greater efficacy with the
(all-E)- and (13Z)-UAB8 isomers as compared to the
natural retinoids. Additionally we generate a new UAB
retinoid, UAB30, whose all-E-isomer has different
nuclear receptor selectivity (RARâ and RARγ subtype-
selective retinoid and RARR antagonist) than (all-E)-
UAB8. A comparison of the activities of UAB8 and
UAB30 isomers in these in vitro assays suggests the
importance of RARR binding/activation in ligand-
induced differentiation of APL cells and RAR binding
in antiproliferative effects in J MML.
mixture was oxidized with MnO2 to give a comparable
mixture of aldehydes (4), which were preparatively
separated using flash chromatography on silica gel.
Each pure isomer of 4 was olefinated as shown to
provide a mixture of either (9Z)- and (9Z,13Z)-5 or (all-
E)- and (13Z)-5. Each mixture was preparatively
separated by HPLC on silica gel (0.5% Et2O, 0.1% THF
in hexane for the former and 1% Et2O, 0.5% THF in
hexane for the latter) using methods similar to those
we previously described.8-11 Individual isomers of ester
5 were then hydrolyzed to the corresponding acids in
KOH, without E/Z-isomerization.12 The isomeric purity
for the final acids (>97%) was verified by NMR and
reverse-phase HPLC.8-11 Experimental yields and se-
lected data for the intermediates and products in
Scheme 1 are summarized in Table 1. The UV/vis
spectral data of the purified isomers are given in Table
1, and the complete 1H NMR assignments are contained
in Table 2.
Biology
Using competitive binding assays as described else-
where,8-10 each isomer was evaluated for its ability to
bind to cytoplasmic retinoic acid binding protein (CRABP)
and nuclear retinoid receptors (RARR, RARâ, RARγ, and
RXRR). The 50% inhibitory concentrations (IC50 values)
were determined from a retinoid dose-response curve.
Functional assays (transient transfection assays) were
used to measure the capacity of retinoids to activate
RARR, RARâ, and RARγ homodimers, RXRR ho-
modimers, or an RXRR/RARR heterodimer using a
chloramphenicol acetyltransferase (CAT) reporter gene
containing TREpal with a thymidine kinase promoter
[(TREpal)2-tk-CAT] as previously described.9,13 Dose-
response curves of RA and UAB retinoids were used to
determine the concentrations causing 50% of the maxi-
Ch em istr y
Scheme 1 summarizes the methods employed to
prepare UAB30. The approach was similar to that
which we previously reported for the syntheses of the
alkyl-substituted UAB retinoids (e.g., UAB8),9,10 except
that R-tetralone was used as the starting enone (1 in
Scheme 1). (Note that atom numbering in Scheme 1 is
nonsystematic to allow ready comparison with RA.) As
noted before,9 a δ-lactone intermediate was not detected
under these conditions but was assumed to be an
intermediate in the production of (9Z)-2. Unlike earlier
reports for related compounds,8-10 the reduction of acid
(9Z)-2 to alcohol 3 was accompanied by isomerization
to give a 1:1 mixture of all-E- and 9Z-isomers. This

Conformationally Defined Retinoic Acid Analogues
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 10 1681
Ta ble 1. Selected Data for New Compounds Produced in
Scheme 1
ture of (9Z)-2 to that generated by molecular mechanics,
Allinger’s MM3(94) calculations were performed as we
described previously for calculating the structures of
UAB1-UAB8 retinoids.8 The calculated structure of
this intermediate was very similar to the solid-state
structure. The ψ5,6,7,8 torsional angles were nearly
planar (-160°) in each structure, like other UAB ret-
inoids with alkyl substituents.8 The ψ7,8,9,10 torsional
angle was -133° for the molecular mechanics structure
and -134.3° in the crystral structure. This angle is
nonplanar due to steric interactions between the C2′
methylene and the C19 methyl group. The molecular
mechanics structures of (9Z)- and (all-E)-UAB30 retin-
oids were also generated. When compared to those of
UAB8 isomers,8 the expected low-energy conformers are
very similar; the aromatic ring of UAB30 retinoids
occupies space that is very similar to that of the alkyl
isolated
percent
yield
UV/visb
Rf λmax
IRc
CdO CdC m/z
MS
a
compound
(9Z)-2
(9Z)- and
ꢀ
86
67
0.35 310 14 000 1673 1618 ND
0.21 264 12 000 (3334) 1630 214.1
(all-E)-3
(9Z)-4
59d
59d
78
92
92
98
93
93
0.48 295 6 000 1662 1609 212.1
0.42 298 8 200 1662 1605 212.1
0.59 328 29 300 1701 1602 322.2
0.57 358 35 000 1708 1605 322.2
0.57 358 30 000 1709 1604 322.2
0.12 328 30 200 1672 1594 294.2
0.09 358 32 500 1672 1598 294.2
0.12 359 26 000 1673 1596 294.2
(all-E)-4
(9Z)-5e
(all-E)-5
(13Z)-5
(9Z)-UAB30f
(all-E)-UAB30
(13Z)-UAB30
a
Values are on silica gel using ether/hexane as eluent: 30%
b
Et2O, 3 and 4; 20% Et2O, 2; 10% Et2O, 5 and UAB30. The
wavelength maximum (nm) and extinction coefficients (M-1 cm-1
)
were obtained in cyclohexane (5) or methanol (UAB30, 4) at room
temperature. ND means not determined. c The IR stretching
frequencies (cm-1) were obtained as thin films on NaCl disks.
Compound 3 OH stretching frequency is reported in parentheses.
i
groups (R1 ) Et; R2 ) Pr) of UAB8.
We previously noted that the 8-s-cis-conformer was
as stable as the 8-s-trans-conformer for different UAB
retinoids.8 The low-energy structures of the 8-s-cis-
conformer of UAB30 isomers were also calculated. The
ψ5,6,7,8 torsional angles were nearly identical to that of
the 8-s-trans-conformer, and the ψ7,8,9,10 angle was -70°.
Further, the 8-s-cis-conformer was slightly more stable
(∆G ≈ 1 kcal/mol) than the corresponding 8-s-trans-
conformer.
d
Yield for the mixture of E/ Z-isomers. e Ester 5 was preparatively
separated on a Whatman Partisil 10 M20/50 normal phase column
with 1.0% Et2O and 0.5% THF in hexane with a 5-mL/min flow
rate.9 Retention times were 138.6 min (13Z), 140.2 min (9Z), and
142.6 min (all-E). Re-injected fractions showed one isomer when
monitoring at either 300 or 340 nm. f UAB30 isomers were
analyzed for isomeric purity by reverse-phase chromatography
using a Spherisorb ODS C-18 column with 1% acetic acid in
acetonitrile (3:7) with a 1.0 mL/min flow rate. Using 340-nm
detection (where each isomer had very similar extinction coef-
ficients), the isomeric ratios were determined from the integrated
peak areas. The isomeric ratios were greater than 95% for each
isomer.
Nuclear Overhauser experiments (NOESY) (2D) were
performed on (9Z)-2 in CDCl3. Significant NOEs were
observed between H-19 and H-2′ (consistent with the
8-s-trans-conformer) and between H-19 and H-8 (con-
sistent with the 8-s-cis conformer). Relative to the very
intense cross-peak intensity found between the H-19
methyl protons and H-10 (indicating the 9Z-configura-
tion), the intensities of the negative cross-peaks ob-
served between H-19/H-2′ and H-19/H-8 were about one-
half this integrated intensity. Similar results were
obtained from NOESY experiments on all-E- and 9Z-
isomers of UAB30. The NOE results are in support of
the MM3 calculations which identified two low-energy
conformers about the C8-C9 bond in the polyene chain
of these retinoids.
mum effect (EC50 values). To compare UAB30 retinoids
to other conformationally restricted UAB retinoids,
namely UAB8, these isomers were evaluated in a skin
antipapilloma assay which measured the capacity of
retinoids to prevent chemically induced papillomas in
mice.14 This assay uses a two-stage model for carcino-
genesis involving initiation by dimethylbenz[a]an-
thracene (DMBA) and promotion by 12-O-tetrade-
canoylphorbol 13-acetate (TPA). RA or retinoid appli-
cation before promotion prevents papillomas in a dose-
dependent manner, and the 50% effective doses (ED50
values) were measured as reported previously.8-10 For
evaluation of effectiveness in APL, UAB8 and UAB30
isomers were screened in an in vitro assay that mea-
sures the capacity of retinoids to induce differentiation
of NB4 cells. This cell line was established from an APL
patient and carries the (15;17) translocation which
involves the RARR receptor. These assays were per-
formed according to the recently published methods of
Breitman and co-workers.15 For the evaluation of
retinoids in J MML, the UAB8 and UAB30 retinoids
were screened for their ability to inhibit spontaneous
granulocyte-macrophage colony-forming units (CFU-
GM) using an in vitro assay with human cells derived
from different J MML patients. These were performed
by the methods developed by Emanuel et al.,5,16 who
first correlated the in vitro activity of (13Z)-RA to its
therapeutic benefit in J MML patients.
CRABP Bin d in g Affin ity. The IC50 values were
evaluated in a chick skin CRABP radioligand binding
assay with [3H]-(all-E)-RA as the radioligand.18 At 100-
fold excess of unlabeled UAB retinoid, (all-E)-UAB30
inhibited the binding of [3H]RA by only 75%; (13Z)-
UAB30 (28% inhibition) and (9Z)-UAB30 (5% inhibition)
had even less affinity for CRABP. The lower affinity of
UAB30 isomers (IC50 > 2000 nM) relative to the all-E-
isomers of RA (IC50 ) 600 nM) is surprising especially
since UAB7 and UAB8 retinoids had binding affinities
comparable to that of RA. Apparently, the aromatic
ring residue of UAB30 does not interact as well as the
bulky alkyl groups of UAB8 in the RA binding site of
CRABP, even though these groups are positioned in a
similar region of space in solution.
Nu clea r Recep tor Bin d in g Affin ity a n d Tr a n -
scr ip tion a l Activity. The all-E-, 9Z-, and 13Z-isomers
of UAB30 were evaluated for their ability to inhibit the
binding of [3H]-(all-E)-RA to nuclear retinoid receptors.
To survey the inhibition process, nuclear receptors
(RARR, RARâ, RARγ, and RXRR) were first exposed to
1000 nM UAB retinoids and 5 nM [3H]-(all-E)-RA. This
Resu lts a n d Discu ssion
Str u ctu r a l Stu d ies. We recently reported17 the
X-ray crystal structure of (9Z)-2, which crystallized in
an 8-s-trans-conformation. To compare the X-ray struc-

1682 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 10
Muccio et al.
Ta ble 2. 1H NMR Chemical Shift Assignments (ppm, TMS) for UAB30 Retinoids at 400 MHz
compound
H-1′
H-2′
H-4
H-3
H-2
H-1
H-8
H-10
H-11
H12
H-14
H-18
9Me
13Me
R-tetralone
(9Z)-2
2.11
1.84
1.82
1.82
1.86
1.86
1.85
1.83
1.84
1.85
1.84
1.84
2.63
2.57
2.70
2.36
2.76
2.50
2.78
2.41
2.78
2.78
2.41
2.79
7.22
7.09
7.09
7.08
7.14
7.14
7.09
7.12
7.09
7.10
7.12
7.08
7.44
7.17
7.16
7.16
7.22
7.23
7.15
7.19
7.15
7.17
7.20
7.15
7.27
7.16
7.15
7.14
7.20
7.19
7.15
7.19
7.15
7.15
7.20
7.15
8.01
7.66
7.58
7.58
7.61
7.64
7.58
7.64
7.58
7.58
7.59
7.58
2.94
2.80
2.80
2.83
2.83
2.86
2.81
2.85
2.80
2.81
2.85
2.79
7.14
6.39
6.36
6.49
6.57
6.51
6.47
6.50
6.52
6.48
6.51
5.79
5.62
5.55
6.04
6.01
6.23
6.11
6.34
6.25
6.12
6.35
11.79
4.29
4.06
10.10
9.65
6.96
6.64
6.95
7.01
6.69
7.00
2.13
1.86
1.85
2.35
2.09
2.08
1.98
2.07
2.09
1.99
2.08
(all-E)-3
(9Z)-3
(all-E)-4
(9Z)-4
(all-E)-5
(9Z)-5
6.28
6.23
7.79
6.32
6.25
7.74
5.79
5.75
5.65
5.82
5.77
5.56
2.37
2.23
2.08
2.38
2.23
2.12
(13Z)-5
(all-E)-UAB30
(9Z)-UAB30
(13Z)-UAB30
Ta ble 3. Summary of IC50 and EC50 Valuesa(nM) for UAB8 and UAB30 Retinoids in Nuclear Receptor Binding and Transcriptional
Activation Assays
IC50 (nM)
RARâ RARγ
EC50 (nM)
retinoid isomer
RARR
RXRR
RARR
RARâ
RARγ
RXRR
RARR/RXRR
(all-E)-RA
6
31
5
8
4
60
>2000
82
22
18
2
27
6
10
>2000
27
32
17
(9Z)-RAb
(all-E)-UAB8
(13Z)-UAB8b
(9Z)-UAB8
(all-E)-UAB30
(13Z)-UAB30
(9Z)-UAB30
14
900
6
371
10
708
>2000
>1000
868
>2000
>2000
284
33
51
>1000
>2000
>2000
>2000
20
150
2
42
190
370
>2000
>2000
220
>2000
>2000
118
68
450
>2000
>1000
>2000
>1000
>1000
>1000
>1000
>2000
35
49
55
110
>1000
>2000
>1000
>2000
>1000
>2000
>2000
>2000
>2000
>2000
a
The IC50 and EC50 values were calculated by a Probit analysis of a dose-response curve using concentrations between 1 and 1000
nM. The standard error in the reported IC50 values was (2.0 nM or less. The standard error in the reported EC50 values was (2.5 nM
b
or less, except for the EC50 value reported for the activation of RARγ by (9Z)-RA, which was (4 nM. IC50 values were taken from Alam
et al.9
was compared to a positive control using 1000 nM
unlabeled (all-E)-RA and a control using no retinoid. For
active retinoids, IC50 values were determined by titra-
tion with varying concentrations of UAB retinoid.
(all-E)-UAB30 was the only UAB30 isomer which
efficiently bound to RARs. The IC50 values (30-50 nM)
for RARs were only about 5-fold greater than that for
(all-E)-RA or (all-E)-UAB8 (Table 3) and comparable to
that for (9Z)-RA. The IC50 values for (9Z)-UAB30 were
>1000 nM indicating that it did not bind to RAR
receptors (Table 3), which is consistent with data on
(9Z)-UAB8. The low affinity for RARs is not shared by
(9Z)-RA, which binds nearly as well as (all-E)-RA to
these receptors. Also, (13Z)-UAB30 was an ineffective
binder to the RAR subtypes, which is different from the
binding properties of (13Z)-UAB8 to these nuclear
receptors.9
UAB30 isomers were evaluated in assays which
measured their functional activity within cells to induce
RARR-, RARâ-, and RARγ-mediated transcriptional
activity. CV-1 cells were transiently transfected with
DNA plasmids from the appropriate RAR subtype and
(TREpal)2-tk-CAT.9,19 Using a dose-response curve, the
ED50 values for retinoid-induced transcription of UAB30
isomers were measured and then compared to those
found for (all-E)- and (9Z)-RA and for (all-E)- and (9Z)-
UAB8 (Table 3). (All-E)-UAB30 induced RARâ-medi-
ated receptor-activated transcription; however, 5-fold
higher concentrations were needed to achieve similar
activation to those observed for (9Z)-RA or (all-E)-UAB8.
(all-E)-UAB30 displayed moderate activity in inducing
RARγ-mediated transcription and had no activity for
inducing RARR transcription. As such this UAB30
isomer displays RARâ and RARγ subtype selectivity and
RARR antagonist properties, which is in sharp contrast
to (all-E)-UAB8sa very efficient pan-RAR agonist.
(13Z)-UAB30 displayed much different nuclear receptor
transcriptional properties than its UAB8 analogue; it
had no activity in inducing RAR-mediated gene expres-
sion, whereas (13Z)-UAB8 is an efficient activator of
gene transcription mediated by RARs. (9Z)-UAB30 was
even a poorer activator of reporter genes mediated by
RARs; it was much less active in these assays than (9Z)-
UAB8.
The binding affinity of the UAB30 isomers to RXRR
was examined next. Only (9Z)-UAB30 inhibited the
binding of 20 nM [3H]-(9Z)-RA to RXRR (Table 3). At
100-fold excess of unlabeled UAB retinoid, (9Z)-UAB30
inhibited 70% of the activity relative to (9Z)-RA. The
IC50 value for this isomer was better than that for (9Z)-
UAB8 and only 4-fold higher than that of (9Z)-RA. The
other isomers were much less effective (IC50 > 2000 nM).
When the effect on RXRR homodimer-mediated tran-
scriptional activity was studied, (9Z)-UAB30 was the
only efficient modulator of the reporter gene (Table 3).
Relative to other 9Z-isomers, transcriptional activation
by (9Z)-UAB30 was about 4-fold less efficient than that
by (9Z)-RA but 2-fold better than that by (9Z)-UAB8.
Of the (9Z)-UAB isomers, UAB30 is the most potent and
selective ligand for activating RXRs.
Since (all-E)-UAB30 bound well to RARR (IC50 ) 35
nm) but did not activate this receptor well (EC50 > 1000
nM), its potential to act as an RARR antagonist was next
evaluated. Fixed (all-E)-RA concentrations ranging
between 1 and 1000 nM were evaluated in the tran-
scriptional assay mediated by RARR homodimers using
two concentrations of (all-E)-UAB30 (10 or 100 nM).
These were compared to a negative control of (all-E)-
RA alone and a positive control using Ro-41-5253, a
known RARR antagonist, at 10 and 100 nM. For 50 and
100 nM (all-E)-RA concentrations, the relative CAT
activity of the cells treated with 100 nM (all-E)-UAB30

Conformationally Defined Retinoic Acid Analogues
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 10 1683
Ta ble 4. ED50 Valuesa (nM) for 4-Day Induction of Differentiation of HL-60 and NB4 Cells and Normalized Percent Inhibitionb of
CFU-GM Colony Growth of J MML Cells from 3 Patients
ED50 (nM)
HL-60 NB4
610
patient J 43 (%)
10-6 10-7
patient J 83 (%)
10-6 10-7
patient J 84 (%)
retinoid isomer
M
M
M
M
10-6
M
10-7
M
10-8
M
(all-E)-RA
(13Z)-RA
(9Z)-RA
(all-E)-UAB8
(13Z)-UAB8
(9Z)-UAB8
(9Z,13Z)-UAB8
(all-E)-UAB30
(13Z)-UAB30
(9Z)-UAB30
120
100
109
100
98
80
79
74
100
94
82
88
86
72
100
87
95
103
49
59
44
57
80
39
52
52
33
5
NTc
NT
NT
NT
100
98
85
61
52
32
NT
51
18
38
23
48
72
44
NT
NT
0
>10 000
>10 000
3 400
320
280
910
NT
NT
NT
NT
13
20
28
25
3
>10 000
>10 000
>10 000
8 000
>10 000
>10 000
>10 000
10 000
59
NT
NT
NT
72
88
79
68
82
76
ED50 values were determined from a dose-response curve using four concentrations of retinoids ranging from 10-2 to 10 µM.
Normalized percent inhibition was calculated by (% inhibition of retinoid)/(% inhibition of (all-E)-RA at 10-6 M). The percent inhibition
a
b
of (all-E)-RA at 10-6 M was 85%, 78%, and 61% for the J 43, J 83, and J 84 patient cells, respectively. c NT means not tested.
decreased relative to those containing only (all-E)-RA.
This decrease in CAT activity is consistent with RA
antagonism by (all-E)-UAB30, but the magnitude of
these effects was 2-fold less than that demonstrated by
Ro-41-5253.
demonstrated potent transactivation activity of RAR
homodimers (Table 3).
In h ibition of J MML CF U-GM Colon y F or m a tion .
In 1991, Emanuel et al.21 demonstrated that the spon-
taneous CFU-GM colony formation observed in J MML
is related to the selective hypersensitivity of J MML
hematopoietic progenitor cells to granulocyte-macroph-
age colony-stimulating factor (GM-CSF). Since this
initial report by the UAB group, others22 have confirmed
these observations of GM-CSF hypersensitivity as a
pathogenetic mechanism in J MML. Additionally in
1991, Castleberry, Emanuel, and co-workers also pre-
sented preliminary evidence of the ability of (13Z)-RA
to inhibit in vitro J MML spontaneous CFU-GM growth
as well as evidence for its possible clinical effectiveness
in J MML.21 The final results of the pilot clinical trial
were published in 1994 and suggested that the effects
of (13Z)-RA in the in vitro screen agree with its efficacy
in treating patients with J MML in the clinic.5 In this
study, we first evaluated (all-E)- and (9Z)-RA in this in
vitro assay and showed that these RA isomers also have
excellent potential to inhibit spontaneous CFU-GM
colony growth in cells isolated from three patients with
J MML (Table 4). We next studied the effects of UAB8
and UAB30 isomers on the inhibition of proliferation
of human J MML cells. UAB8 isomers were evaluated
in this assay using cells from two patients (Table 4).
(all-E)-UAB8 was as effective as (all-E)-RA in control-
ling the spontaneous growth of these CFU-GM colonies.
However, (9Z)- and (13Z)-UAB8 were less effective than
the corresponding RA isomers in controlling growth. In
particular for patient J 84 cells, there is a 2-fold decrease
in activity of these retinoids. For UAB30 isomers, the
most potent isomer is (13Z)-UAB30 which has activity
in two patients only slightly less than that of RA. In
contrast to UAB8 isomers, the all-E-isomer of this
retinoid is less active than RA and (all-E)-UAB8. Also,
(9Z)-UAB30 exhibits lower activity in this assay, similar
to that of (9Z)-UAB8.
Ch em op r even tion of Mou se Sk in P a p illom a .
The UAB30 retinoids were next evaluated for their
activity in preventing the chemical induction of papil-
lomas on mouse skin.14 Both (all-E)-RA and (all-E)-
UAB8 were very effective in this chemopreventive
assay. At a 45.9-nmol dose, they completely prevented
tumor formation (95% for RA; 99% for UAB8). The ED50
values for these retinoids were 3.0 (RA) and 2.5 (UAB8)
nmol.8 (all-E)-UAB30 was less effective than this RA
or UAB8 isomer in the skin antipapilloma assay; it only
reduced 59% of the papillomas at 45.9 nmol. Using
dose-response curves, an ED50 could not be accurately
determined for this retinoid, consistent with our efforts
to quantify the activity of other marginally active UAB
retinoids.8 Previously, we8 correlated (in structure-
activity relationships) the high activity of (all-E)-UAB8
(containing large alkyl groups at R1 and R2) to its
efficient binding and/or activation of RARs. Using these
trends, the reduced activity of (all-E)-UAB30 may be
due to its poor ability to activate RARγ and/or lack of
ability to activate RARR (Table 3), especially in com-
parison to (all-E)-UAB8 which has excellent RARγ
activity. The (9Z)-UAB30 isomer was also less effective
than (9Z)-RA in this assay; it prevented only 54% of
tumor formation at 45.9 nmol, a dose at which (9Z)-RA
was highly (95%) active.8 This low activity, however,
is comparable to that of (9Z)-UAB8 in this assay.8 Since
both (9Z)-UAB8 and (9Z)-UAB30 are RXR-selective
retinoids, they are not effective activators of RAR-
mediated pathways, particularly those mediated by
RARâ and RARγ nuclear receptors. Chandraratna et
al.20 showed that RARâ- and RARγ-selective retinoids
are excellent inhibitors of TPA-induced ornithine de-
carboxylase (ODC) activity in skin, an assay which
correlates well with the skin antipapilloma assay.14
They suggested that these RARâ- and RARγ-selective
retinoids are potent inhibitors of ODC activity in vivo
by antagonizing AP-1-mediated gene expression through
RARR binding. If this is the case, then (all-E)-UAB30
(a weak RARR antagonist) should not be an efficient
AP-1 antagonist, but (all-E)-UAB8 should have activity
to repress AP-1 gene transcription in addition to its
There are no consistent chromosomal abnormalities
in J MML, and none that appear to involve the retinoid
receptors. However one very plausible mechanism23 to
explain the apparent modulating effects of RA on GM-
CSF hypersensitivity in J MML is through the RA-
induced antagonistic effects of RARR on the transcrip-
tion factor, AP-1 (c-jun/c-fos). The GM-CSF signal
transduction cascade through Ras appears to have as
one of its major endpoints an upregulation or activation

1684 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 10
Muccio et al.
of the AP-1 response element. This response element
is activated by c-jun/c-fos, and when stimulated by
extracellular signaling (e.g., GM-CSF), induction of AP-1
leads to cellular proliferation, among other processes.
In fact in Northern blot experiments investigating early
response gene expression in J MML, Emanuel and co-
workers22 have noted an apparent constitutive upregu-
lation of both c-jun and c-fos, but not c-myc. As
reviewed by Pfahl24 and demonstrated in epithelial
cells,25 the mechanism by which RA represses gene
transcription mediated by AP-1 is cell line-specific and
may be mediated by either RARs or RXRs.
activity to induce terminal differentiation in this cell
line correlates well with their ability to activate RARR/
RXRR-mediated transcription that presumably removes
the negative block to normal myeloid stem cell dif-
ferentiation. The data presented here support this idea.
As we previously demonstrated,8 both (all-E)-UAB8 and
(13Z)-UAB8 efficiently activate this heterodimer, whereas
the RXR-selective retinoids, (9Z)-UAB8 or (9Z)-UAB30,
do not. As shown in Table 3, (all-E)-UAB30 binds well
but does not activate RARR-mediated transcription,
consistent with its inactivity in this assay.
In conclusion, we have shown that substituting a
fused benzyl ring for alkyl groups has little effect on
the conformation on the polyene chain between C-8 and
C-15, but it markedly changes the biological properties
of the conformationally constrained UAB retinoids. In
particular, the nuclear receptor transactivational pro-
files of all-E-isomers are modified dramatically. For
UAB30, this isomer displays selectivity for activation
of RARâ- and RARγ-mediated transcription (an RARR
antagonist), whereas (all-E)-UAB8 is a potent pan-RAR
agonist. The transactivational profiles of the 9Z-
isomers are very similar; both 9Z-isomers of UAB8 and
UAB30 are RXRR-selective compounds, with the latter
displaying the most potency and selectivity. These
retinoids also exhibit very different efficacies in several
biological assays, and the changes in their activities
correlate well with their individual transcriptional
properties. For example, the high activity of (all-E)-
UAB8 over (all-E)-UAB30 and (9Z)-UAB8/(9Z)-UAB30
in the prevention of skin papillomas, in the induction
of differentiation of NB4 cells, and in the inhibition of
proliferation of J MML cells is consistent with its
demonstrated potent activity to activate gene expression
mediated by each RAR subtype and/or its putative
activity to repress gene expression of AP-1 through RAR
subtypes. The most exciting aspect of the work pre-
sented here is the higher efficacy of (all-E)- and (13Z)-
UAB8 over (all-E)-RA in their ability to differentiate
NB4 cells. The all-E-isomer of UAB8 is 2-fold less toxic
than this isomer of RA,29 and its pharmacokinetics
allows higher concentrations in plasma for prolonged
times. Taken together, these two UAB8 isomers may
be improved agents for clinical use in the treatment of
APL patients.
The precise mechanism for the high activity of (13Z)-
RA in inhibiting the proliferation of these myeloid
progenitor cells in J MML remains an area of investiga-
tion. Assuming a receptor-based mechanism, (13Z)-RA
isomer may isomerize to either (all-E)- or (9Z)-RA and
interact with RAR/RXR receptors. We show here for
the first time that these latter two isomers are as active
as (13Z)-RA in preventing proliferation in J MML cells,
supporting this assumption. To probe RAR/RXR-medi-
ated pathways, we investigated the ability of receptor-
selective retinoids to prevent proliferation of J MML
cells. The high activity of (all-E)-UAB8 in this in vitro
J MML assay and reduced activity of either (9Z)-UAB8
or (9Z)-UAB30, two RXR-selective retinoids, are con-
sistent with a RAR-mediated pathway of transrepres-
sion. (all-E)-UAB30 is less active than the other natural
RA isomers and (all-E)-UAB8. Since it binds well to
each RAR subtype, but does not activate transcription
for RARR (an RARR antagonist), this lower activity in
the in vitro J MML assay relative to (all-E)-UAB8 may
be due to its RARR antagonism, or in other words, lack
of AP-1 antagonism. Ligand binding is a necessary but
not sufficient condition for AP-1 antagonism. This is
not an unusual property of retinoids, since both Dawson
and Chandraratna’s group26 have identified retinoids
that are RARR antagonists but do not induce RAR-
mediated transrepression of AP-1-mediated transcrip-
tion.
In d u ction of Differ en tia tion in HL-60 a n d NB4
Cells. To explore the potential of UAB retinoids for
treatment of APL, we evaluated UAB8 and UAB30
isomers in in vitro assays which measure their ability
to induce differentiation of two APL leukemia cell lines,
HL-60 and NB4 cells. Even though HL-60 cells were
isolated from APL patients, only the NB4 cell line
contains the t(15;17) translocation involving the RARR
to form the RARR/PML fusion protein. Because this
characteristic mutation is stably carried in NB4 cells,
it is a better model for judging retinoid efficacy for this
disease. In NB4 cells, the ED50 values of two UAB
retinoids were much lower than that of RA after
treatment for 3 days: (all-E)-UAB8 (ED50 ) 475 nM)
was 10-fold more active and (13Z)-UAB8 was 3-fold
Exp er im en ta l Section
Ch em istr y. 1H NMR spectra were obtained at 400.1 MHz
(Bruker DRX spectrometer) in CDCl3. NOE experiments were
performed on degassed samples using 2D phase-sensitive
NOESY experiments with different mixing times (between 250
and 2000 ms). Typically, 1 s mixing times were used with 16
pulses for phase cycling and 2 dummy scans. The data were
processed with line broadening of 0.3 Hz in each dimension
and zero-filled to yield 512 × 4096 2D contour plots. The
integrated intensities of the negative cross-peaks were deter-
mined using standard Bruker NMR software features.
UV/vis spectra were recorded on an AVIV 14DS spectro-
photometer in cyclohexane or methanol solutions (Fisher,
Spectrograde). IR spectra were recorded using a Nicolet FT
IR spectrometer on thin films. HPLC separations were
performed on a Gilson HPLC gradient system using 25-mL
pump heads and an ISCO V′4 variable wavelength detector.
The column employed was a Whatman Partisil 10 M20/50
(500- × 22-mm i.d.) with a flow rate of 5 mL/min and
monitoring by UV/vis detection at 340 nm. TLC chromatog-
raphy was performed on precoated 250-µm silica gel GF glass
more active (ED50 ) 1400 nM) than (all-E)-RA (ED50
)
4700 nM) in this differentiation assay. The 9Z- and 9Z,-
13Z-isomers of UAB8 and the three UAB30 isomers
were essentially inactive (ED50 >10 000 nM) at this time
point. After day 4, the (all-E)- and (13Z)-UAB8 isomers
were 2-fold more active than RA (Table 4). In contrast
to UAB8 isomers, none of the UAB30 isomers were
active. Using receptor-selective retinoids, Kizaki et al.27
and Chen et al.28 recently demonstrated that retinoid

Conformationally Defined Retinoic Acid Analogues
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 10 1685
plates (Analtech, Inc.; 5 × 10 cm). Solvents and liquid starting
materials were distilled prior to use. Reactions and purifica-
tions were conducted with deoxygenated solvents, under inert
gas (N2) and subdued lighting. Selected data for the inter-
mediates and products in Scheme 1 are contained in Table 1.
reaction mixture was partitioned between ether and water.
The ether layer was washed once with brine, dried (Na2SO4),
and concentrated under vacuum to give ester as a mixture of
mainly two configurational isomers. By this method were
prepared the following.
(2E,4E,6Z,8E)- a n d (2Z,4E,6Z,8E)-Eth yl 8-(3′,4′-Dih y-
d r o-1′(2′H)-n a p h th a len -1′-ylid en e)-3,7-d im eth yl-2,4,6-oc-
ta tr ien oa te ((9Z)- a n d (9Z,13Z)-5). This preparation em-
ployed a suspension of NaH (0.060 g, 2.5 mmol) in dry THF (1
mL), a solution of triethyl phosphonosenecioate (0.14 g, 0.53
mmol) in dry THF (1 mL), HMPA (0.041 g, 0.21 mmol), and a
solution of (9Z)-4 (0.089 g, 0.42 mmol) in dry THF (1 mL) to
give a 2:1 mixture of esters (9Z)- and (9Z,13Z)-5 (0.12 g, 78%
yield). This mixture was separated by HPLC on silica gel
using 0.5% Et2O, 0.1% THF in hexane.
(2Z,4E)-4-(3′,4′-Dih yd r o-1′(2′H)-n a p h th a len -1′-ylid en e)-
3-m eth yl-2-bu ten oic Acid ((9Z)-2). Zinc dust (1.75 g, 26.8
mmol) was stirred with 5% HCl (5 mL) for 2 min at room
temperature. The mixture was allowed to settle, and the liquid
was carefully removed by pipet. In a similar fashion the Zn
was washed, under nitrogen, with water (3 × 5 mL), acetone
(3 × 5 mL), and ether (2 × 8 mL). After residual ether was
removed under a stream of nitrogen, the flask containing the
Zn dust was strongly heated with a Bunsen burner flame for
30 s. The cooled Zn dust was suspended in anhydrous dioxane
(3 mL), and the stirred suspension was heated to reflux.
A
(2E,4E,6E,8E)- a n d (2Z,4E,6E,8E)-Eth yl 8-(3′,4′-Dih y-
d r o-1′(2′H)-n a p h th a len -1′-ylid en e)-3,7-d im eth yl-2,4,6-oc-
ta tr ien oa te ((a ll-E)- a n d (13Z)-5). This preparation em-
ployed a suspension of NaH (0.050 g, 2.1 mmol) in dry THF (1
mL), a solution of triethyl phosphonosenecioate (0.11 g, 0.40
mmol) in dry THF (1 mL), HMPA (0.048 g, 0.22 mmol), and a
solution of (all-E)-4 (0.065 g, 0.31 mmol) in dry THF (1 mL) to
give a 2:1 mixture of esters (all-E)- and (13Z)-5 (0.12 g, 78%
yield). This mixture was separated by HPLC on silica gel
using 1% Et2O, 0.5% THF in hexane.
Gen er a l P r oced u r e for th e Hyd r olysis of In d ivid u a l
Isom er s of Ester 5. To a solution of the ester 5 (1 equiv) in
methanol (final concentration 0.061 M) was added an aqueous
solution of 2 M KOH (10 equiv). The solution was heated at
reflux, and the reaction progress was monitored by TLC. After
90 min the hot solution was poured into a beaker of ice (40 g)
and acidified with 10% HCl until pH 2. The mixture was then
extracted with Et2O, which was washed with brine, dried (Na2-
SO4), and concentrated under reduced pressure to give the
product. NMR revealed that the hydrolysis occurred without
isomerization. The following acids were synthesized by this
method.
(2E,4E,6E,8E)-8-(3′,4′-Dih yd r o-1′(2′H )-n a p h t h a len -1′-
ylid en e)-3,7-d im et h yl-2,4,6-oct a t r ien oic Acid ((a ll-E)-
UAB30). This preparation utilized a solution of KOH (0.258
g, 4.61 mmol) in water (2 mL) and a warm solution of ester
(all-E)-5 (0.122 g, 0.378 mmol) in methanol (10 mL) to provide
acid (all-E)-UAB30 (0.104 g, 93% yield) as a yellow solid: mp
192-197 °C (cyclohexane).
(2E,4E,6Z,8E)-8-(3′,4′-Dih yd r o-1′(2′H )-n a p h t h a len -1′-
yliden e)-3,7-dim eth yl-2,4,6-octatr ien oic Acid ((9Z)-UAB30).
This preparation utilized a solution of KOH (0.15 g, 2.8 mmol)
in water (1 mL) and a warm solution of ester (9Z)-5 (0.073 g,
0.23 mmol) in methanol (3 mL) to give acid (9Z)-UAB30 (0.065
g, 98% yield) as a yellow solid: mp 182-185 °C (cyclohexane).
(2Z,4E,6E,8E)-8-(3′,4′-Dih yd r o-1′(2′H )-n a p h t h a len -1′-
ylid en e)-3,7-d im et h yl-2,4,6-oct a t r ien oic Acid ((13Z)-
UAB30). This preparation employed a solution of KOH (0.085
g, 1.50 mmol) in water (1 mL) and a warm solution of ester
(13Z)-5 (0.040 g, 0.12 mmol) in methanol (4 mL) to give acid
(13Z)-UAB30 (0.034 g, 93% yield) as a yellow solid: mp 180-
185 °C (cyclohexane).
solution of freshly distilled R-tetralone (0.520 g, 3.56 mmol),
ethyl bromosenecioate (1.47 g, 7.12 mmol), and anhydrous
dioxane (2.5 mL) was prepared under nitrogen, and a portion
of this solution (0.5 mL) was added to the heated Zn suspen-
sion. This produced an exothermic reaction, and the remain-
der of the solution containing R-tetralone was then added
during 10 min at a rate sufficient to control reflux. The final
reaction mixture was stirred at reflux for 8 h and then cooled
to room temperature. Water (5 mL) was added, the mixture
was stirred for 20 min, and ether (20 mL) was added. The
suspension was filtered through a pad of Celite and the filter
washed well with ether (40 mL). The filtrate was extracted
with 15% HCl (40 mL), water (40 mL), 1 N NaOH (40 mL),
and an additional amount of water (30 mL). The basic wash
and final water wash were combined, adjusted to pH 1-2 with
15% HCl, and extracted with ether (2 × 100 mL). The organic
layer was dried (Na2SO4) and concentrated under vacuum to
provide (9Z)-2 as yellow crystals (0.70 g, 86% yield): mp 153-
154 °C (1:1 hexane-ethyl acetate).
(2Z,4E)- a n d (2E,4E)-4-(3′,4′-Dih yd r o-1′(2′H)-n a p h th a -
len -1′-ylid en e)-3-m eth yl-2-bu ten -1-ol ((9Z)- a n d (a ll-E)-
3). A solution of acid (9Z)-2 (0.11 g, 0.48 mmol) in anhydrous
THF (10 mL) was cooled to -78 °C, and 1 M LiAlH4/ether (0.50
mL, 0.50 mmol) was added dropwise with stirring under
nitrogen. The dry ice-acetone bath was removed, and the
reaction mixture was brought to room temperature and stirred
for 3 h. The reaction mixture was cooled to 0 oC, and methanol
(0.2 mL) followed by 10% HCl (10 mL) was added dropwise.
This was allowed to warm to room temperature, and the
mixture was extracted with ether (2 × 20 mL). The ether layer
was washed with brine (1 × 20 mL), dried (Na2SO4), and
concentrated under vacuum to give an oily residue (0.113 g).
This was placed on a flash silica gel column (1 × 30 cm) and
eluted with 30% Et2O-hexane to give a 1:1 mixture of (9E)-
and (all-E)-3 (0.068 g, 67% yield). The alcohol mixture was
carried on in this form without further purification.
(2Z,4E)- a n d (2E,4E)-4-(3′,4′-Dih yd r o-1′(2′H)-n a p h th a -
len -1′-ylid en e)-3-m eth yl-2-bu ten a l ((9Z)- a n d (a ll-E)-4).
To a stirred solution of alcohol 3 (0.065 g, 0.30 mmol) in dry
CH2Cl2 (10 mL) at 0 °C, under nitrogen, was added 4 Å
molecular sieves (1.5 g) followed by activated MnO2 (0.53 g,
o
6.1 mmol), and the mixture was stirred at 0 C for 3 h. The
Biology. The chick skin CRABP binding assay measured
IC50 values for retinoid binding to CRABP-II using a radiola-
beled competition assay.8-10 The IC50 values for retinoids with
RARs and RXRs were measured with a radioligand competi-
tion assay.8-10 The nuclear receptor transcriptional activity
assays were performed using CV-1 cells. Transient transfec-
reaction mixture was filtered through a pad of flash silica gel,
and the filter was washed with cold 50% CH2Cl2/ether (100
mL). The filtrate was concentrated to dryness under vacuum
to give a residual oil (0.05). This was placed on a flash silica
gel column (1 × 15 cm) and eluted with 10% Et2O-hexane to
give (9Z)-4 (0.018 g), (all-E)-4 (0.016 g), and starting alcohol
3 (0.007 g). The total yield of 4 was 0.034 g (59% yield based
on unrecovered starting alcohol).
Gen er a l P r oced u r e for th e Olefin a tion of In d ivid u a l
Isom er s of Ald eh yd e 4. NaH (1.2 equiv) was washed in dry
THF three times to eliminate mineral oil. At 0 °C under N2,
freshly distilled triethyl phosphonosenecioate (1.1 equiv) in
THF was added. After the mixture stirred for 30 min, HMPA
(0.2 equiv) followed by either (9Z)- or (all-E)-4 (1 equiv, final
concentration 0.8 M) was added. After the mixture reacted
for 120 min at room temperature, water was added and the
tion of these cells with
a DNA plasmid was performed
essentially as described in Alam et al.9 The mouse skin
antipapilloma assay measured the ED50 values for retinoid
tumor inhibition on the dorsal skin of mice. The inhibition of
mouse skin papilloma by retinoids was performed according
to a modification of the procedure developed by Verma and
Boutwell14 as reported previously by Muccio et al.8
In h ibition of (a ll-E)-RA In d u ced -RARr Tr a n scr ip -
tion a l Activity in CV-1 Cells. The antagonistic effects of
UAB30 retinoids on the transcription of RARR were deter-
mined in CV-1 cells. The methods are nearly identical to those

1686 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 10
Muccio et al.
used to perform transcriptional assays. These studies utilized
the CAT reporter gene containing TREpal with a thymidine
kinase promoter [(TREpal)2-tk-CAT] as previously described
by Vaezi et al.11 Twenty-four hours after transfection, cells
were treated with (all-E)-RA (1-1000 nM) alone or together
with the indicated UAB retinoid (100 or 1000 nM). After 24
h, cells were harvested and the transcriptional effects were
measured in relative CAT activity, after correction with â-gal
activity for transfection efficiency. Similar studies were done
with a known RARR antagonist, Ro-41-5253. All experiments
were performed in triplicate, and the data were averaged.
Antagonism was defined as a lowering of the relative CAT
activity of transcription induced by (all-E)-RA in the presence
of retinoid as compared to positive controls which did not
contain retinoid. Both (all-E)-UAB30 and Ro-41-5253 exhib-
ited antagonism at each concentration of (all-E)-RA used to
induce transcription, but the effects were most noticeable when
(all-E)-RA was 50 nM.
Ack n ow led gm en t. These studies were supported by
grants PO1 CA34968 (D.D.M., W.J .B., D.L.H., B.P.S.),
AICR 93B39 (B.P.S.), RO1 CA59446 (B.P.S.), and UO1
CA60407 (P.D.E.).
Refer en ces
(1) Rosen, J .; Day, A.; J ones, T. K.; J ones, E. T. T.; Nadzan, A. M.;
Stein, R. B. Intracellular Receptors and Signal Transducers and
Activators of Transcription Superfamilies: Novel Targets for
Small-Molecule Drug Discovery. J . Med. Chem. 1995, 38, 4855-
4874.
(2) (a) Mangelsdorf, D. J .; Umesono, K.; Evans, R. M. The Retinoid
Receptors. In THE RETINOIDS Biology, Chemistry and Medi-
cine; 2nd ed.; Sporn, M. B., Roberts, A. B., Goodman, D. S., Eds.;
Raven Press: New York, 1994; pp. 319-349. (b) Gudas, L.
Retinoids and Vertebrate Development. J . Biol. Chem. 1994,
269, 15399-15402.
(3) Nadzan, A. M. Retinoids for the Treatment of Oncological
Diseases. Annu. Rep. Med. Chem. 1995, 30, 119-128.
(4) Degos, L.; Dombret, H.; Chomienne, C.; Daniel, M.-T.; Miclea,
J .-M.; Chastang, C.; Castaigne, S.; Fenauz, P. All-trans-Retinoic
Acid as a Differentiating Agent in the Treatment of Acute
Promyelocytic Leukemia. Blood 1995, 85, 2643-2653.
(5) Castleberry, R. P.; Emanuel, P. D.; Zuckerman, K. S.; Cohn, S.;
Strauss, L.; Byrd, R. L.; Homans, A.; Chaffee, S.; Nitschke, R.;
Gualtieri, R. J . A Pilot Study of Isotrentinoin in the Treatment
of J uvenile Chronic Myelogenous Leukemia. N. Engl. J . Med.
1994, 331, 1680-1684.
(6) Emanuel, P. D.; Shannon, K. M.; Castleberry, R. P. J uvenile
Myelomonocytic Leukemia: Molecular Understanding and Pros-
pects for Therapy. Mol. Med. Today 1996, 2, 468-475.
(7) Muccio, D. D.; Brouillette, W. J . (UAB Research Foundation)
Retinoid Compounds. U.S. Patent 5,094,783, March 10, 1992.
(8) Muccio, D. D.; Brouillette, W. J .; Alam, M.; Vaezi, M.; Sani, B.
P.; Venepally, P.; Reddy, L.; Li, E.; Norris, A. W.; Simpson-
Herren, L.; Hill, D. L. Conformationally Defined 6-s-trans-
Retinoic Acid Analogues. 3. Structure-Activity Relationships for
Nuclear Receptor Binding, Transcriptional Activity, and Cancer
Chemoprevention. J . Med. Chem. 1996, 39, 3625-3635.
(9) Alam, M.; Zhestkov, V.; Sani, B. P.; Venepally, P.; Levin, A. A.;
Kazmer, S.; Li, E.; Norris, A. W.; Zhang, X.-k.; Lee, M. O.; Hill,
D. L.; Lin, T.-H.; Brouillette, W. J .; Muccio, D. D. Conforma-
tionally Defined 6-s-trans-Retinoic Acid Analogues. 2. Selective
Agonists for Nuclear Receptor Binding and Transcriptional
Activity. J . Med. Chem. 1995, 38, 2303-2310.
(10) Vaezi, M. F.; Alam, M.; Sani, B. P.; Rogers, T. S.; Simpson-
Herren, L.; Wille, J . J .; Hill, D. L.; Doran, T. I.; Brouillette, W.
J .; Muccio, D. D. A Conformationally Defined 6-s-trans-Retinoic
Acid Isomer. Synthesis, Chemopreventive Activity, and Toxicity.
J . Med. Chem. 1994, 37, 4499-4507.
(11) Vaezi, M. F.; Robinson, C. Y.; Hope, K. D.; Brouillette, W. J .;
Muccio, D. D. Preparation of the 9-cis, 13-cis, and All trans-
Isomers of R- and â2-Retinal. Org. Prep. Proc. Int. 1987, 19, 187-
195.
(12) Hale, R. L.; Burger, W.; Perry, C. W.; Liebman, A. A. Preparation
of High Specific Activity All Trans-R-Retinyl-11-3H Acetate. J .
Labeled Compds. Radiopharm. 1977, 13, 123-135.
(13) Zhang, X.-k.; Hoffmann, B.; Tran, P.; Graupner, G.; Pfahl, M.
Retinoid X Receptor is an Auxiliary Protein for Thyroid Hormone
and Retinoic Acid Receptors. Nature (London) 1992, 355, 441-
446.
In d u ction of Differ en tia tion of HL-60 a n d NB4 Cells.
Human leukemia cell lines HL-60 and NB4 were used for
measuring the ability of retinoids to induce terminal dif-
ferentiation. Both cell lines are predominantly promyelocytes
and are induced by RA to mature to cells with many charac-
teristics of mature granulocytes. The induction of this dif-
ferentiation is the basis for assaying one of the biological
activities of RA analogues. For the differentiation assay, HL-
60 or NB4 cells were grown in nutrient medium with fetal
bovine serum as detailed previously by Breitman.15 The UAB
retinoids and RA were prepared in ethanol at concentrations
from 0.01 to 10 mM and then diluted 100-fold into cell cultures.
After 3- and 4-day incubation periods, the ability of the cells
to reduce nitroblue tetrazolium was measured. The percent-
age of cells that gained this marker was measured microscopi-
cally. The percentage of positive cells was then plotted as a
function of the retinoid concentration to determine ED50
values.
In h ibition of J MML Sp on ta n eou s CF U-GM Colon y
F or m a tion . (all-E)- and (9Z)-RA, UAB8, and UAB30 retin-
oids were tested for their ability to inhibit spontaneous CFU-
GM colony growth in three J MML patient samples as de-
scribed in detail elsewhere by Emanuel et al.5,16 Briefly, after
obtaining parental consent and with the approval of the
Institutional Review Board, J MML peripheral blood and bone
marrow samples were collected from patients and shipped by
overnight delivery to the University of Alabama at Birming-
ham. Mononuclear cells were separated by density gradient
centrifugation, washed, and frozen in aliquots. Aliquots of
J MML patient samples were later thawed, washed, and set
up in 0.3% soft agar clonal assays in McCoys’ 5A medium
supplemented with nutrients as well as 15% fetal bovine
serum. No growth factors or other exogenous stimuli were
added to the cultures. UAB retinoids and RA isomers were
dissolved in either 100% ethanol or DMSO. Appropriate
dilutions were made, and controls using the respective ethanol
or DMSO concentration were set up in parallel. Varying doses
of retinoids were added only once, 24 h after initiation of the
soft agar assays, by pipetting the retinoid solution on top of
the soft agar and allowing the solution to enter the agar and
interact with the cells by diffusion. The cultures were
incubated for 14 days at 37 °C in a humidified atmosphere
with 5% CO2. CFU-GM colonies (1 colony has >40 cells) were
scored, and the resultant amount of inhibition of spontaneous
CFU-GM colony growth was compared to the (13Z)-RA.
(14) Verma, A.; Boutwell, R. Vitamin A Acid (Retinoic Acid), a Potent
Inhibitor of 12-O-Tetradecanoylphorbol-13-acetate-induced Or-
nithine Decarboxylase Activity in Mouse Epidermis. Cancer Res.
1977, 37, 2196-2201.
(15) (a) Breitman, T. R. Growth and Differentiation of Human
Myeloid Leukemia Cell Line HL60. Methods Enzymol. 1990, 190,
118-130. (b) Taimi, M.; Breitman, T. R. Growth, Differentiation,
and Death of Retinoic Acid-Treated Human Acute Promyelocytic
Leukemia NB4 Cells. Exp. Cell Res. 1997, 230, 69-75.
(16) (a) Emanuel, P. D.; Bates, L. J .; Zhu, S.-W.; Castleberry, R. P.;
Gualtieri, R. J .; Zuckerman, K. S. The Role of Monocyte-Derived
Hemopoietic Growth Factors in the Regulation of Myeloprolif-
eration in J uvenile Chronic Myelogemous Leukemia. Exp. He-
matol. 1991, 19, 1017-1024. (b) Emanuel, P. D.; Bates, L. J .;
Zhu, S.-W.; Castleberry, R. P.; Gualtieri, R. J .; Zuckerman, K.
Selective Hypersensitivity to Granulocyte-Macrophage Colony-
Stimulating Factor by J uvenile Chronic Myeloid Leukemia
Hematopoietic Progenitors. Blood, 1991, 77, 925-929.
(17) Lake, C. H.; Alam, M. A.; Muccio, D. D.; Brouillette, W. J . A
Structural Model for a New Class of Conformationally Con-
strained Retinoid: (2Z,4E)-4-[3′, 4′-dihydro-1′(2′H)-naphthalene-
1′-ylidene]-2-butenoic acid. J . Chem. Crystallogr. 1997, 27, 231-
235.
Molecu la r Mod elin g. Retinoid structures were generated
with Sybyl version 6.2 (Tripos Inc., St. Louis, MO) on a Silicon
Graphics Indigo 2 workstation. The structure of (9Z)-2 was
built using the recently reported X-ray crystal structure.17 This
structure crystallized in the 8-s-trans-conformation. To gener-
ate 8-s-cis-conformations, this structure was rotated about the
C8-C9 bond followed by energy minimization using Allinger’s
MM3(94) force fields. The structures of final UAB30 acids
were generated from these 8-s-cis- and 8-s-trans-structures of
2. The thermodynamic parameters were calculated according
to previously described methods.8

Conformationally Defined Retinoic Acid Analogues
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 10 1687
(18) (a) Sani, B. P.; Titus, B. C.; Banerjee, C. K. Determination of
Binding Affinities of Retinoids to Retinoic Acid-Binding Protein
and Serum Albumin. Biochem. J . 1978, 171, 711-717. (b) Sani,
B. P. Cellular Retinoic Acid-Binding Protein and the Action of
Retinoic Acid. In Chemistry and Biology of Synthetic Retinoids;
Dawson, M. I., Okamuar, W. H., Eds.; CRC Press: Boca Raton,
FL, 1990; pp 365-384.
(19) Graupner, G.; Wills, K. N.; Tzukerman, M.; Zhang, X.-k.; Pfahl,
M. Dual Regulatory Role for Thyroid Hormone Receptors Allows
Control of Retinoic Acid Receptor Activity. Nature (London)
1989, 340, 653-656.
(20) (a) Chandraratna, R. A. S.; Gillett, S. J .; Song, T. K.; Attard, J .;
Vuligonda, S.; Garst, M. E.; Arefieg, T.; Gil, D. W.; Wheeler, L.
Synthesis and Pharmacological Activity of Conformationally
Restricted, Acetylenic Retinoid Analogues. BioMed. Chem. Lett.
1995, 5, 523-527. (b) Nagpal, S.; Athanikar, J .; Chandraratna,
R. A. S. Separation of Transcription and AP1 Antagonism
Functions of Retinoic Acid Receptor R. J . Biol. Chem. 1995, 270,
923-927.
(21) Castleberry, R. P.; Emanuel, P.; Gualtieri, R.; Zuckerman, K.;
Cohn, S.; Strauss, L.; Byrd, R.; Homans, A.; Chaffee, S.;
Nitshcke, R. Preliminary Experience with 13-Cis Retinoic Acid
(CRA) in the Treatment of J uvenile Chronic Myelogenous
Leukemia (J CML). Blood 1991, 78 (Suppl. 1), 170a.
(22) (a) Lapidot, T.; Cohen, A.; Grunberger, T.; Dick, J .; Freedman,
M. H. Aberrant Growth Properties of J uvenile Chronic Myelog-
enous Leukemia (J CML) CD34+ Cells in Vitro and in Vivo using
SCID Mouse Assays, Blood 1993, 82 (Suppl. 1), 197a. (b)
Cambier, N.; Menot, M. L.; Fenaux, P.; Wattel, E.; Baruchel,
A.; Chomienne, C. GM-CSF Hypersensitivity in CD34+ Purified
Cells in J uvenile and Adult Chronic Myelomonocytic Leuke-
mia: Effects of Retinoids. Blood 1995, 86 (Suppl 1), 791a.
(23) Emanuel, P. D.; Sokol, J . M.; Castleberry, R. P. Characterization
of Early Response Gene Expression in J uvenile Myelomonocytic
Leukemia Syndrome (J MML). Blood 1995, 86 (Suppl 1), 728a.
(24) (a) Yang Yen, H.-F.; Zhang, X.-K.; Graupner, G.; Tzukerman,
M.; Sakamoto, B.; Karin, M.; Pfahl, M. Antagonism between
Retinoic Acid Receptors and AP-1: Implications for Tumor
Promotion and Inflammation. New Biol. 1991, 3, 1206-1219.
(b) Schule, R.; Rangarajan, P.; Yang, N.; Kliewer, S.; Ransone,
L. J .; Bolado, J .; Verma, I. M.; Evans, R. M. Retinoic acid is a
negative regulator of AP-1-responsive genes. Proc. Natl. Acad.
Sci. U.S.A. 1991, 88, 6092-6096. (c) Pfahl, M. Nuclear Receptor/
AP-1 Interactions. Endocrine Rev. 1993, 14, 651-658.
(25) Salbert, G.; Fanjul, A.; Piedrafita, F. J .; Lu, X. P.; Kim, S.-J .;
Tran, P.; Pfahl, M. Retinoic Acid Receptors and Retinoid X
Receptor-R Down-Regulate the Transforming Growth Factor-â1
Promoter by Antagonizing AP-1 Activity. Mol. Endocrinol. 1993,
7, 1347-1356.
(26) (a) Fanjul, A.; Dawson, M. I.; Hobbs, P. D.; J ong, L.; Cameron,
J . F.; Harlev, E.; Graupner, G.; Lu, X.-P.; Pfahl, M. A New Class
of Retinoids with Selective Inhibition of AP-1 Inhibits Prolifera-
tion. Nature 1994, 372, 107-111. (b) Nagpal, S.; Athanikar, J .;
Chandraratna, R. A. S. Separation of Transactivation and AP1
Antagonism Functions of Retinoic Acid Receptor R. J . Biol.
Chem. 1995, 270, 923-927.
(27) (a) Kizaki, M.; Nakajima, H.; Mori, S.; Koike, T.; Morikawa, M.;
Ohta, M.; Saito, M.; Koeffler, H. R.; Ikeda, Y. Novel Retinoic
acid, 9-Cis Retinoic Acid, in Combination with All-Trans Retinoic
Acid Is an Effective Inducer of Differentiation of Retinoic Acid-
Resistance HL-60 Cells. Blood 1994, 83, 3289-3297. (b) Kizaki,
M.; Dawson, M. I.; Heyman, R.; Eistner, E.; Morosetti, R.;
Pakkala, S.; Chen, D.-L.; Ueno, H.; Chao, W.-r.; Morikawa, M.;
Ikeda, Y.; Heber, D.; Pfahl, M.; Koeffler, H. R. Effects of Novel
Retinoid X Receptor-Selective Ligands on Myeloid Leukemia
Differentiation and Proliferation in Vitro. Blood 1996, 87, 1977-
1984.
(28) Chen, J .-Y.; Clifford, J .; Zusi, C.; Starrett, J .; Tortolani, D.;
Ostrowski, J .; Reczek, P. R.; Chambon, P.; Gronemeyer, H. Two
Distinct Actions of Retinoid-Receptor Ligands. Nature 1996, 382,
819-822.
(29) Lin, T.-H.; Rogers, T. S.; Hill, D. H.; Simpson-Herren, L.; Farnell,
D. R.; Kochhar, D. M.; Alam, M.; Brouillette, W. J .; Muccio, D.
D. Murine Toxicology and Pharmacology of UAB-8, a Confor-
mationally Constrained Analogue of Retinoic Acid. Toxicol. Appl.
Pharmacol. 1996, 139, 310-316.
J M970635H