Heteroarotinoids inhibit head and neck cancer cell lines in vitro and in vivo through both RAR and RXR retinoic acid receptors

Article abstract of DOI:10.1021/jm990292i

A class of less toxic retinoids, called heteroarotinoids, was evaluated for their molecular mechanism of growth inhibition of two head and neck squamous cell carcinoma (HNSCC) cell lines SCC-2 and SCC-38. A series of 14 heteroarotinoids were screened for growth inhibition activity in vitro. The two most active compounds, one that contained an oxygen heteroatom (6) and the other a sulfur heteroatom (16), were evaluated in a xenograph model of tumor establishment in nude mice. Five days after subcutaneous injection of 10⁷ SCC-38 cells, groups of 5 nu/nu mice were gavaged daily (5 days/week for 4 weeks) with 20 mg/kg/day of all-trans-retinoic acid (t-RA, 1), 10 mg/kg/day of 6, 10 mg/kg/day of 16, or sesame oil. After a few days, the dose of t-RA (1) was decreased to 10 mg/kg/day to alleviate the side effects of eczema and bone fracture. No significant toxic effects were observed in the heteroarotinoid groups. All three retinoids caused a statistically significant reduction in tumor size as determined by the student t-test (P < 0.05). Complete tumor regression was noted in 3 of 5 mice treated with t-RA (1), 4 of 5 mice treated with 16, 1 of 5 mice treated with 6, and 1 of 5 mice treated with sesame oil. Reverse transcriptase polymerase chain reaction (RT- PCR) was used to determine that the expression levels of RARα, RXRα, and RXRβ were similar in the two cell lines, while RARβ expression was higher in SCC-2 over SCC-38, and RARγ expression was higher in SCC-38 over SCC-2. Receptor cotransfection assays in CV-1 cells demonstrated that 16 was a potent activator of both RAR and RXR receptors, while 6 was selective for the RXR receptors. Transient cotransfection assays in CV-1 cells using an AP-1 responsive reporter plasmid demonstrated that t-RA (1), 6, and 16 each inhibited AP-1-driven transcription in this cell line. In conclusion, the growth inhibition activity of the RXR-selective 6 and the more potent growth inhibition activity of the RAR/pan-agonist 16 implicate both RARs and RXRs in the molecular mechanism of retinoid growth inhibition. Moreover, the chemoprevention activity and the lack of toxicity of heteroarotinoids demonstrate their clinical potential in head and neck cancer chemoprevention.

Full text of DOI:10.1021/jm990292i

4434
J . Med. Chem. 1999, 42, 4434-4445
Heter oa r otin oid s In h ibit Hea d a n d Neck Ca n cer Cell Lin es in Vitr o a n d in Vivo
Th r ou gh Both RAR a n d RXR Retin oic Acid Recep tor s
David Zacheis,† Arindam Dhar,§ Shennan Lu,|,§ Matora M. Madler,‡ J ozef Klucik,‡ Chad W. Brown,‡
Shengquan Liu,‡ Francis Clement,‡ Shankar Subramanian,‡ G. Mahika Weerasekare,‡ K. Darrell Berlin,*,‡
Michael A. Gold,§ J ohn R. Houck, J r.,† Kenneth R. Fountain,3 and Doris M. Benbrook*,§,|
Departments of Otorhinolaryngology, Obstetrics and Gynecology, and Biochemistry and Molecular Biology,
University of Oklahoma Health Sciences Center, P.O. Box 26901, Oklahoma City, Oklahoma 73190,
Department of Chemistry, Oklahoma State University, Stillwater, Oklahoma 74078, and Department
of Chemistry, Truman State University, Kirksville, Missouri 63501
Received J une 8, 1999
A class of less toxic retinoids, called heteroarotinoids, was evaluated for their molecular
mechanism of growth inhibition of two head and neck squamous cell carcinoma (HNSCC) cell
lines SCC-2 and SCC-38. A series of 14 heteroarotinoids were screened for growth inhibition
activity in vitro. The two most active compounds, one that contained an oxygen heteroatom (6)
and the other a sulfur heteroatom (16), were evaluated in a xenograph model of tumor
establishment in nude mice. Five days after subcutaneous injection of 107 SCC-38 cells, groups
of 5 nu/ nu mice were gavaged daily (5 days/week for 4 weeks) with 20 mg/kg/day of all-trans-
retinoic acid (t-RA, 1), 10 mg/kg/day of 6, 10 mg/kg/day of 16, or sesame oil. After a few days,
the dose of t-RA (1) was decreased to 10 mg/kg/day to alleviate the side effects of eczema and
bone fracture. No significant toxic effects were observed in the heteroarotinoid groups. All three
retinoids caused a statistically significant reduction in tumor size as determined by the Student
t-test (P < 0.05). Complete tumor regression was noted in 3 of 5 mice treated with t-RA (1), 4
of 5 mice treated with 16, 1 of 5 mice treated with 6, and 1 of 5 mice treated with sesame oil.
Reverse transcriptase polymerase chain reaction (RT-PCR) was used to determine that the
expression levels of RARR, RXRR, and RXRâ were similar in the two cell lines, while RARâ
expression was higher in SCC-2 over SCC-38, and RARγ expression was higher in SCC-38
over SCC-2. Receptor cotransfection assays in CV-1 cells demonstrated that 16 was a potent
activator of both RAR and RXR receptors, while 6 was selective for the RXR receptors. Transient
cotransfection assays in CV-1 cells using an AP-1 responsive reporter plasmid demonstrated
that t-RA (1), 6, and 16 each inhibited AP-1-driven transcription in this cell line. In conclusion,
the growth inhibition activity of the RXR-selective 6 and the more potent growth inhibition
activity of the RAR/RXR pan-agonist 16 implicate both RARs and RXRs in the molecular
mechanism of retinoid growth inhibition. Moreover, the chemoprevention activity and the lack
of toxicity of heteroarotinoids demonstrate their clinical potential in head and neck cancer
chemoprevention.
In tr od u ction
Retinoids have been known to elicit useful pharma-
cological responses in a variety of disorders.1 Activities
of endogenous retinoids such as trans-retinoic acid (t-
RA, 1), 9-cis-retinoic acid (9-c-RA, 2), and 13-cis-retinoic
acid (13-c-RA, 3) are well-known in cancer therapy,
although all have some toxic characteristics.2 The
synthetic retinoid 4-(hydroxyphenyl)retinamide (4,
4-HPR), currently under clinical investigation, is also
known to possess some undesirable properties such as
night blindness, although it appears to be dose-depend-
ent and reversible.3
Certain heteroarotinoids have demonstrated antican-
the growth inhibitory activity of a variety of hetero-
cer activity as well as low toxicity.4-8 We now report
arotinoids 5-19 (Chart 1) against head and neck
squamous cell carcinoma (HNSCC) cell lines, SCC-2 and
SCC-38. The heteroarotinoids possess a range of flex-
ibility in terms of the linking groups attached to the aryl
ring and, in most cases, between two aryl rings. It was
reasoned that such flexibility might better allow accom-
modation of the molecule in the receptor site and
* Corresponding authors. For biology: e-mail, Doris-Benbrook@
OUHSC-edu. For chemistry: e-mail, kdberlin@bmb-fs1.biochem.
okstate.edu.
† Department of Otorhinolaryngology, OUHSC.
§ Department of Obstetrics and Gynecology, OUHSC.
| Department of Biochemistry and Molecular Biology, OUHSC.
‡ Department of Chemistry, OSU.
3 Department of Chemistry, TSU.
10.1021/jm990292i CCC: $18.00 © 1999 American Chemical Society
Published on Web 09/21/1999

Heteroarotinoids Inhibit Cancer Cell Lines
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 21 4435
Ch a r t 1
therefore lead to an activated complex of ligand-
receptor. The area of ligand-receptor binding has been
reviewed.9
when treated with 13-c-RA (3) for a median follow-up
of 54 months.17 Despite these encouraging results, a
large number of relapses have been detected upon
cessation of certain agents, and there have been numer-
ous side effects, especially at high doses. These side
effects include mucocutaneous dryness, dermatitis, con-
junctivitis, photodermatosis, impaired night vision, hy-
pertriglyceridemia, and transient increases in serum
liver function enzymes. Efforts to prevent relapses and
to reduce toxicity have included co-treatment with
R-tocopherol and the development of synthetic reti-
noids.7,18
Prior to clinical application, the chemoprevention
activities of retinoids are usually evaluated against
HNSCC cell lines in vitro and in animal models in vivo.
Although vitamin A is the least toxic of the natural
retinoids, it is also the weakest inhibitor of growth in
HNSCC cell lines in vitro and in xenograph growth in
vivo, followed by 13-c-RA (3).19-24 The reduced toxicity
of certain heteroarotinoids7 prompted an evaluation in
the present study to determine if such heterocycles
inhibited the growth of HNSCC cell lines in vitro and
in vivo and to explore the molecular mechanism of
action of such agents.
Head and neck cancers often arise from a field of
epithelial injury caused by carcinogens in tobacco smoke
which, in turn, leads to multiple premalignant lesions
and the risk of primary and second primary tumor
formation.10,11 Retinoids have been shown to reverse oral
premalignant lesions and prevent second primaries in
patients treated for head and neck cancers. The use of
retinoids for the prevention and treatment of head and
neck cancers was inspired by the observations that
vitamin A deficiency caused keratinizing squamous
metaplasia in mucociliary epithelium and that replen-
ishment of retinoids restored normal epithelial dif-
ferentiation.12,13 Leukoplakia and erythroplakia are
premalignant head and neck mucosal lesions. Several
studies have attempted to use retinoids to prevent or
reverse these lesions back to normal mucosa in high-
risk patients. In a low-dose experiment that showed
minimal toxicity to the patients, the natural retinoid
13-c-RA (3) was found to be superior to â-carotene in
reducing the rate of oral leukoplakia progression.14 The
synthetic retinoid 4-HPR (4) has been shown to decrease
relapses and new localizations of surgically treated oral
leukoplakia from 30% to 6%.15
Resu lts a n d Discu ssion
Many head and neck epithelial malignancies are
multifocal, and the development of a second primary
tumor either during treatment or after successful treat-
ment for a primary tumor occurs at a rate of 4-7%/
year.16 A population of these patients had significant
reductions in development of second primary tumors
Ch em istr y. The discussion of the chemistry is in an
order with respect to various properties of the selected
heteroarotinoids. Known activities of 5,7 6,7 12,6 14,4 and
157 in certain assays prompted us to make a change to
a somewhat different flexible linking group as in 7.
Although some amide linking groups have been exam-

4436 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 21
Zacheis et al.
Sch em e 1
The use of a combination of LiOCl/Clorox was ultimately
successful. Esterification of 32 required a long reaction
time to obtain 10, apparently due to the increased steric
hindrance around the carbonyl group of 32.
In view of the known toxicity of TTNPB,7 the synthe-
sis of the close analogue 11 (Scheme 5) would also serve
as a model for comparison of activity with that of
heteroarotinoid 10. Acetylation of known29 33 also was
demanding in terms of avoiding or minimizing cleavage
of the ether linkage in 30. Using the conditions outlined
resulted in the generation of 34 (75%). As with the
conversion of 31 to 32, generation of acid 35 from ketone
34 was slow because of hindrance, although 35 was
isolated in moderate yield (44%). Esterification of 35 to
produce 11 (67%) required a long reaction time as
anticipated.
The unusual heteroarotinoid 13 was selected for
inclusion in the study since it possessed a rigid long side
chain and contained two heteroatoms in the fused ring
unit. Other heteroarotinoids with long side chains had
exhibited strong activity in several assays.5,6 The influ-
ence of two heteroatoms could also be assessed. The
addition of a nearly 3-fold excess of vinylmagnesium
bromide to commercial 36 resulted in essentially a
quantitative yield of slightly crude 37 (Scheme 6).
Conversion of the vinyl alcohol 37 to the phosphonium
salt 38 (85%) was effected by conditions previously
developed.5,7 Generation of the Wittig reagent of 38 as
shown was followed by the addition of ethyl 3-methyl-
4-oxocrotonate in a normal condensation to provide
heteroarotinoid 13, albeit in modest yield (28.8%).
The flexible heteroarotinoid 16 could be obtained via
the conversion of 39 f 40 f 41 f 16 as illustrated
(Scheme 7). The nitration of known5 39 afforded the
6-isomer 40 in modest yield (27%) since other nitrated
products appeared to have formed. Reduction of the
nitro group in 40 to the amino function in 41 (50%)
progressed reasonably well under mild conditions. N-
Acylation of 41 with the acid chloride shown produced
the target 16 (72%), again under quite mild conditions.
Biology. 1. In Vitr o Gr ow th In h ibition Scr een .
A series of retinoids were screened for growth inhibition
activity in the SCC-2 and SCC-38 HNSCC cell lines. The
activities of 14 heteroarotinoids containing a sulfur
heteroatom (14-16), an oxygen heteroatom (5-10, 12,
13), or a nitrogen heteroatom (17-19), as well as a
structurally related arotinoid (11) void of a heteroatom,
were compared to that of 9-c-RA (2) (Table 1). Although
most of the compounds inhibited growth, some of them
stimulated growth. Three of the compounds with oxygen
heteroatoms (5, 8, 9) had opposite effects in the two cell
lines. The Student t-test was used to demonstrate that
there were no statistically significant differences be-
tween the responses of the two cell lines to the entire
set of retinoids (P > 0.05) or between the activities of
groups with different types of heteroatoms (P > 0.05).
The oxygen heteroarotinoid 6 and sulfur heteroarotinoid
16 were the most active in both cell lines and were
therefore chosen for evaluation in a xenograft model.
ined for useful activity25 and even for RARR selectivity,26
no record of the use of a C(O)-N-methoxy linker could
be found in the literature. The excess bulk of this group
was considered as altering the rotational barrier of the
C-N bond and thereby possibly influencing receptor
binding and activation.
Generation of the acid chloride from known 208 and
subsequent treatment with the appropriate hydroxyl-
amine as shown gave 21 (Scheme 1). O-Methylation
proceeded under mild conditions to yield 7 (48%).
Creating and assessing ester 8 for useful activity was
considered a test of the influence on activity by flanking
methyl groups of the aryl ring with respect to freedom
of rotation of the internal ester linkage. In addition, it
is known5,7 that alkyl groups adjacent to the heteroatom
have a positive influence on biological activity. Thus,
ester 8 possesses both such features. Excess methyl-
magnesium bromide added to known 2227 in boiling
ether over 48 h led to lactol 23 (Scheme 2). Dehydration
of 23 under standard conditions yielded ether 24. Direct
esterification of 24 gave new ester 8 (50%) which was
purified by chromatography.
Surprisingly, when 22 was treated with the same
excess of methyl Grignard reagent as used to obtain 23,
but in THF, ether 25 (76%) formed (Scheme 3). Perhaps
the much longer reaction time was significant, but the
solvent was found to be important as well. Esterification
of 25 under the usual conditions produced the target
ester ether 9 (52%). This heteroarotinoid 9 possessed
the geminal methyl flanking group that enhanced
activity in several assays in related systems which
contained more rigid linking groups attached to the aryl
ring.5 Consequently, an assessment could be made of
the alteration in activity between 8 and 9.
Retaining the flexible ester linkage between the two
aryl rings was desirable in 10. In part, the rationale
for 10 was based on the strong activity of (E)-4-[2-
(5,6,7,8-tetrahydro-5,5,8,8-tetramethyl-2-naphthalenyl)-
1-propenyl]benzoic acid (TTNPB), although it is very
toxic.7 The incorporation of an oxygen atom into system
10 might reduce toxicity sufficiently to be tolerable and
yet retain useful activity. Dimethylation of 26 as shown
was facile and gave 27 in high yield (84%) (Scheme 4).
Base-catalyzed hydrolysis of 27 led to acid 28 (69%), a
reported compound28 but not previously well-character-
ized. Reduction of 28 to 29 was straightforward, and
alcohol 29 was cyclized with the incorporation of an
acetone unit to yield ether 30 (75%). Acetylation of 30
proved more difficult than anticipated, but the use of
H3CNO2 with AlCl3 and AcCl was successful and gave
31 (51%) without cleavage of the methoxy ether group.
Oxidative cleavage of the acetyl group in 31 to give acid
32 gave only a modest yield (21%) due to side reactions.
2. Xen ogr a ft Mod el. The 6 and 16 compounds were
evaluated for biological activity in a xenograft nude
mouse model. Athymic nu/nu mice were injected sub-
cutaneously with 107 SCC-2 or SCC-38 cells. The effects
of retinoids in mice injected with SCC-2 cells could not

Heteroarotinoids Inhibit Cancer Cell Lines
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 21 4437
Sch em e 2
Sch em e 3
Sch em e 4
Sch em e 5
Tumors developed from the SCC-38 cell line under-
went a reduction in volume during days 9-14, followed
by a stabilization of tumor size (Figure 1, panels B-D).
The tumors in the groups treated with t-RA (1) and 16
stabilized starting on days 12 and 14, respectively, at
sizes that were significantly smaller than the tumors
in the control group (P < 0.05, t-test), which were also
stable during this period. The sizes of the tumors in the
group treated with 6 were not significantly smaller than
the tumors of the control group until day 26 (P <0.05,
t-test). The last day of tumor measurements demon-
strated growth in the control group compared to con-
tinued stabilization or further reduction in the treat-
ment groups. The average tumor volumes of the four
treatment groups were significantly different from each
other after day 23 as determined by ANOVA (P < 0.05).
Complete tumor regression was observed in 3 of 5
mice treated with t-RA (1), 4 of 5 mice treated with 16,
and 1 of 5 mice treated with 6. In the control group, 1
of 5 sesame oil-treated mice had complete tumor regres-
sion.
3. R ecep t or E xp r ession . Retinoids regulate gene
expression through two classes of nuclear retinoic acid
receptors, RAR and RXR, each of which have three
isoforms labeled R, â, and γ.9 The expression patterns
of retinoic acid receptors in SCC-2 and SCC-38 were
evaluated by reverse transcriptase polymerase chain
reaction (RT-PCR). Both cell lines expressed RARR,
RXRR, and RXRâ at similar levels (Figure 2). The
intensities of the âactin control bands for each cell line
were similar, but the intensities of the RARâ and RARγ
be evaluated because these mice developed very large
tumors and died rapidly of disease (data not shown).
Mice injected with SCC-38, however, developed stable
tumors and survived the duration of the month-long
experiment. Mice bearing SCC-38 xenografts were
arbitrarily divided into 4 treatment groups of 5 mice
each and were gavaged with 0.1 mL of sesame oil
containing either 20 mg/kg/day of t-RA (1), 10 mg/kg/
day of 6, 10 mg/kg/day of 16, or no drug control. After 1
week of treatment, the dose of t-RA (1) was decreased
to 10 mg/kg/day due to excessive eczema and bone
fractures in this group of mice. These toxicities were
not evident in the other three treatment groups. There
was no significant difference in the animal weights
between the treatment groups (Figure 1, panel A).

4438 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 21
Zacheis et al.
Sch em e 6
Sch em e 7
strate that the concentrations needed to achieve this
activation were above pharmacologically obtainable
levels.
5. An ti-AP -1 Activity. The ligand-bound retinoic acid
receptors have been shown to repress transcriptional
activation by AP-1 transcription factors.30,31 This anti-
AP-1 activity has been implicated in the mechanism of
growth inhibition.32,33 The ability of t-RA (1), 9-c-RA (2),
6, and 16 to repress AP-1-driven transcription was
evaluated using transient transfection assays of an AP-
1-driven reporter plasmid into SCC-2 cultures (Figure
3). All the retinoids repressed AP-1 activity, and 16
exhibited the greatest repression. Similar experiments
using SCC-38 cultures could not be performed due to
the poor transfection efficiency of this line.
Ta ble 1. Percent in Vitro Growth Inhibition of Two HNSCC
Cell Lines, SCC-2 and SCC-38, by a 3-Day Exposure to 10 µM
Heteroarotinoids and 9-c-RA (2)
Con clu sion s
atom
compd
SCC-2
SCC-38
In this report, the potential of certain heteroarotinoids
for chemoprevention of head and neck cancers was
confirmed by growth inhibition of HNSCC cell lines in
vitro and prevention of tumor establishment in a
xenograph nude mouse model in vivo. The SCC-38
xenograph tumors in this study exhibited a reduction
during days 9-14, followed by a stabilization of tumor
size (Figure 1). The curves in all four groups were
remarkably similar for days 9-14, suggesting that this
initial regression in size was due to factors intrinsic to
the xenograph model such as absorption of diluent and
nonviable tumor cells and resolution of the host’s
inflammatory response to tumor injection. The last
tumor measurement in the control group, however,
demonstrated significant growth compared to continued
stabilization or further reduction in the treatment
groups. This implies that the model is reflective of tumor
establishment rather than tumor growth. Establish-
ment of tumor implants appears to have occurred over
the initial 28 days following injection, while growth of
the established tumors did not occur until after day 28.
The efficacy of the two heteroarotinoids in prevention
of the establishment of SCC-38 xenograph tumors is
indicated by the significant decrease in size of the
tumors in the groups treated with 6 and 16. The
complete regression of 4 out of 5 xenograph tumors in
the group treated with 16, in comparison to 1 out of 5
in the control group, is a more convincing demonstration
of heteroarotinoid chemoprevention activity.
O
O
O
O
O
O
O
O
S
S
S
N
N
N
5
6
7
8
9
10
12
13
14
15
16
17
18
19
11
-6
19
3
15
-1
52
1
13
18
17
-8
9
6
1
-2
1
17
37
13
7
15
12
10
-33
2
10
64
12
9
5
15
13
9-c-RA (2)
bands for each cell line varied, indicating that SCC-2
expressed higher levels of RARâ in comparison to SCC-
38 and that SCC-38 expressed higher levels of RARγ in
comparison to SCC-2.
4. Nu clea r Retin oid Recep tor Activa tion . While
t-RA (1) and arotinoid compounds only activate the RAR
receptors, 9-c-RA (2) and some heteroarotinoids activate
both RAR and RXR receptors.9 To determine the speci-
ficity of retinoic acid receptor activation by 6 and 16
cotransfection assays using the RARâ RARE-tk-CAT
reporter plasmid were performed. All six nuclear recep-
tors were activated by 6 and 16 in a dose-responsive
manner. The concentration that induced half-maximal
activity (EC50 value) was derived as a measure of
receptor potency, and the maximal activity relative to
9-c-RA (2) was derived as a measure of efficacy for each
receptor (Table 2). Heteroarotinoid 16 was a potent
activator of all six receptors, while 6 was selective for
the RXR receptors. Although the efficacy of RAR activa-
tion by 6 ranged from 24-58% of the activation by 9-c-
RA (2), the greater than 10 000 nM potencies demon-
The molecular mechanism of this growth inhibition
most likely involves the regulation of gene expression
through the nuclear retinoic acid receptors. In support
of this hypothesis, the use of t-RA (1) to treat squamous
cell carcinoma HNSCC cell lines resulted in decreased
expression of two genes that are associated with in-
creased proliferation and that are commonly overex-

Heteroarotinoids Inhibit Cancer Cell Lines
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 21 4439
F igu r e 1. Xenograft model of retinoid-induced tumor shrinkage. An amount of 107 SCC-38 cells was injected into the flanks of
18 nude mice. Five days later, 10 mg/kg of t-RA (1) (9), 6 (b), 16 (2), or sesame oil (O) was administered by gavage for 5 consecutive
days/week. Each mouse was weighed 3 times/week, and the average weights for each group are shown in panel A. Tumor volumes
were measured with calipers. The data in panels B-D represent the average and standard deviations of tumor sizes of 6 mice/
group treated with t-RA (1), 6, or 16, respectively. Retinoid treatment caused a statistically significant decrease in tumor size
starting at day 12 for t-RA (P ) 0.025), day 26 for 6 (P ) 0.051), and day 14 for 16 (P ) 0.016) as determined by the Student
T-test.
pressed in head and neck cancers, namely the trans-
forming growth factor R (TGF-R) and the epidermal
growth factor receptor (EGF-R).34 The growth inhibition
activities of compounds, such as t-RA (1), that activate
all RAR receptors are not likely to be strongly affected
by alterations in the pattern of receptor expression. In
agreement with this suggestion, a study using a panel
of six HNSCC cell lines revealed that receptor expres-
sion patterns were unrelated to a response to t-RA (1)
treatment.21 In contrast, the activities of receptor-
selective retinoids can be affected by the receptor
expression patterns in head and neck lesions. An
investigation of premalignant, dysplastic, and malig-
nant head and neck cancers found that RARâ expression
was suppressed in more than 50% of all lesions.23
However, RARγ expression was not suppressed and has
been implicated in squamous differentiation and t-RA
response in HNSCC.35,36 The role of RARγ in the
mechanism of retinoid growth inhibition in HNSCC cell
lines has been illustrated by the use of sense and
antisense RARγ expression in the SqCC/Y1 line.36 While
t-RA (1) treatment and sense RARγ expression caused
growth inhibition, AP-1 repression and decreased ex-
pression of EGF-R and of squamous differentiation
markers, anti-sense RARγ, antagonized these activities.
Therefore, the potencies of RARâ and RARγ activation
by 16 (EC50’s of 1004 and 152 nM, respectively) are
amenable to use against head and neck lesions that
have decreased expression of RARâ and consistent
expression of functional RARγ.
Both SCC-2 and SCC-38 cell lines evaluated in this
study expressed all of the human RAR and RXR nuclear
retinoic acid receptors and are therefore susceptible to
all retinoic acid receptor-selective compounds. The
potent growth inhibition activity of 6, which is selective
for the RXR receptors, suggests that RXR activation is
sufficient for growth inhibition of SCC-2 and SCC-38.
The greater activity of 16, which is an RAR/RXR pan-
agonist, indicates that RARs also contribute to the
mechanism of growth inhibition by heteroarotinoids.
Additive and synergistic activity of RAR- and RXR-
selective retinoids has been noted in other cell lines in

4440 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 21
Zacheis et al.
Exp er im en ta l Section
Ch em ica l Meth od s. Gen er a l. IR spectra were recorded
on a Perkin-Elmer 2000 FT-IR as films or as KBr pellets. GC-
MS data were obtained from a HP G1800A GCD System GC
electron ionization detector. FAB MS experiments were per-
formed on a VG ZAB-2SE HRMS unit, while EI MS experi-
ments were accomplished with an HP GC-MS ENGINE model
5989B unit. All 1H and 13C NMR spectra were taken on a
Varian Inova 400 MHz or Varian XL-400 MHz BB spectrom-
eter operating at 399.99 and 100.0 Hz, respectively, and signals
were referenced to TMS. Melting points were determined with
a Thomas-Hoover melting point apparatus and were uncor-
rected. Syntheses were executed, unless otherwise indicated,
under an atmosphere of N2. Elemental analyses were per-
formed by Galbraith Laboratories, Knoxville, TN, or by
Atlantic Microlab, Inc., Norcross, GA. All chromatography was
performed with the chromatotron unless otherwise noted and
utilized silica gel, pF 254 containing gypsum (EM Science).
Heteroarotinoids 5,7 6,7 12,6 14,6 15,7 17,39 18,39 and 1939
were prepared by known methods. Although the intermediates
and final products appeared to be relatively stable in light,
precautions were taken to minimize exposure to any light
source and to the atmosphere. In the purified state, all
materials were stored in the cold and dark for indefinite
periods without significant decomposition.
Biologica l Meth od s. Cell Lin es. The SCC-2 cell line was
derived from a patient with T1,N0 well-differentiated squa-
mous cell carcinoma of the mandibular alveolus. The SCC-38
cell line was derived from a patient with a T2,N2a moderately
well-differentiated squamous cell carcinoma of the base of the
tongue. Cells were maintained in Eagles minimal essential
media (MEM) containing Earle’s salts and L-glutamine, supple-
mented with nonessential amino acids, antibiotic-antimycotic,
sodium pyruvate, and 10% fetal bovine serum (FBS) in 95%
air-5% CO2. Only lots of FBS that had negligible quantities
of retinoic acid as determined by HPLC (<10-8 M) were used.
Retin oid s. All heteroarotinoids (5-10, 12-19), arotinoid
11, t-RA (1; Sigma Chemical Co., St. Louis, MO), and 9-c-RA
(2; Bio-Mol, Plymouth Meeting, PA) were dissolved in dimethyl
sulfoxide (DMSO) for in vitro studies or highly refined sesame
oil (Croda, Inc., Mill Hall, PA) for in vivo studies, were
protected from light, and were stored at -70 °C. For the tissue
culture studies, retinoids were added from 1000X stocks,
achieving a final concentration of 0.01% DMSO in the culture
medium. For the animal studies, the retinoids were dissolved
at a concentration equal to the daily dose of 0.1 mL of sesame
oil.
F igu r e 2. RAR expression profile of SCC-2 and SCC-38. RT-
PCR reaction products, obtained with primers specific for each
receptor, were electrophoresed through 1.5% agarose gels
containing ethidium bromide and photographed. The expected
size of the specific bands for RARR, RARâ, RARγ, and RXRâ
is approximately 200 base pairs. The expected size for RXRR
is 400 base pairs. The lanes contain RT-PCR reaction products
from SCC-2 (2), SCC-38 (38), or no RNA control (0) and
molecular weight markers (M) (1 kb DNA ladder, GibcoBRL,
Gaithersburg, MD; the top band corresponds to 505 bp’s
followed by 396, 344, 288, 220, 201, 154, 134, and 75).
vitro.37 The inhibition of HNSCC xenograph tumor
growth by an RAR-selective retinoid, observed by oth-
ers,38 and by RXR-selective retinoid 6, in this study,
demonstrates that either RAR or RXR activation is
sufficient for growth inhibition. The more potent activity
of RAR/RXR pan-agonist 16 suggests that activation of
both RAR and RXR receptors is also complimentary in
vivo.
Since all of the retinoids used in the animal model
repressed AP-1-driven transcription to similar extents,
the contribution of anti-AP-1 activity to the mechanism
of growth inhibition cannot be ruled out. Although the
most biologically active heteroarotinoid 16 exhibited the
greatest repression of AP-1 activity, it was also the most
active RARE transactivator and, therefore, cannot be
used to differentiate the two retinoid activities. It is
most likely, however, that both anti-AP-1 and RARE
transactivation activities are involved in the mechanism
of retinoid growth inhibition to different degrees, de-
pending upon the inner cellular environment. A series
of heteroarotinoids are currently being investigated for
selected anti-AP-1 activity in the absence of RARE
transactivation activity. These types of selective com-
pounds can be used to evaluate the role of AP-1
repression in the mechanism of growth inhibition.
Meth yl N-(4-Meth oxyoxop h en yl)-4,4-d im eth yl-3,4-d i-
h yd r o-2H-ben zo[b]p yr a n -6-ylh yd r oxa m a te (7). A mixture
of powdered KOH (0.022 g, 0.38 mmol) and DMSO (1 mL) was
stirred for 5 min, and then ester 21 (0.090 g, 0.25 mmol) was
added, followed by an immediate addition of H3C-I (0.031 mL,
0.51 mmol) with stirring. After stirring for 0.5 h, the resulting
mixture was poured into H2O (10 mL). Extraction (H2CCl2, 3
× 10 mL) of the aqueous layer and combining the extracts gave
a solution which was washed with water and brine. After
drying (MgSO4), the solution was evaporated to a colorless
solid which was recrystallized (hexanes:EtOAc, 4:1) to give 7
(0.045 g, 48%): mp 104-106 °C; IR (KBr) 1722, 1670 (CdO)
1
cm-1; H NMR (DCCl3) δ 1.29 [s, 6 H, (H3C)2], 1.83 [t, 2 H, J
) 5.6 Hz, CCH3], 3.71 [s, 3 H, NOCH3], 3.92 [s, 3 H, OCH3],
4.23 [t, 2 H, J ) 5.6 Hz, OCH2], 6.77 [d, 1 H, Ar-H], 7.47 [dd,
1 H, Ar-H], 7.59 [dd, 2 H, Ar-H], 7.63 [d, 1 H, Ar-H], 8.04
[dd, 2 H, Ar-H]; 13C NMR (DCCl3) ppm 30.53 [ArCCH2], 30.74
[(H3C)2], 37.05 [ArCCH2], 51.15 [OCH3], 62.17 [NOCH3], 63.32
[ArOCH2], 116.59-156.38 [Ar-C], 166.38 [CdO], 167.97 [Cd
O]. Anal. (C21H23NO5) C, H, N.
In conclusion, the growth inhibitory activity of RXR-
selective 6 and the more potent growth inhibition
activity of RAR/RXR pan-agonist 16 implicate both
RARs and RXRs in the molecular mechanism of retinoid
growth inhibition of HNSCC cell lines. Both agents
exhibit anti-AP1 activity that may contribute to the
mechanism of action. Moreover, the chemoprevention
activity and lack of toxicity of heteroarotinoids demon-
strate their clinical potential in head and neck cancer
chemoprevention.
Meth yl 4-[(6-Hyd r oxy-2,4,4,5,7-p en ta m eth yl-4H-ben zo-
[b]p yr a n -6-yl)ca r bon yloxy]ben zoa te (8). To a cloudy mix-
ture of the chromenol 24 (0.150 g, 0.688 mmol), methyl
monoterephthalic acid (0.161 g, 0.894 mmol), and H2CCl2 (10
mL) were added slowly dicyclohexylcarbodiimide (0.425 g, 2.06
mmol) and a catalytic amount (7 mg) of DMAP. After the
resulting cloudy solution was stirred (rt) for 24 h and filtered,

Heteroarotinoids Inhibit Cancer Cell Lines
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 21 4441
Ta ble 2. Potency (EC50, nM) and Efficacy (maximal response relative to 9-c-RA (2)) of Heteroarotinoidsa
potency (efficacy)
RAR
RXR
compd
R
â
γ
R
â
γ
6
16
>10000 (24)
>10000 (58)
>10000 (38)
294 (35)
84 (80)
1897 (46)
68 (49)
42 (35)
302 (43)
225 (72)
1004 (73)
152 (86)
a
The efficacy was derived by dividing the maximal activity induced by 6 or 16 by the maximal activity induced by 9-c-RA (2). All
compounds exhibited maximal activity at a concentration of 10 µM. The maximal activity induced by 10 µM 9-c-RA (2) was 4.2-fold
induction for RARR, 2.6-fold for RARâ, 2.4-fold for RARγ, 3.8-fold for RXRR, 4.0-fold for RXRâ, and 3.7-fold for RXRγ.
OCH3], 4.40 [q, 2 H, J ) 7.2 Hz, OCH2], 6.93 [s, 1 H, Ar-H],
7.29 [d, 2 H, Ar-H], 7.73 [s, 1 H, Ar-H], 8.12 [s, 2 H, Ar-H];
13C NMR (DCCl3) ppm 14.31 [H2C-CH3], 26.79 [(H3C)2], 29.76
[(H3C)2], 34.67 [ArC(CH3)2], 56.11 [ArCOCH2], 61.03 [CH2-
CH3], 70.33[ArOCH3], 74.86 [H2CO], 108.79-158.29 [Ar-C],
163.59, 165.93 [CdO]. Anal. (C24H28O6) C, H.
E t h yl 4-[(3-Met h oxy-5,6,7,8-t et r a h yd r o-5,5,8,8-t et r a -
m eth yln a p h th a len -2-yl)ca r bon yloxy]ben zoa te (11). To a
cloudy mixture of acid 35 (0.300 g, 1.14 mmol) and ethyl
4-hydroxybenzoate (0.300 g, 1.8 mmol) were added dicyclo-
hexylcarbodiimide (0.750 g, 3.6 mmol) and a catalytic amount
of DMAP (10 mg). After the resulting clear solution had been
stirred at rt (24 h) and then filtered, it was evaporated to a
F igu r e 3. Repression of AP-1-driven transcription. SCC-2
cultures transiently cotransfected with an AP-1 responsive
reporter plasmid (ColCAT) and a transfection control plasmid
(pJ ATLac) were treated with 10 µM of the indicated retinoid.
After 48 h of treatment, cell lysates were prepared and CAT
expression was quantitated as a measure of AP-1 activity. The
results are presented as the average and standard deviations
of 4 determinations.
heavy oil. Chromatography of the oil (hexane:EtOAc, 4:1) gave
ester 11 (0.312 g, 67%) as a white solid: mp 104-106 °C; IR
(KBr) 1722 (CdO) cm-1 1H NMR (DCCl3) δ 1.28 [s, 6 H,
;
(H3C)2], 1.30 [s, 6 H, (H3C)2], 1.39 [t, 3 H, J ) 7.2 Hz, CH2CH3],
1.68 [2 t, 4 H, J ) 6.0 Hz, (CH2)2], 3.89 [s, 3 H, OCH3], 4.37 [q,
2 H, J ) 7.2 Hz, CH2CH3], 6.90 [s, 1 H, Ar-H], 7.26 [d, 2 H,
Ar-H], 7.93 [s, 1 H, Ar-H], 8.07 [d, 2 H, Ar-H]; 13C NMR
(DCCl3) ppm 14.34 [CH2CH3], 31.65 [(H3C)2], 31.86 [(H3C)2],
33.75 [CH2], 34.82 [CH2], 34.84 [ArC(CH3)2], 35.09 [ArC(CH3)2],
56.07 [OCH3], 61.03 [CH2CH3], 110.07-157.71 [Ar-C], 163.85
[CdO], 166.03 [CdO]. Anal. (C22H23NO5) C, H.
the filtrate was cooled (0 °C) overnight and filtered again.
Evaporation of the solvent gave a heavy oil which was
subjected to chromatography (H2CCl2). A white solid (0.130 g,
50%) 8 was obtained: mp 142-143 °C; IR (KBr) 1736 (CdO)
cm-1; 1H NMR (DCCl3) δ 1.44 [s, 3 H, CH3], 1.57 [s, 3 H, CH3],
2.12 [s, 3 H, Ar-CH3], 2.27 [s, 3 H, Ar-CH3], 3.98 [s, 3 H,
OCH3], 4.39 [s, 1 H, CdCH], 6.72 [s, 1 H, Ar-H], 8.18 [dd, 2
H, Ar-H], 8.30 [dd, 2 H, Ar-H]; 13C NMR (DCCl3) ppm 14.88
[CH3], 16.23 [CH3], 18.78 [CH3], 31.79 [ArCCH2], 32.54 [(H3C)2],
52.51 [OCH3], 109.00-148.51 [CdC] and [Ar-C], 163.83 [Cd
O], 166.13 [CdO], 168.07 [CdO]. Anal. (C23H24O5) C, H.
Meth yl 4-[(2,2,4,4,5,7-Hexa m eth yl-3,4-d ih yd r o-2H-ben -
zo[b]p yr a n -6-yl)ca r bon yloxy]ben zoa te (9). To a cloudy
mixture of 25 (0.085 g, 0.366 mmol) and methyl monotereph-
thalic acid (0.115 g, 0.641 mmol) in H2CCl2 (10 mL) were added
dicyclohexylcarbodiimide (0.132 g, 0.641 mmol) and a catalytic
amount (4 mg) of DMAP. The resulting solution was stirred
(18 h) at rt, filtered, and evaporated to a heavy oil which was
chromatographed (H2CCl2). The oil crystallized when treated
with dry hexane and gave 9 (0.075 g, 52%): mp 95-96.5 °C;
Eth yl (2E,4E,6E)-7-(2,3-Dih yd r o-1,4-ben zod ioxa n -6-yl)-
3-m eth yl-2,4,6-octa tr ien oa te (13). To a stirred suspenion
of 38 (4 g, 7.53 mmol) in dry ether (40 mL) was added dropwise
n-BuLi (14 mL, 0.906 M, 21.7 mmol in hexanes) over 5 min.
The resulting dark red solution was stirred for 0.5 h, and then
ethyl 3-methyl-4-oxocrotonate (l.3 mL, 9.15 mmol) in dry ether
(10 mL) was added dropwise. After 10 h of stirring at rt, the
solution was diluted (hexanes, 100 mL), filtered, and evapo-
rated to a light yellow oil. Crystals formed upon standing and
were recrystallized (absolute ethanol) to give 13 (0.683 g,
28.85%): mp 86-87 °C; IR 1690 (CdO) cm-1; 1H NMR (DCCl3)
δ 1.3 [t, J ) 7.1 Hz, 3 H, H(19)], 2.20 [s, 3 H, H(10)], 2.37 [d,
J ) 0.98 Hz, 3 H, H(15)], 4.17 [q, J ) 7.1 Hz, 2 H, H(18)], 4.27
[s, 4 H, H(2,3)], 5.78 [bs, 1 H, H(16)], 6.33 [d, J ) 15.1 Hz, 1
H, H(13)], 6.51 [d, J ) 11.2 Hz, H(11)], 6.87-7.04 [m, 3 H,
H(5,7,8)]. Anal. (C16H22O4) C, H.
Meth yl 4-[(2,3-Dih yd r o-2,2,4,4-tetr a m eth yl-2H-1-ben -
zoth iop yr a n -6-yl)ca r ba m oyl]ben zoa te (16). To a solution
of amine 41 (1.6 g, 7.32 mmol) in dry benzene (90 mL) and
pyridine (9.2 mL) at rt was added slowly, with stirring, freshly
made methyl monoterephthaloyl chloride (2.20 g, 10.84 mmol).
After stirring for 12 h (rt), the mixture was poured into water,
and the resulting mixture was extracted (EtOAc, 4 × 200 mL).
The combined extracts were washed with 6 N HCl (4 × 50
mL), H2O (3 × 400 mL), saturated NaHCO3 (2 × 400 mL),
H2O (50 mL), and brine (200 mL). The dried (Na2SO4) organic
solution was evaporated to a light yellow solid which was
purified via chromatography over silica gel (hexane:EtOAc:
Et2NH, 74:25:1). Recrystallization (hexane:EtOAc, 3:1) of the
solid gave 16 (2.00 g, 72%): mp 154-155 °C; a further
recrystallized analytical sample of 16 melted at 162-164 °C;
IR (KBr) 3395-3390 (N-H), 1725 (CdO), 1690-1680 (HNCd
O) cm-1; 1H NMR (DCCl3) δ 1.41 [s, 6 H, (H3C)2], 1.42 [s, 6 H,
SC(CH3)2], 1.96 [s, 2 H, CH2], 3.95 [s, 3 H, CH3], 7.11 [d, 1 H,
Ar-H], 7.33 [d, 1 H, Ar-H], 7.76 [s, 1 H, Ar-H], 7.91 [d, 2 H,
Ar-H], 7.93 [bs, 1 H, Ar-H], 8.12 [d, 2 H, Ar-H], 8.13 [s, 1
H, Ar-H]; 13C NMR (DCCl3) ppm 31.51 [C(CH3)], 32.48 [SC-
(CH3)2], 35.94 [CCH3], 42.17 [SC(CH3)2], 52.43 [CH2], 54.33
[OCH3], 118.61-143.66 [Ar-C], 164.72 [NHCdO], 166.21
1
IR (KBr) 1730 (CdO) cm-1; H NMR (DCCl3) δ 1.35 [s, 6 H,
(H3C)2], 1.48 [s, 6 H, (H3C)2], 1.87 [s, 2 H, CH2], 2.08 [s, 3 H,
Ar-CH3], 2.29 [s, 3 H, Ar-CH3], 3.98 [s, 3 H, OCH3], 6.61 [s,
1 H, Ar-H], 8.18 [d, 2 H, Ar-H], 8.29 [d, 2 H, Ar-H]; 13C
NMR (DCCl3) ppm 15.17 [Ar-CH3], 16.24 [Ar-CH3], 27.87
[(H3C)2], 30.78 [(H3C)2], 32.13 [Ar-C(CH3)2], 52.51 [OCH3],
53.33 [ArOC(CH3)2], 73.11 [OC(CH3)2], 118.09-150.80 [Ar-
C], 163.97 [CdO], 166.19 [CdO]. Anal. (C24H28O5) C, H.
Eth yl 4-[(1,1,4,4-Tetr a m eth yl-7-m eth oxy-3,4-d ih yd r o-
1H-2-ben zo[b]p yr a n -6-yl)ca r bon yloxy]ben zoa te (10). To
a cloudy mixture of acid 32 (0.100 g, 0.382 mmol) and ethyl
4-hydroxybenzoate (0.095 g, 0.573 mmol) were added dicyclo-
hexylcarbodiimide (0.118 g, 0.573 mmol) and a catalytic
amount of DMAP (7 mg). The yellow solution was stirred at
rt (48 h) and then filtered. Evaporation of the solution gave a
heavy oil which was chromatographed (hexane:EtOAc, 3:1).
An oil resulted and was treated with 0.5 mL of hexane:EtOAc
(4:1) and chilled. Crystallization occurred to give solid 10
(0.110 g, 70%): mp 106.5-108 °C; IR 1722 (CdO) cm-1 1H
;
NMR (DCCl3) δ 1.32 [s, 6 H, (H3C)2], 1.40 [s, 3 H, J ) 7.2 Hz,
CH3], 1.56 [s, 6 H, (H3C)2], 3.62 [s, 2 H, H2CO], 3.94 [s, 3 H,

4442 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 21
Zacheis et al.
[HOCdO]; MS (EI) calcd for C22H25O3SN [M+] 383.1555, found
381.1552. Anal. (C22H25O3SN) C, H, N.
[Ar-C]; MS (EI) calcd for C15H22O2 [M+] 234, found 234. The
sample of 25 was used at once to prepare 9.
2-(3-Meth oxyp h en yl)-2-m eth ylp r op a n en itr ile (27). An
orange mixture of KOH (20.0 g. 0.356 mol), triethylbenzylam-
monium bromide (TEBA; 2.0 g, 7.46 mmol), and (3-methoxy-
phenyl)acetonitrile (26; 20.0 g, 0.135 mol) was heated at 80
°C for 0.5 h. Methyl iodide (11.50 g, 0.810 mol) was added
slowly over 2 h, and the resulting yellow mixture was heated
(80 °C) for 1 h. Additional KOH (20.0 g, 0.356 mol) was then
added along with TEBA (2.0 g, 7.46 mmol), and the new
mixture was stirred vigorously for 0.5 h. Methyl iodide (11.50
g, 0.081 mol) was added over 1 h, and the resulting mixture
was stirred with heating (80-100 °C) for 5 h. After cooling to
rt, the reaction mixture was extracted (toluene, 5 mL). The
combined organic extracts were washed with water and brine
and then dried (MgSO4). Evaporation of the solvent gave an
oil (3.8 g, 84%) which was vacuum distilled (100-101 °C/∼2
mmHg) to a colorless oil (13.4 g, 57%) of 27: IR (neat) 2242
CtN cm-1; 1H NMR (DCCl3) δ 1.72 [s, 6 H, (H3C)2], 3.83 [s, 3
H, OCH3], 6.84 [2d 1 H, Ar-H], 7.01 [t, 1 H, Ar-H], 7.05 [2d,
1 H, Ar-H], 7.31 [t, 1 H, Ar-H]; 13C NMR (DCCl3) ppm 29.08
[(H3C)2], 36.76 [Ar-C(CH3)2], 55.30 [ArOCH3], 111.40-159.87
[Ar-C], 123.44 [CtN]; MS (EI) calcd for C11H13NO [M+] 175,
found 175. This sample of 27 was used to prepare 28.
N-(4-Meth oxyoxop h en yl)-4,4-d im eth yl-3,4-d ih yd r o-2H-
ben zo[b]p yr a n -6-ylh yd r oxa m ic Acid (21). Acid 207 (0.300
g, 1.40 mmol) and thionyl chloride (15 mL) were stirred at rt
for 16 h. Excess thionyl chloride was evaporated under
aspirator pressure, and the residual acid choride (an oil) was
further dried under higher vacuum (∼2 mmHg) to remove
traces of thionyl chloride. The hydroxylamine (0.253 g, 1.40
mmol) was dissolved in THF (10 mL), and then NaHCO3 (0.176
g, 2.1 mmol) was added at rt. The above acid chloride was
dissolved in a minimum of dry ether and then was added
slowly to the solution of the hydroxylamine at 0 °C. The
resulting mixture was stirred (rt) for 18 h, filtered, and
evaporated in vacuo to yield a solid. Chromatography of the
solid (hexanes:EtOAc, 1:10) gave 21 (0.250 g, 50%): mp 70-
71 °C; IR (KBr) 3428 (O-H), 1721, 1634 (CdO) cm-1; 1H NMR
(DCCl3) δ 1.13 [s, 6 H, (CH3)2], 1.78 [t, 2 H, J ) 5.6 Hz, CCH2)]
3.90 [s, 3 H, OCH3], 4.19 [t, 2 H, J ) 5.6 Hz, H2CO], 6.68 [d,
2 H, Ar-H], 7.20 [dd, 1 H, Ar-H], 7.25 [dd, 1 H, Ar-H], 7.34
[d, 1 H, Ar-H], 7.96 [d, 2 H, Ar-H], 9.18 [s, 1 H, OH]. Anal.
(C20H21NO5) C, H. Acid 21 was used at once to obtain 7.
2,3-Dih yd r o-2,6-d ih yd r oxy-2,4,4,5,7-p en t a m et h yl-2H -
ben zop yr a n (23). To a solution of 1.50 mL (4.50 mmol) of 3
M methylmagnesium bromide in ether (5 mL) was added (rt)
the coumarol 2227 (0.100 g, 0.45 mmol) in ether (5 mL). The
resulting solution was stirred for 48 h (rt). After the solution
was allowed to cool to rt, a saturated solution (∼10 mL) of
aqueous H4NCl was added slowly. Separation of the layers was
followed by extraction (ether, 3 × 10 mL) of the organic layer
to give extracts which were washed with water and brine. After
drying (MgSO4), the solution was evaporated to give the lactol
23 as a colorless solid (0.57 g, 54%): mp 105.5-106.5 °C; IR
2-(3-Meth oxyp h en yl)-2-m eth ylp r op a n oic Acid (28). A
mixture of 27 (3.8 g, 0.0217 mol), KOH (5.00 g, 0.089 mol),
diethylene glycol (30 mL), and water (5 mL) were heated at
130-140 °C for 3 days and turned brown. After cooling to rt,
the solution was poured into cold water. Extraction (benzene,
8 × 40 mL) of the aqueous layer followed, and then the residual
aqueous layer was made acidic (concd HCl, pH ∼ 3). The new
aqueous layer was extracted (ether, 4 × 30 mL), and the
combined extracts were dried (MgSO4). Evaporation of the
solvent gave a yellow solid which recrystallized (hexane) to a
white solid 28 (3.81 g, 69%): mp 79-80 °C (lit.28 mp 80-81.5
°C). Since spectral data were scarce on 28, the following data
1
(KBr) 3472, 3408 (O-H) cm-1 ; H NMR (DCCl3) δ 1.41 [s, 3
H, CCH3], 1.58 [s, 3 H, CCH3], 1.62 [s, 3 H, C(OH)-CH3], 1.90
[d, 1 H, J ) 14 Hz, CH2], 1.99 [d, 1 H, J ) 14 Hz, CH2], 2.18
[s, 3 H, Ar-CH3], 2.38 [s, 3 H, Ar-CH3], 4.31 [s, 1 H, OH],
6.52 [s, 1 H, Ar-H]; 13C NMR (DCCl3) ppm 14.75 [Ar-CH3],
15.86 [Ar-CH3], 28.92 [(H3C)2], 29.95 [(H3C)2], 30.93 [CH2],
32.14 [C(CH3)2], 50.07 [H3C(COAr)], 95.79 [C(OH)], 117.34-
147.25 [Ar-C]. MS (EI) calcd for C14H20O3 [M+] 236, found 236.
Lactol 23 was converted to 24.
were taken: IR (KBr) 3434-2538 (CO2H), 1702 (CdO) cm-1
;
1H NMR (DCCl3) δ 1.58 [s, 6 H, (H3C)2], 3.80 [s, 3 H, ArOCH3],
6.80 [ddd, 1 H, Ar-H], 6.95 [t, 1 H, Ar-H], 6.99 [ddd, 1 H,
Ar-H], 7.25 [t, 1 H, Ar-H]; MS (EI) calcd for C11H14O3 [M+]
194, found 194. Acid 28 was converted directly to 29.
2-(3-Meth oxyp h en yl)-2-m eth yl-1-p r op a n ol (29). To a
mixture of LiAlH4 (2.23 g, 58.9 mmol) in dry THF (19 mL)
was added dropwise acid 28 (3.81 g, 19.6 mmol) in dry THF
(5 mL) over 0.5 h. After being heated at reflux (48 h), the
resulting mixture was allowed to cool to rt and was then
treated cautiously with ethanol/water to destroy residual
LiAlH4. A white solid formed and was filtered, and the
remaining aqueous layer was extracted (ether, 5 × 30 mL).
The combined extracts were washed with brine, dried (MgSO4),
and evaporated to a colorless oil 29 (2.6 g, 75%): IR (neat)
6-Hydr oxy-2,4,4,5,7-pen tam eth yl-4H-1-ben zopyr an (24).
A mixture of lactol 23 (0.100 g, 0.42 mmol), a catalytic amount
of p-toluenesulfonic acid (0.025 g), 4A molecular sieves (1.0
g), and toluene was boiled for 2 h. After cooling to rt and
filtering, the filtrate was washed with a saturated solution of
NaHCO3 (∼10 mL) and then with brine. The dried (Na2SO4)
organic solution was evaporated to yield 24 (0.082 g, 89%) as
a light yellow oil: IR (neat) 3576 (O-H) cm-1; 1H NMR (DCCl3)
δ 1.43 [s, 6 H, (H3C)2)], 1.85 [s, 3 H, CdC-CH3], 2.19 [s, 3 H,
Ar-CH3], 2.33 [s, 3 H, Ar-CH3], 4.33 [s, 1 H, OH], 4.39 [s, 1
H, CdCH2], 6.60 [s, 1 H, Ar-H]; 13C NMR (DCCl3) ppm 14.38
[Ar-CH3], 15.79 [Ar-CH3], 18.87 [CdCCH2], 31.82 [(H3C)2],
32.42 [C(CH3)2], 108.74 [CdCH], 116.12 [CdCCH3], 121.91-
148-30 [Ar-C]; MS (EI) calcd for C14H18O2 [M+] 218, found
218. This chromenol 24 was used directly to prepare hetero-
arotinoid 8.
3392 (O-H) cm-1 1H NMR (DCCl3) δ 1.32 [s, 6 H, (H3C)2],
;
3.59 [s, 2 H, H2COH], 3.81 [s, 3 H, ArOCH3], 6.75 [2 d, 1 H,
Ar-H], 6.93 [t, 1 H, Ar-H], 6.97 [2 d, 1 H, Ar-H], 7.27 [t, 1
H, Ar-H]; MS (EI) calcd for C11H16O2 [M+] 180, found 180.
The sample of 29 was used immediately to prepare ether 30.
1,1,4,4-Tetr a m eth yl-6-m eth oxy-3,4-d ih yd r o-1H-2-ben -
zop yr a n (30). A solution of alcohol 29 (0.989 g, 5.6 mmol),
acetone (30 mL), and concd HCl (10 mL) was heated and
stirred at 45-50 °C for 2 h and at rt for 12 h. The deep yellow
solution was allowed to cool to rt and was then poured into
ice-water. Two layers were separated, and the organic layer
was extracted (ether, 5 × 20 mL). The combined extracts were
washed with water (30 mL), saturated NaHCO3 (30 mL), and
brine and then was dried (MgSO4). Evaporation of the solvent
gave a colorless oil 30 (0.92 g, 75%) which was used directly
to prepare 31. Spectral data for 30 were as follows: 1H NMR
(DCCl3) δ 1.26 [s, 6 H, (H3C)2], 1.51 [s, 6 H, (H3C)2], 3.58 [s, 2
H, H2CO], 3.80 [s, 3 H, ArOCH3], 6.73 [dd, 1 H, Ar-H], 6.82
[d, 1 H, Ar-H], 6.99 [d, 1 H, Ar-H]; MS (EI) calcd for C14H20O2
[M+] 220, found 220.
2,2,4,4,5,7-Hexa m eth yl-6-h yd r oxy-3,4-d ih yd r o-2H-ben -
zo[b]p yr a n (25). To a solution of methylmagnesium bromide
(1.50 mL, 34.50 mmol, 3 M) in THF (5 mL) was added
coumarol 2227 (0.100 g, 0.45 mmol) in THF (5 mL) at rt. The
resulting solution was stirred for 4 days at reflux and was then
allowed to cool. Saturated aqueous H4NCl was added, and two
layers separated. Extraction (ether, 3 × 10 mL) of the aqueous
layer and combining the extracts gave an organic solution
which was washed with H2O and brine. The dried (MgSO4)
solution was evaporated to a colorless solid (0.086 g, 76%) of
25: mp 62.5-64.5 °C; IR (KBr) 3458 (O-H) cm-1 1H NMR
;
(DCCl3) δ 1.31 [s, 6 H, (H3C)2], 1.47 [s, 6 H, (H3C)2], 1.84 [s, 2
H, CH2], 2.17 [s, 3 H, Ar-CH3], 2.37 [s, 3 H, Ar-CH3], 4.25 [s,
1 H, OH], 6.50 [s, 1 H, Ar-H]; 13C NMR (DCCl3) ppm 14.77
[Ar-CH3], 15.85 [Ar-CH3], 27.84 [(H3C)2], 30.88 [(H3C)2], 31.98
[ArC(CH3)2], 53.68 [CH2], 72.58 [OC(CH3)2], 117.56-146.68
7-Aceto-1,1,4,4-tetr a m eth yl-6-m eth oxy-3,4-d ih yd r o-1H-
2-ben zop yr a n (31). To a solution of 30 (0.989 g, 4.18 mmol),
acetyl chloride (0.74 mL, 10.0 mmol), and H3CNO2 (15 mL)

Heteroarotinoids Inhibit Cancer Cell Lines
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 21 4443
was added very slowly and portionwise AlCl3 (1.33 g, 10.0
mmol) over 1 h. The resulting brown mixture was stirred at
rt for 24 h and was then poured cautiously into ice-water (50
mL). Dilute HCl (2 M, ∼5 mL) was added cautiously to destroy
residual AlCl3. The layers were separated, the aqueous layer
was extracted (H2CCl2, 3 × 10 mL), and the combined organic
extractssoriginal layer were washed with H2O (20 mL),
saturated aqueous NaHCO3 (25 mL), and brine. Evaporation
of the solvent gave a solid which was by purified by chroma-
tography (H2CCl2) and yielded a colorless solid 31 (0.562 g,
THF (25 mL) was added dropwise vinylmagnesium bromide
[prepared from Mg (1.52 g, 62.53 mmol) and vinyl bromide
(10 g, 93.45 mmol) in THF (10 mL)] over 0.25 h. After stirring
at rt for 10 h, the mixture was cooled in a water bath, and a
saturated aqueous solution (20 mL) of H4NCl was added
dropwise to effect decomposition. Separation and extraction
(ether, 4 × 50 mL) of the aqueous layer gave a solution of
combined extracts which was washed with saturated, aqueous
NaCl. Drying of the organic solution and evaporation of the
solvent gave slightly crude 37 (4.78 g, 103%) as a yellow oil:
IR (neat) 3700-3100 (O-H) cm-1; 1H NMR (DCCl3) δ 1.60 [s,
3 H, H(10)], 4.23 [s, 4 H, H(2,3)], 5.11 [dd, J cis ) 10.6 Hz, J gem
) 1.08 Hz, 1 H, H(12)], 5.27 [dd, J trans ) 17.2 Hz, J gem ) 1.08
Hz, J gem ) 1.08 Hz, 1 H, H(12)], 6.12 [dd, J trans )17.3 Hz, J cis
) 10.6 Hz, H (11)] 6.8-6.99 [m, Ar-H, 3 H, H(5,7,8)]; HRMS
calcd for C12H14O3 [M+] 206.0942, found 206.0938. This sample
of 37 was used without further purification to prepare salt 38.
1
51%): mp 115-117 °C; IR (KBr) 1662 (CdO) cm-1; H NMR
(DCCl3) δ 1.29 [s, 6 H, (H3C)2], 1.52 [s, 6 H, (H3C)2], 2.60 [s, 3
H, C(O)CH3], 3.59 [s, 2 H, OCH2], 3.91 [s, 3 H, ArOCH3], 6.85
[s, 1 H, Ar-H], 7.50 [s, 1 H, Ar-H]; MS (EI) calcd for C16H22O3
[M+] 262, found 262. Ketone 31 was used at once to prepare
acid 32.
1,1,4,4-Tetr a m eth yl-6-m eth oxy-3,4-d ih yd r o-1H-2-ben -
zop yr a n -7-ca r boxylic Acid (32). To ketone 31 (0.560 g, 0.216
mmol) was added a solution of LiOCl (Bioguard-LiOCl 29%,
inert 71%)/Clorox [5 g LiOCl (Bioguard), 20 mL Clorox] and
95% ethanol (10 mL) in the normal manner.40 The resulting
light yellow solution was boiled (24 h) and then allowed to cool
to rt. The basic solution was extracted (ether, 3 × 15 mL), and
the combined extracts were washed with water (10 mL) and
brine and then were dried (MgSO4). The original aqueous layer
was acidified (2 M HCl, 10 mL, ∼pH 3) and then extracted
(ether, 4 × 20 mL). The combined organic layers were washed
with water (20 mL) and brine and then dried (MgSO4).
Evaporation of the solvent gave a colorless oil which crystal-
lized (EtOAc) to give 32 (0.420 g, 21%): mp 147-149 °C; IR
3443-2538 [C(O)O-H], 1691 [CdO] cm-1; 1H NMR (DCCl3) δ
1.30 [s, 6 H, (H3C)2], 1.54 [s, 6 H, (H3C)2], 3.61 [s, 2 H, H2CO],
4.07 [s, 3 H, ArOCH3], 6.93 [s, 1 H, Ar-H], 7.89 [s, 1 H, Ar-
H]; MS (EI) calcd for C15H20O4 [M+] 264, found 264. The acid
32 was used directly to obtain the target compound 10.
[3-(2,3-Dih yd r o-1,4-b en zd ioxa n -6-yl)-2-b u t en -1-yl]t r i-
p h en ylp h osp h on iu m Br om id e (38). To a stirred suspension
of Ph3P‚HBr (2.61 g, 4.92 mmol) in dry methanol (20 mL) was
added dropwise alcohol 37 (1.01 g, 4.92 mmol) in dry methanol
(30 mL) over 10 min at rt under N2. After stirring at rt for 10
h, the suspension was concentrated to about 5 mL and was
then transferred to a large beaker. Dry ether was added slowly
with stirring to complete the precipitation. Filtration gave a
solid which was washed (dry ether) and then was dissolved in
dry methanol (∼15 mL). Ether was added to cloudiness, and
the resulting mixture was chilled in the freezer. Filtration and
drying the solid gave fine, white crystals of 38 (2.22 g,
84.92%): mp 234-235 °C; 1H NMR (DCCl3) δ 1.58 [d, 3 H,
H(01)], 4.22 [s, 4 H, H(2,3)], 4.84 [dd, J HP ) 15.1 Hz, J HH
)
8.1 Hz, H(12)], 5.57 [m, 1 H, H(11)], 6.88 [m, 3 H, H(5,7,8)],
7.65-7.93 [m, 14 H, Ar-H, P(C6H5)3]. In view of the rarity of
such salts, an elemental analysis was obtained and resulted
in the identification of a monohydrate. Anal. (C30H28O2PBr‚
H2O) C, H. Salt 38 was used directly to obtain 13.
1-(3-H yd r oxy-5,6,7,8-t et r a h yd r o-5,5,8,8-t et r a m et h yl-
n a p h th a len -2-yl)eth a n on e (34). To a solution of ether 3332
(0.050 g, 0.23 mmol), acetyl chloride (0.049 g, 0.69 mmol), and
H3CNO2 (5 mL) was added slowly and portionwise AlCl3 (0.092
g, 28.9 mmol) over 1 h. After stirring at rt (4 h), the brown
solution was poured into ice-water (10 mL). Two layers were
separated, and the aqueous layer was extracted (ether, 2 ×
10 mL). The combined organic layers were washed with H2O
(20 mL), saturated aqueous NaHCO3 (15 mL), and brine. After
drying (MgSO4) and evaporation of the solvent from the
solution, ketone 34 (0.045 g, 75%) was obtained as a light
2,2,4,4-Tetr a m eth yl-6-n itr oth ioch r om a n (40). To a solu-
tion of thioether 395 (5.13 g, 24.8 mmol) in freshly distilled
Ac2O (5 mL) at 0 °C was added slowly a mixture of cold, concd
HNO3 (3.54 mL) and Ac2O (9 mL) over 10 min. After being
stirred for 2 h, the reaction mixture was poured into a solution
of saturated, aqueous NaHCO3. Extraction (H2CCl2, 3 × 40
mL) of the water layer and combining the extracts and organic
layer gave a solution which was washed with water and brine
and then was dried (Na2SO4). Evaporation of the solvent gave
a solid which was recrystallized (hexane) to yield 40 (1.6 g,
27%): mp 103-107 °C; 1H NMR (DCCl3) δ 1.10 [s, 3 H, (H3C)2],
1.37 [s, 3 H, (H3C)2], l.52 [s, SC(CH3)2], 1.56 [s, 3 H, SC(CH3)2],
2.03 [m, 3 H, CH2], 8.01 [d, 1 H, Ar-H], 8.24 [d, 2 H, Ar-H].
The sample of 40 was used directly to prepare amine 41.
yellow solid: mp 96-98 °C; IR (KBr) 1664 cm-1 1H NMR
;
(DCCl3) δ 1.27 [s, 6 H, (H3C)2], 1.30 [s, 6 H, (H3C)2], 1.68 [2t,
4 H, J ) 6.0 Hz, (H2C)2], 2.59 [s, 3 H, CH3], 3.89 [s, 3 H, OCH3],
6.85 [s, 1 H, Ar-H], 7.73 [s, 1 H, Ar-H], 11.87 [s, 1 H, OH];
MS (EI) calcd for C17H24O2 [M+] 260, found 260. Ketone 34
was converted to acid 35.
2,2,4,4-Tetr a m eth yl-6-a m in oth ioch r om a n (41). To chro-
man 40 (0.800 g, 3.18 mmol) in acetic acid (29 mL) and water
(6 mL) was added dropwise with stirring TiCl3/HCl (33.00 g,
21.39 mmol). After stirring for 2 h (rt), the reaction mixtrure
was cooled (0 °C), and NaOH (30%, 130 mL) was added slowly.
Extractions (EtOAc, 4 × 35 mL, and H2CCl2, 2 × 40 mL) of
the mixture followed, and combining the extracts and organic
layer gave a solution which was washed with H2O and
saturated NaHCO3 (2 × 50 mL). The dried (Na2SO4) solution
was evaporated to an oil which was separated via chromatog-
raphy over silica gel (hexane:H2CCl2, 1:1) and led to 41 as a
white solid (0.352 g, 50%): mp 57-59 °C; IR (KBr) 3450, 3360
3-Meth oxy-5,6,7,8-tetr a h yd r o-5,5,8,8-tetr a m eth yln a p h -
th a len e-2-ca r boxylic Acid (35). A mixture of ketone 34
(0.900 g, 3.46 mmol), a solution of LiOCl and Clorox [40 g
LiOCl (Bioguard), 80 mL Clorox], and 95% ethanol (45 mL)
was boiled for 24 h and then allowed to cool to rt. An aqueous
solution of Na2S2O5 (25%, 35 mL) was slowly added to the
reaction mixture which was then acidified with 6 M HCl (10
mL, pH ∼ 3). Extraction (ether, 4 × 20 mL) of this mixture
and then washing of the combined extracts with H2O (20 mL)
and brine gave a new solution which was dried (MgSO4).
Evaporation of the solvent gave a very light yellow solid 35
(0.420 g, 44%): mp 136.5-138 °C; IR (KBr) 3269 (broad)
1
(N-H) cm-1; H NMR (DCCl3) δ 1.36 [s, 6 H, (H3C)2], 1.39 [s,
6 H, SC(CH3)2], 1.90 [s, 2 H, CH2], 3.50 [bs, 2 H, NH2], 6.44
[d, 1 H, Ar-H], 6.75 [s, 1 H, Ar-H], 9.92 [d, 1 H, Ar-H]. The
sample of 41 was used directly to prepare 16.
1
[C(O)O-H], 1731 [CdO] cm-1; H NMR (DCCl3) δ 1.28 [s, 6
H, (H3C)2], 1.31 [s, 6 H, (H3C)2], 1.69 [t, 4 H, J ) 5 Hz, CH2],
4.06 [s, 3 H, OCH3], 6.93 [s, 1 H, Ar-H], 8.12 [s, 1 H, Ar-H];
13C NMR (DCCl3) ppm 31.58 [(H3C)2], 31.67 [(H3C)2], 33.83
[CH2], 34.60 [CH2], 34.65 [ArC(CH3)2], 35.09 [ArC(CH3)2], 56.47
[OCH3], 109.15-155.53 [Ar-C], 165.60 [CdO]; MS (EI) calcd
for C16H22O3 [M+] 262, found 262. Acid 35 was used directly
to prepare ester 11.
In Vitr o Gr ow th In h ibition . Cultures were plated with
96-well microtiter plates in volumes of 150 µL at a concentra-
tion of 2500 cells/well. The next day, retinoids were added at
4X concentrations in 50 µL of media resulting in 10-9, 10-8
,
10-7, 10-6, and 10-5 M final concentrations of each retinoid.
Control cultures were treated with the same volume of DMSO.
After 3 days of treatment, the cell density in each well was
determined by fixing the cells in trichloroacetic acid and
2-(2,3-Dih yd r o-1,4-ben zod ioxa n -6-yl)-3-bu ten -2-ol (37).
To a solution of commercial ketone 36 (4.0 g., 22.45 mmol) in

4444 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 21
Zacheis et al.
staining the cytoplastic proteins with sulforhodamine B (SRB).
After rinsing, the SRB was solubilized in TrisHCl, and the
optical density of each culture was determined with a MR600
microtiterplate reader. Each experiment was performed in
triplicate, and the three values for each treatment were
averaged. The average OD of the treated cultures was divided
by that of the control cultures treated with solvent alone. To
determine the percent growth inhibition, this ratio was
subtracted from 1 and multiplied by 100.
efficiency by dividing by the âGal expression levels. Each
experiment was performed in triplicate and repeated at least
twice.
An ti-AP -1 Activity. The SCC-2 cells were plated at 3 ×
105 cells/well in 6-well plates 24 h before transfection. The
cultures were cotransfected with a CAT reporter plasmid
driven by the collagenase promoter (Col-CAT) and the pJ AT-
Lac plasmid DNA using FuGENE6 transfection reagent (Boeh-
ringer Mannheim Corp., Indianapolis, IN) according to the
manufacturer’s instructions. After incubation for 24 h, the
media was replenished, and retinoids were added at a con-
centration of 10-5 M. After another 48 h, CAT and âGal
expressions were quantitated as described above. All values
were divided by the value of control cultures treated with
DMSO to determine the percent activities. Each experiment
was performed in duplicate and repeated twice.
In Vivo Tu m or Xen ogr a p h s. Cells in log phase growth
were harvested by trypsinization, resuspended in MEM culture
medium, and centrifuged at 3000 rpm for 10 min. The pellets
were resuspended in MEM culture medium at a concentration
of 1 × 107 cells/mL before implantation into mice. Five-week
old female athymic nu/nu mice (Harlan Sprague-Dawley,
Indianapolis, IN) were housed in a laminar flow room under
sterile conditions at 83-85 °F. The mice were quarantined for
1 week prior to the beginning of the study and were allowed
access to autoclaved food (Purina 5001 mouse/rat sterilizable
diet, St. Louis, MO) and water ad libitum. Animals were
injected with 1 × 107 cells into the right scapular region with
a 24-gauge needle/1 cm3 tuberculin syringe (Becton Dickinson,
Rutherford, NJ ). Twenty-four hours after tumor implantation,
animals were randomized into 4 treatment groups of 5 animals
each. Retinoids were administered daily po beginning 5 days
after tumor implantation with a 20-gauge intragastric feeding
tube (Popper & Sons, New Hyde Park, NY), 5 days/week, at
doses of 10 or 20 mg/kg/day in 0.1 mL of super-refined sesame
oil (Corda, Inc., Parisppany, NJ ). Tumors were measured with
calipers thrice weekly, and tumor volumes were calculated
using the formula: volume ) length × width × heigth. Animal
weights were recorded thrice weekly, and clinical signs of
overall health status and cutaneous toxicities were recorded
weekly.
Ack n ow led gm en t. We gratefully acknowledge par-
tial support of this work by the National Institutes of
Health through grants from the National Cancer Insti-
tute (CA-73639 and F3CA83168). We are also pleased
to acknowledge funding for the Varian Inova 400 MHz
NMR spectrometer in the Oklahoma Statewide Shared
NMR Facility by the National Science Foundation (BIR-
9512269), the Oklahoma State Regents for Higher
Education, the W. M. Keck Foundation, and Conoco, Inc.
The HNSCC cell lines were kindly supplied by Thomas
Carey from the University of Michigan. The RAR
expression vectors and reporter plasmids were gifts of
Magnus Pfahl, and the RXR expression vectors were
gifts of Pierre Chambon.
Refer en ces
Recep tor Exp r ession . RNA was isolated using the Quick-
Prep Kit (Amershsam Pharmacia Biotech, Piscataway, NJ ),
and cDNA was synthesized with Superscript reverse tran-
scriptase (GibcoBRL, Gaithersburg, MD) according to the
manufacturer’s instructions. The specific primers and anneal-
ing conditions used for the PCR reactions were previously
described and demonstrated to be specific for the individual
receptor subtypes.41 The RXR expression was not evaluated
because a human RXRγ gene has not been cloned. All PCR
reactions were performed under the same conditions in a 50-
µL reaction volume containing 5 µL of cDNA from reverse
transcription, 1.25 U of Taq DNA polymerase, and the buffer
provided. Initial denaturation was performed at 94 °C for 5
min, followed by 35 cycles of 45 s at 94 °C, 45 s at the indicated
annealing temperature, and 1 min at 72 °C with an ending of
1 5-min cycle at 72 °C. Negative control reactions consisted of
rt-PCR reactions without added RNA.
(1) (a) Moon, R. C.; Mehta, R. G.; Rao, K. V. N. Retinoids and Cancer
In Experimental Animals. In The Retinoids-Biology, Chemistry,
and Medicine; Sporn, M. B., Roberts, A. B., Goodman, D. S., Eds.;
Raven Press: New York, 1994. (b) Hong, W. K.; Iti, L. M.
Retinoids in Human Cancer. In The Retinoids-Biology, Chem-
istry, and Medicine; Sporn, M. B., Roberts, A. B., Goodman, D.
S., Eds.; Raven Press: New York, 1994.
(2) (a) Armstrong, R. B.; Ashenfelter, K. O.; Eckhoff, C.; Levin, A.
A.; Shapiro, S. S. General and Reproductive Toxicology of
Retinoids. In The Retinoids-Biology, Chemistry, and Medicine;
Sporn, M. B., Roberts, A. B., Goodman, D. S., Eds.; Raven
Press: New York, 1994. (b) Kamm, J . J .; Ashenfleter, K. O.;
Ehmann, C. W. Preclinical and Clinical Toxicology of Selected
Retinoids. In The Retinoids; Sporn, M. B., Roberts, A. B.,
Goodman, D. S., Eds.; Academic Press: Orlando, 1984.
(3) Gross, E. G.; Helfgott, M. A. Retinoids and the Eye. Dermatol.
Clin. 1992, 10, 521-531.
(4) Waugh, K. M.; Berlin, K. D.; Ford, W. T.; Holt, E. M.; Carroll,
J . P.; Schomber, P. R.; Schiff, L. J . Synthesis and Characteriza-
tion of Selected Heteroarotinoids. Pharmacological Activity as
Assessed in Vitamin A Deficient Hamster Tracheal Organ
Cultures. Single-Crystal X-ray Diffraction Analysis of 4,4-
Dimethylthiochroman-6-yl Methyl Ketone 1,1-Dioxide and Ethyl
(E)-4-[2-(4,4-Dimethylthiochroman-6-yl)-1-propenyl]benzoate. J .
Med. Chem. 1985, 27, 116-124.
(5) Spruce, L. W.; Gale, J . B.; Berlin, K. D.; Verma, A. K.; Breitman,
T. R.; J i, X.; van der Helm, D. Novel Heteroarotinoids: Synthesis
and Biological Activity. J . Med. Chem. 1991, 34, 430-439.
(6) Spruce, L. W.; Rajadhyaksha, S. N.; Berlin, K. D.; Gale, J . B.;
Miranda, E. T.; Ford, W. T.; Blossey, E. C.; Verma, A. K.;
Hossain, M. B; van der Helm, D.; Breitman, T. R. Heteroaroti-
noids. Synthesis, Characterization, and Biological Activity in
Terms of an Assessment of These Systems to Inhibit Induction
of Ornithine Decarboxylase Activity and to Induce Terminal
Differentiation of HL- 60 Cells. J . Med. Chem. 1987, 30, 1474-
1482.
Recep tor Activa tion . The CV-1 cells were plated in 6-well
culture plates at a density of 7.5 × 104 cells/well and, after 24
h, were cotransfected with a retinoic acid receptor expression
vector (pECE-hRARR, pECE-hRARâ, pECE-hRARγ, pSG5-
mRXRR, pSG5-mRXRâ, or pSG5-mRXRγ), a retinoic acid
responsive reporter plasmid (âRARE-tk-CAT), and a trans-
fection control plasmid (pJ ATLac) using FuGENE6 transfec-
tion reagent (Boehringer Mannheim Corp., Indianapolis, IN).
The âRARE-tkCAT plasmid contained the chloramphenicol
acetyltransferase gene (CAT) driven by the thymidine kinase
(tk) promoter and the RARâ retinoic acid response element
(RARE). The pJ ATLac plasmid contained the â-galactosidase
(âGal) gene driven by the âactin promoter. About 16-18 h
after transfection, retinoids or DMSO (solvent) was added to
the cells. The retinoids were administered at 10-fold dilutions
of concentrations ranging from 10-5 to 10-10 M. The concentra-
tion of DMSO in all treated and control cultures was 0.01%
which was not cytotoxic to the cells. After incubation for 24 h,
cell lysates were prepared and assayed for reporter expression
using CAT and âGal ELISA kits (Boehringer Mannheim Corp.,
Indianapolis, IN) according to the manufacturer’s instructions.
The CAT expression levels were corrected for transfection
(7) Benbrook, D. M.; Madler, M. M.; Spruce, L. W.; Birckbichler, P.
J .; Nelson, E. C.; Subramanian, S.; Weerasekare, G. M.; Gale,
J . B.; Patterson, J r., M. K.; Wang, B.; Wang, W.; Lu, S.; Rowland,
T. C.; DiSivestro, P.; Lindamood, C.; Hill, D. L.; Berlin, K. D.
Biologically Active Heteroarotinoids Exhibit Anticancer Activity
and Decreased Toxicity. J . Med. Chem. 1997, 40, 3567-3583.
(8) Benbrook, D. M.; Subramanian, S.; Gale, J . B.; Liu, S.; Brown,
C. W.; Boehm, M. F.; Berlin, K. D. Synthesis and Characteriza-
tion of Heteroarotinoids Demonstrates Structure-Activity Re-
lationships. J . Med. Chem. 1998, 41, 3753-3757.

Heteroarotinoids Inhibit Cancer Cell Lines
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 21 4445
(9) (a) Mangelsdorf, D. J .; Umesono, K.; Evans, R. M. The Retinoid
Receptors. In The Retinoids-Biology, Chemistry, and Medicine;
Sporn, M. B., Roberts, A. B., Goodman, D. S., Eds.; Raven
Press: New York, 1994. (b) Dawson, M. I. Retinoids. In Burger’s
Medicinal Chemistry and Drug Discovery, 5th ed.; Wolff, M. E.,
Ed.; Wiley: New York, 1996.
(10) Cooper, J . S.; Pajak, T. F.; Rubin, P.; Tupchong, L.; Brady, L.
W.; Leibel, S. A.; Laramore, G. E.; Marcial, V. A.; Davis, L. W.;
Cox, J . D. Second Malignancies in Patients Who Have Head and
Neck Cancers: Incidence, Effect on Survival and Implications
for Chemoprevention Based on the RTOG Experience. Int. J .
Radiat. Oncol. Biol. Phys. 1989, 19, 449-456.
(11) Slaughter, D. P.; Southwick, H. W.; Smejkal, W. Field Cancer-
ization in Oral Stratified Squamous Epithelium: Clinical Im-
plications of Multicentric Origin. Cancer 1953, 6, 963-968.
(12) Lotan, R. Squamous Cell Differentiation Markers in Normal,
Premalignant, and Malignant Epithelium: Effects of Retinoids.
J . Cell. Biochem. 1993, 17, 167-174.
(13) Wolbach, S. B.; Howe, P. R. Tissue Changes Following Depriva-
tion of Fat Soluble A Vitamin. J . Exp. Med. 1925, 42, 753-777.
(14) Lippman, S. M.; Batsakis, J . G.; Toth, B. B.; Weber, R. S.; Lee,
J . J .; Martin, J . W.; Hays, G. L.; Goepfert, H.; Hong, W. K.
Comparison of Low-Dose Isotretinoin Therapy with â-Carotene
for Prevention of Oral Carcinogenesis. N. Engl. J . Med. 1993,
328, 15-20.
(27) Wang, B.; Liu, S.; Borchardt, R. T. Development of a Novel
Redox-Sensitive Protecting Group for Amines Which Utilize a
Facilitated Lactonization Reaction. J . Org. Chem. 1995, 60, 539-
543.
(28) Nakatani, K.; Numata, S.; Inoue, T.; Fujisawa, K. K.; Kawasaki,
T. I.; Toyama, T.; Tachibana, H.; Udagave, T.; Gohbara, M. UP
254135, U.S. Patent 4397864; Chem. Abstr. 1982, 97, 5964e.
(29) Boehm, M. F.; Zhang, L.; Zhi, L.; Badea, B. A.; White, S. K.;
Mais, D. E.; Berger, E.; Suto, M. C.; Goldman, M. E.; Heyman,
R. A. Synthesis and Structure-Activity Relationships of Novel
Retinoid X Receptor-Selective Retinoids. J . Med. Chem. 1994,
37, 2930-2941.
(30) Mangelsdorf, D. J .; Umesono, K.; Evans, R. M. In The Retinoids-
Biology, Chemistry, and Medicine; Sporn, M. B., Roberts, A. B.,
Goodman, D. S., Eds.; Raven Press: New York, 1994; pp 319-
349.
(31) Pfahl, M. Nuclear Receptor/AP-1 Interaction. Endocrinol. Rev.
1993, 14, 651-658.
(32) Chen, J . Y.; Penco, S.; Ostrowski, J .; Balaguer, P.; Pons, M.;
Starrett, J . E.; Preczek, P.; Chambon, P.; Gronemeyer, H. RAR-
Specific Agonist/Antagonists Which Dissociate Transactivation
and AP1 Tranrepression Inhibit Anchorage-Dependent Cell
proliferation. EMBO J . 1995, 14, 1187-1197.
(33) Fanjul, A.; Dawson, M. I.; Hobbs, P. D.; J ong, L.; Cameron, J .
F.; Harlev, E.; Graupner, G.; Lu, C.-P.; Pfahl, M. A Novel Class
of Retinoids with Selective Inhibition of AP-1 Exhibit Anti-
Proliferative Activity. Nature 1994, 372, 107-111.
(34) Grandis, J . R.; Zeng, Q.; Tweardy, D. J . Retinoic Acid Normalizes
the Increased Gene Transcription Rate of TGF-Alpha and EGFR
in Head and Neck Cancer Cell Lines. Nature Med. 1996, 2, 237-
240.
(35) Issing, W. J .; Wustrow, T. P. Expression of Retinoic Acid
Receptors in Squamous Cell Carcinomas and Their Possible
Implication in Chemoprevention. Anticancer Res. 1996, 16,
2373-2377.
(36) Oridate, N.; Esumi, N.; Lotan, D.; Hing, W. K.; Rochette-Egly,
C.; Chambon, P.; Lotan, R. Implication of Retinoic Acid Receptor
Gamma in Squamous Differentiation and Response to Retinoic
Acid in Head and Neck SqSCC/Y1 Squamous Carcinoma Cells.
Oncogene 1996, 12, 2019-2028.
(37) Lotan, R.; Dawson, M. I.; Zou, C.-C.; J ong, L.; Lotan, D.; Zou,
C.-P. Enhanced Efficacy of Combinations of Retinoic Acid- and
Retinoid X Receptor-Selective Retinoids and R-Interferon in
Inhibition of Cervical Carcinoma Cell proliferation. Cancer Res.
1995, 55, 232-236.
(38) Shalinsky, D. R.; Bischoff, E. D.; Lamph, W. W.; Zhang, L.;
Boehm, M. F.; Davies, P. J . A.; Nadzan, A. M.; Heyman, R. A. A
Novel Retinoic Acid Receptor-Selective Retinoid, ALRT1550, Has
Potent Antitumor Activity Against Human Oral Squamous
Carcinoma Xenograph in Nude Mice. Cancer Res. 1997, 57, 162-
168.
(39) Dhar, A.; Liu, S.; Klucik, J .; Berlin, K. D.; Madler, M. M.; Lu,
S.; Ivey, R. T.; Zacheis, D.; Brown, C. W.; Nelson, E. C.;
Birckbichler, P. J .; Benbrook, D. M. Synthesis, Structure-
Activity Relationships, and RARγ-Ligand Interactions of Nitro-
gen Heteroarotinoids. J . Med. Chem. 1999, 42, 3602-3614.
(40) Madler, M. M.; Klucik, J .; Soell, P. S.; Brown, C. W.; Lui, S.;
Berlin, K. D.; Benbrook, D. M.; Birckbichler, P. J .; Nelson, E. C.
LiOCl/NaOCl as a Novel Oxidative Mixture for Methyl Ketones
and Methyl Carbinols. Org. Prep. Proc. Int. 1998, 30, 230-234.
(41) Benbrook, D. M.; Lu, S.; Flangan, C.; Shen-Gunther, J .; Angros,
L. H.; Lightfoot, S. A. Biological Assay for Activity and Molecular
Mechanism of Retinoids in Cervical Tumor Cells. Gyn. Oncol.
1997, 66, 114-121.
(15) Chiesa, F.; Tradati, N.; Marazzara, M.; Rossi, N.; Boraccchi, p.;
Mariani, L.; Formelli, F.; Giardini, R.; Costa, A.; DePalo, G.;
Veronesi, U. Fenretinide (4HPR) in Chemoprevention of Oral
Leukoplakia. J . Cell. Biochem. 1993, 17 (Suppl), 255-261.
(16) Kurie, J . M.; Lippman, S. M.; Hong, W. K. Potential of Retinoids
in Cancer Prevention. Cancer Treat. Rev. 1994, 20, 1-10.
(17) Benner, S. E.; Pajak, T. F.; Lippman, S. M.; Early, C.; Hong, W.
K. Prevention of Second Primary Tumors with Isotretinoin in
Patients with Squamous Cell Carcinoma of the Head and Neck:
Long Term Followup. J . Natl. Cancer Inst. 1994, 86, 140-141.
(18) Dimery, I. W.; Hong, W. K.; Lee, J . J .; Guillory-Perez, C.; Pham.
F.; Fritsche, J r., H. A.; Lippman, S. M. Phase I Trial of Alpha-
Tocopherol Effects on 13-cis-Retinoic Acid Toxicity. Ann. Oncol.
1997, 8, 85-89.
(19) Shalinsky, D. R.; Bischoff, E. D.; Gregory, M. L.; Gottaris, M.
M.; Hayes, J . S.; Lamph, W. W.; Heyman, R. A.; Shirley, M. A.;
Cooke, T. A.; Davies, P. J .; Thomazy, V. Retinoid-Induced
Suppression of Squamous Cell Differentiation in Human Oral
Squamous Cell Carcinoma Xenographs (line 1483) in Athymic
Nude Mice. Cancer Res. 1995, 55, 3183-3191.
(20) Scher, R. L.; Saito, W.; Dodge, R. K.; Richtsmeier, W. J .; Fine,
R. L. Fenretinide-Induced Apoptosis of Human Head and Neck
Squamous Carcinoma Cell Lines. Otol. Head Neck Surg. 1998,
118, 464-471.
(21) Cooper, M. P.; Klassen, I.; Brakenhoff, R. H.; Cloos, J .; Snow,
G. B.; Braakhuis, B. J . All-trans-Retinoic Acid Induced Gene
Expression and Growth Inhibition in Head and Neck Cancer Cell
Lines. Oral Oncol. 1997, 33, 270-274.
(22) Eicher, S. A.; Lotan, R. Differential Effects of Retinoic Acid and
N-(4-Hydroxyphenyl)retinamide on Head and Neck Squamous
Cell Carcinoma Cells. Laryngoscope 1996, 106, 1471-1475.
(23) Lotan, R. Retinoids and Their Receptors in Modulation of
Differentiation, Development, and Prevention of Head and Neck
Cancers. Anticancer Res. 1996, 16, 2415-2419.
(24) Mourad, W. A.; Bruner, J . M.; Vallieres, E.; McName, C.;
Alabdulwahed, S.; Scott, K.; Oldring, D. J . The Effect of High
Dose Vitamin A on the Morphology and Proliferative Ability of
Xenograph Lung and Neck Cancer. In Vivo 1996, 10, 329-33.
(25) Kagechika, H.; Kawachi, E.; Hashimoto, I.; Himi, T.; Shudo, K.
Retinobenzoic Acids. 1. Structure-Activity Relationships of
Aromatic Acids With Retinoidal Activity. J . Med. Chem. 1988,
31, 2182-2192.
(26) Apfel, C.; Bauer, F.; Crettaz, M.; Forni, L.; Kamber, M.;
Kaufmann, F.; LeMotte, P.; Pirson, W.; Klaus, M. A Retinoic
Acid Receptor R Antagonist Selectivity Counteracts Retinoic Acid
Effects. Proc. Natl. Acad. Sci. U.S.A. 1992, 89, 7129-7133.
J M990292I