Recent studies have shown that nuclear transcription factor cyclic adenosine monophosphate response element binding protein (CREB) is overexpressed in many different types of cancers. Therefore, CREB has been pursued as a novel cancer therapeutic target. Naphthol AS-E and its closely related derivatives have been shown to inhibit CREB-mediated gene transcription and cancer cell growth. Previously, we identified naphthamide 3a as a different chemotype to inhibit CREB's transcription activity. In a continuing effort to discover more potent CREB inhibitors, a series of structural congeners of 3a was designed and synthesized. Biological evaluations of these compounds uncovered compound 3i (666-15) as a potent and selective inhibitor of CREB-mediated gene transcription (IC50 = 0.081 ± 0.04 μM). 666-15 also potently inhibited cancer cell growth without harming normal cells. In an in vivo MDA-MB-468 xenograft model, 666-15 completely suppressed the tumor growth without overt toxicity. These results further support the potential of CREB as a valuable cancer drug target.
Recent studies have shown that nuclear transcription factor cyclic adenosine monophosphate response element binding protein (CREB) is overexpressed in many different types of cancers. Therefore, CREB has been pursued as a novel cancer therapeutic target. Naphthol AS-E and its closely related derivatives have been shown to inhibit CREB-mediated gene transcription and cancer cell growth. Previously, we identified naphthamide 3a as a different chemotype to inhibit CREB's transcription activity. In a continuing effort to discover more potent CREB inhibitors, a series of structural congeners of 3a was designed and synthesized. Biological evaluations of these compounds uncovered compound 3i (666-15) as a potent and selective inhibitor of CREB-mediated gene transcription (IC50 = 0.081 ± 0.04 μM). 666-15 also potently inhibited cancer cell growth without harming normal cells. In an in vivo MDA-MB-468 xenograft model, 666-15 completely suppressed the tumor growth without overt toxicity. These results further support the potential of CREB as a valuable cancer drug target.
The cAMP-response element
binding protein (CREB) is a nuclear transcription
factor that can be activated to initiate gene transcription in response
to hormones, growth factors, and neuronal activity.[1,2] These
stimuli activate intracellular protein serine/threonine kinases such
as mitogen-activated protein kinase (MAPK), protein kinase A (PKA),
protein kinase B (PKB/Akt), and p90 ribosomal S6 kinase (p90RSK).[3] All these kinases have been shown
to be able to phosphorylate Ser133 in CREB.[1,3] Phosphorylation
at Ser133 is crucial in CREB’s binding with histone acetyl
transferase and mammalian transcription coactivator CREB-binding protein
(CBP) and its paralog p300 to initiate CREB-dependent gene transcription.
The binding interaction between CREB and CBP/p300 is mediated by the
activation domain in CREB called kinase-inducible domain (KID) and
KID-interacting (KIX) domain in CBP/p300.[4] Three protein phosphatases, protein phosphatase 1 (PP1),[5] protein phosphatase 2A (PP2A),[6] and phosphatase and tensin homolog (PTEN),[7] have been shown to dephosphorylate Ser133 in phosphorylated
CREB to turn off CREB-dependent gene transcription.The protein
kinases leading to CREB activation are frequently overactivated,
while the three phosphatases to dephosphorylate CREB are often inactivated
in various cancer cells. Therefore, it was predicted that CREB would
be overactivated in cancer cells. Consistent with this prediction,
CREB and phosphorylated CREB have been consistently shown to be overexpressed
in cancer tissues from brain,[8,9] breast,[10,11] lung,[12] prostate,[13] and bone marrow.[14] Because of
its aberrant activation in cancer cells, CREB has been pursued as
a novel cancer therapeutic target.[3] We
recently identified naphtholAS-E (1, Figure 1) as a cell-permeable inhibitor of CREB-mediated
gene transcription through inhibiting KID-KIX interaction,[15] the essential protein–protein interaction
to activate CREB-dependent gene transcription.[4] Consistent with the important roles of CREB in the maintenance of
cancer cells, we found that 1 and its close related derivatives
selectively inhibited proliferation of a large panel of cancer cell
lines from different organs in the low micromolar concentration range
without harming normal cells in vitro.[16]
Figure 1
Chemical
structures of previously reported CREB inhibitors: naphthol
AS-E (1) and compounds 2 and 3a. Compound 2 is rapidly transformed into 3a through an O,N-acyl transfer reaction
at pH 7.4.
Chemical
structures of previously reported CREB inhibitors: naphtholAS-E (1) and compounds 2 and 3a. Compound 2 is rapidly transformed into 3a through an O,N-acyl transfer reaction
at pH 7.4.During our course of studies to
improve the aqueous solubility
and biological activity of 1, we designed and synthesized
compound 2 (Figure 1). Compound 2 presented significantly improved antiproliferative activity
against a panel of different cancer cells.[17] Unexpectedly, we found that 2 was rapidly converted
into 3a under physiological conditions and was considered
as a prodrug of 3a, where a long-range O,N-acyl transfer reaction was involved (Figure 1).[17] While 2 displayed in vivo antibreast cancer activity, its CREB inhibition
potency remained modest.[17] In this report,
we detail our optimization of 3a and identification of 3i (666-15) as a potent CREB inhibitor with highly
efficacious in vivo antibreast cancer activity.
Results and Discussion
Analog
Design Rationale
A series of structural congeners
of 3a shown in Figure 2 was designed
to improve its biological activities and physicochemical properties.
Compound 3a contains a phenolic hydroxyl group that is
a potential site for glucuronidation, which would limit its metabolic
stability and bioavailability.[18−20] To test if this potential metabolic
liability can be removed without compromising bioactivity, compound 3b was designed to interrogate the role of the phenol group
in 3a in contributing to its bioactivity. Compound 3a also has relatively high polar surface area (PSA, 123.2
Å2) and high cLogP (5.30) (Table 1). To improve these two physicochemical parameters, compounds 3c,d were designed by removing one of the conjugated
planar naphthyl rings. Truncating one of the naphthyl ring systems
into a benzene system decreases the PSA to ∼98 Å2 and cLogP to ∼4.9 (Table 1). Compounds 3e–g were designed to probe the role of
the primary amino group in 3a. If this primary amino
group tolerates structural changes, additional functional groups may
be attached to the primary amino group. Analogs 3h–j were designed by varying the lengths of the linker and side
chain to understand their roles in biological activities. As presented
in Table 1, compounds 3e–j show decreased PSA and 3g–i also present decreased cLogP compared to 3a.
Figure 2
Chemical structures
of newly designed structural congeners (3b–j) of 3a. The key structural
difference between 3b–d and 3a is highlighted in red.
Table 1
Physiochemical Properties and CREB
Inhibition Activity of 3a–j
compd
PSAa (Å2)
cLogPa
CREB inhibition IC50 (μM)b
3a
123.2
5.30
2.22 ± 0.38
3b
86.9
6.41
4.69 ± 1.28
3c
98.0
4.91
10.05 ± 2.29
3d
97.9
4.92
5.30 ± 1.41
3e
96.1
5.42
>50
3f
92.45
5.87
18.53 ± 8.68
3g
105.8
5.26
7.30 ± 1.66
3h
100.2
4.43
0.30 ± 0.12
3i
100.2
4.83
0.081 ± 0.04
3j
100.0
5.39
5.23 ± 0.36
The polar surface
area (PSA) and
calculated log P (cLogP) values were computed
from their global energy minima using QikProp.
CREB inhibition refers to inhibition
of CREB-mediated gene transcription in HEK 293T cells using a CREB
reporter assay. The IC50 was presented as the mean ±
SD of at least two independent experiments in triplicate or >50
in
the cases where the IC50 was not reached at the highest
tested concentration (50 μM).
Chemical structures
of newly designed structural congeners (3b–j) of 3a. The key structural
difference between 3b–d and 3a is highlighted in red.The polar surface
area (PSA) and
calculated log P (cLogP) values were computed
from their global energy minima using QikProp.CREB inhibition refers to inhibition
of CREB-mediated gene transcription in HEK 293T cells using a CREB
reporter assay. The IC50 was presented as the mean ±
SD of at least two independent experiments in triplicate or >50
in
the cases where the IC50 was not reached at the highest
tested concentration (50 μM).
Chemistry
The synthesis of compounds 3b–j is presented in Schemes 1–7 and is overall similar to the synthesis
of 3a as described before.[17] All the final products were prepared in good to excellent yields.
This synthesis of 3b is shown in Scheme 1. Mitsunobu coupling (Ph3P/DEAD)[21] between 1 and Boc-protected 3-amino-1-propanol
(A1) gave 6b, whose Boc protecting group
was removed under acidic condition to generate free base 7b after neutralization with NaHCO3. Amide formation between
amine 7b and previously reported acid 5a(17) under the BOP/DIPEA coupling condition
yielded amide 8b. Deprotection of Boc in 8b with 2 N HCl delivered product 3b. Compound 3c was prepared in a similar fashion with the exception of a need for 7c as the key intermediate (Scheme 2). The commercially available starting materials methyl salicylate
(1c) and A1 were coupled together under
Mitsunobu reaction condition. Saponification of methyl ester 4c generated acid 5c, which was then coupled
with aniline 9 to yield 6c with MsCl as
the activating reagent.[22] The activating
reagent MsCl was found to be superior to BOP in achieving high selectivity
for forming desired amide versus the alternative undesired ester.[17] Removal of the Boc group from 6c provided amine 7c, which was further coupled with acid 5a followed by acidic deprotection of Boc to give desired
product 3c. The activating reagent MsCl was again found
to be superior to BOP in achieving high selectivity for forming desired
amide versus the alternative ester and was selected for all the subsequent
amide formation reactions. Compound 3d was prepared in
two steps by coupling between amine 7a(17) and acid 5c followed by Boc deprotection (Scheme 3).
Scheme 1
Synthesis of Compound 3b
Scheme 7
Synthesis of Compound 3j
Scheme 2
Synthesis of Compound 3c
Scheme 3
Synthesis of Compound 3d
The preparation of morpholine
substituted compound 3e is shown in Scheme 4. The morpholine side
chain was incorporated into 1a(23) by Mitsunobu reaction with A2. Saponification of ester 4e provided acid 5e, which was then coupled with
amine 7a by MsCl/Et3N to afford amide 8e. Treatment of 8e with HCl yielded corresponding
hydrochloride salt 3e. A similar sequence of reactions
was employed to prepare 3f and 3g. The side
chain in 4g was installed by O-alkylation of naphthol 1a, while the N-methyl group in 4f was introduced by methylation of carbamate 4a by NaH/MeI.[24]
Scheme 4
Synthesis of Compounds 3e–g
Compounds 3h–j having different
linker and side chain lengths were synthesized as shown in Schemes 5–7. Intermediate 7h was prepared essentially the same as that for 7a(17) with the use of A4 as
the Mitsunobu coupling partner followed by saponification, amide formation,
and Boc deprotection. Amide coupling between the amine 7h and acid 5h generated amide 8h, whose
Boc was removed under acidic condition to provide 3h.
Intermidiates 8i and 8j were prepared by
assembling building blocks 5a and 7h, 5h and 7a, respectively (Schemes 6 and 7).
Final deprotection of Boc in 8i and 8j delivered
desired compounds 3i and 3j uneventfully.
Scheme 5
Synthesis of Compound 3h
Scheme 6
Synthesis of Compound 3i
Inhibition of CREB-Mediated Gene Transcription by 3b–j
The newly synthesized final compounds 3b–j were evaluated for their activity
in inhibiting CREB-mediated gene transcription in HEK 293T cells using
a CREB Renilla luciferase (RLuc) reporter assay.[15] In this assay, HEK 293T cells were transfected
with a RLuc reporter under the control of a synthetic CREB promoter
containing three copies of cAMP-response elements (CRE). The transfected
cells were then treated with different concentrations of compounds
for 30 min before the induction of RLuc synthesis by forskolin (10
μM), an activator of adenylate cyclase to activate CREB’s
transcription activity.[25] The results from
this CREB reporter assay are summarized in Table 1, where the concentrations required to inhibit 50% of CREB’s
transcription activity (IC50) are shown. For comparison
purpose, the potency of previously reported compound 3a (IC50 = 2.22 μM) was also included in Table 1.[17]In comparison
to 3a, compound 3b without the phenol group
showed about 2-fold decrease of activity in CREB inhibition (IC50 = 4.69 μM), indicating that the phenol group in 3a has a beneficial effect on CREB inhibition. Therefore,
the rest of the compounds were designed to retain this crucial phenol
group. Compounds 3c and 3d, with one of
the naphthyl rings being trimmed down to a benzene ring, displayed
approximately 2- to 5-fold less potent CREB inhibition activity than 3a, suggesting that the two naphthyl rings could not be simplified
to phenyl rings without compromising CREB inhibitory activity. Replacement
of the primary amino group in 3a with morpholine (3e), N-methylamino (3f), or
hydroxyl (3g) group resulted in a total loss of CREB
inhibition activity for 3e (IC50 > 50 μM)
and significant decrease in CREB inhibition activity for 3f (IC50 = 18.53 μM) and 3g (IC50 = 7.30 μM). These data indicated that the primary amino group
at the side chain of 3a is critical for maintaining CREB
inhibition activity and not suitable for even minor modifications
like methylation.We then focused on the modification of the
lengths of the linker
between the two naphthyl rings and the side chain in compound 3a to interrogate their roles in CREB inhibition activity.
Specifically, compounds 3h–j with
two- or three-carbon chains were designed and synthesized. Gratifyingly,
compound 3h with a two-carbon linker and a two-carbon
side chain showed significantly increased CREB inhibition activity
(IC50 = 0.30 μM) compared to 3a. Furthermore,
compound 3i with a two-carbon linker and a three-carbon
side chain exhibited even more potent CREB inhibition activity (IC50 = 81 nM), which is approximately 28-fold improvement over 3a. The enhancement of CREB inhibition activity seen for 3i is very structure-specific because its structural isomer 3j, which has a three-carbon linker and two-carbon side chain,
was a much weaker CREB inhibitor with an IC50 of 5.23 μM.
The CREB inhibitory potency difference among 3a and 3h–j demonstrated that the length of the
linker between the two naphthyl rings can dramatically affect the
activity and the length of the side chain also has a critical role
on the CREB inhibition activity.
Inhibition of Cancer Cell
Proliferation by 3b–j
We
also evaluated the antiproliferative activity
of compounds 3b–j in four different
cancer cell lines: A549 (non-small-cell lung cancer), MCF-7 (breast
cancer), MDA-MB-231 (breast cancer), and MDA-MB-468 (breast cancer)
using the colorimetric MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium
bromide] assay.[16,26] The concentrations required to
inhibit 50% of the cancer cell growth (GI50) are presented
in Table 2. As reported before,[17] compound 3a was a submicromolar
inhibitor of proliferation of all the four cancer cell lines tested.
The analogs 3b–e all presented less
potent antiproliferative activity than 3a, in agreement
with their reduced CREB inhibition potency. Although 3e was inactive in the CREB reporter assay, it showed weak growth inhibition
activity in MDA-MB-468 cells (GI50 = 13.74 μM) but
no activity in the other three cancer cell lines. It is unlikely that
the weak activity in MDA-MB-468 cells was a result of inhibition of
CREB’s transcription activity. Similar discrepancy was also
observed for 3f, which exhibited weak CREB inhibition
activity while displaying robust antiproliferative activity in these
four cancer cell lines. The GI50 for 3f is
0.26, 1.65, 0.26, and 0.20 μM in A549, MCF-7, MDA-MB-231, and
MDA-MB-468 cells, respectively. Compound 3g displayed
modest CREB inhibition activity, but it was a rather weak inhibitor
of proliferation in all four cancer cell lines tested with GI50 values ranging from 10.19 to 82.67 μM. Finally, in
the series of compounds 3h–j with
different lengths of the linker and side chain, we observed that potent
CREB inhibitor 3i also potently inhibited cancer cell
growth. In MDA-MB-231 and MDA-MB-468 cells, the GI50 for 3i was 73 and 46 nM, respectively. In A549 and MCF-7 cells,
it exhibited robust activity as well with GI50 of 0.47
and 0.31 μM. Compared to 3i, 3h retained
reasonable CREB inhibition activity and inhibition of cancer cell
growth while 3j was much less potent. Therefore, compound 3i represents the most potent CREB inhibitor bearing potent
anticancer activity reported to date.[15−17,27,28]
Table 2
Antiproliferative
Activities of 3a–ja
GI50 (μM)
compd
A549
MCF-7
MDA-MB-231
MDA-MB-468
3a
0.29 ± 0.01
0.14 ± 0.03
0.37 ± 0.13
0.22 ± 0.07
3b
0.95 ± 0.77
1.72
1.00 ± 0.58
1.62 ± 0.48
3c
2.44 ± 0.40
2.30 ± 0.67
4.61 ± 2.20
1.82 ± 0.20
3d
2.32 ± 0.35
2.66 ± 1.30
5.82 ± 4.61
2.32 ± 0.39
3e
>100
>100
>100
13.74 ± 3.98
3f
0.26 ± 0.06
1.65 ± 0.44
0.26 ± 0.04
0.20 ± 0.02
3g
39.15 ± 32.09
28.99 ± 10.27
82.67 ± 24.52
10.19 ± 1.47
3h
1.16 ± 0.05
0.81 ± 0.78
1.21 ± 0.15
0.25 ± 0.01
3i
0.47 ± 0.06
0.31 ± 0.10
0.073 ± 0.04
0.046 ± 0.04
3j
2.17 ± 0.11
1.89 ± 0.60
2.71 ± 0.32
1.85 ± 0.23
GI50 is the concentration
required to inhibit the cancer cell growth by 50% as evaluated by
the MTT assay. The compounds were incubated with cells for 72 h. The
GI50 was presented as mean ± SD of at least two independent
experiments in duplicates or >100 in the cases where GI50 was not reached at the highest tested concentration (100 μM).
When SD was not presented, only one experiment was performed in duplicate.
GI50 is the concentration
required to inhibit the cancer cell growth by 50% as evaluated by
the MTT assay. The compounds were incubated with cells for 72 h. The
GI50 was presented as mean ± SD of at least two independent
experiments in duplicates or >100 in the cases where GI50 was not reached at the highest tested concentration (100 μM).
When SD was not presented, only one experiment was performed in duplicate.Previously, it was shown that 3a only weakly inhibited
CREB-CBP interaction (IC50 = 19.72 ± 1.78 μM)
as assayed by a split RLuc complementation assay.[17] We also investigated if the more potent CREB inhibitor 3i could inhibit CREB-CBP interaction using the same assay.
It was also found to be a rather weak inhibitor of CREB-CBP interaction
with IC50 = 18.27 ± 2.81 μM. We conclude that 3i inhibits CREB’s transcription activity in living
cells independent of direct CREB or CBP binding interaction. Further
studies are needed to understand if 3i will modulate
the upstream components of CREB activation.[3] Or alternatively, an unbiased chemoproteomics approach[29] may be utilized to identify the direct target
of 3i to understand its mechanism of inhibiting CREB-mediated
gene transcription.The results presented above showed that
the bioactivities of 3a are very sensitive to structural
modifications to either
increase or decrease its activity. The physicochemical property parameters
like PSA[30] and cLogP[31] that are associated with cell membrane permeability do
not seem to be the major determinants. For example, compounds 3h–j bear similar PSA, but their bioactivities
do not correlate with their cLogP (Table 1).
To identify the structural basis for the observed bioactivity differences
among 3a and 3h–j, we
performed conformational searches to identify their global conformational
minima using MacroModel. The conformational ensemble was generated
by systematically rotating all the rotatable bonds in 3a and 3h–j. The identified global
conformational minima are shown in Figure 3. All the four compounds form an intramolecular hydrogen bond between
the protonated ammonium nitrogen and amide carbonyl oxygen. However,
the more potent CREB inhibitors 3h and 3i adopt a more compact conformation by forming π–π
stacking interaction between one of the naphthyl rings and chlorophenyl
ring (Figure 3). On the other hand, the same
naphthyl ring in the less potent compounds 3a and 3j do not form π–π stacking interaction
with the chlorophenyl ring by assuming a more extended conformation
at their global minima. These differences suggest that the unique
conformation associated with 3h and 3i may
contribute to their potent CREB inhibitory activity and antiproliferative
activity.
Figure 3
Conformation of the global energy minimum of 3a (A), 3j (B), 3h (C), and 3i (D). The
distance between the amide oxygen and ammonium nitrogen was labeled
in each conformation to indicate formation of an intramolecular hydrogen
bond.
Conformation of the global energy minimum of 3a (A), 3j (B), 3h (C), and 3i (D). The
distance between the amideoxygen and ammonium nitrogen was labeled
in each conformation to indicate formation of an intramolecular hydrogen
bond.
In the CREB RLuc reporter assay with transfected
HEK 293T cells, compound 3i was very potent in inhibiting
CREB’s transcription activity. In order to investigate if 3i also inhibited endogenous CREB target gene expression,
the transcript level of nuclear receptor related 1 protein (Nurr1/NR4A2), a well-defined CREB target gene in HEK 293T
cells, was evaluated.[17,32] The cells were treated with 3i followed by stimulation with forskolin (10 μM). Then
the relative mRNA of Nurr1/NR4A2 was determined by
quantitative reverse transcription polymerase chain reaction (qRT-PCR).
As shown in Figure 4, forskolin robustly stimulated Nurr1/NR4A2 level to ∼31-fold. 3i dose-dependently
inhibited transcription of Nurr1/NR4A2. Significant
inhibition was observed even at 50 nM of 3i. In contrast,
the weaker CREB inhibitor 3a only started to show significant
inhibition at 1000 nM (Figure S1 in Supporting
Information). These results are consistent with those from
the CREB reporter assay.
Figure 4
Compound 3i decreased endogenous
CREB target gene
expression. HEK 293T cells were treated with different concentrations
of 3i followed by treatment with forskolin. Then the
relative mRNA level of Nurr1/NR4A2 was determined
by qRT-PCR analysis.
Compound 3i decreased endogenous
CREB target gene
expression. HEK 293T cells were treated with different concentrations
of 3i followed by treatment with forskolin. Then the
relative mRNA level of Nurr1/NR4A2 was determined
by qRT-PCR analysis.To investigate 3i’s selectivity on different
transcription activators, we employed RLuc reporter assays to monitor
individual transcription factor activity in HEK 293T cells. VP16-CREB
is a fusion protein by fusing the potent activation domain VP16 to
the full-length CREB. It requires CREB-CRE interaction for transcriptional
activation, but it is a constitutively active transcription factor
independent of phosphorylation as opposed to wild type CREB.[33] As shown in Table 1 and
Figure 5, 3i potently (IC50 = 81 nM) and efficaciously inhibited CREB’s transcription
activity in HEK 293T cells. On the other hand, it showed much less
efficacious inhibition of VP16-CREB and p53-mediated gene transcription.
And even this weak inhibition only occurred at high concentrations
(>1 μM). In a separate transcription reporter assay with
NF-κB,
much higher concentrations of 3i were required to inhibit
NF-κB-mediated gene transcription (IC50 = 5290 nM, Figure S2), which is distinct from 1 and its phosphate.[32] Collectively, these
results indicate that 3i selectively inhibited CREB-mediated
gene transcription.
Figure 5
Compound 3i selectively inhibited CREB-mediated
gene
transcription. HEK 293T cells were transfected with indicated combinations
of plasmids. Then the cells were treated with different concentrations
of 3i before RLuc activity measurement. Forskolin (Fsk,
10 μM) was added to CRE-Rluc only transfected cells at 30 min
after drug treatment to stimulate CREB’s activity. The RLuc
activity was normalized to the protein concentration and presented
as relative luciferase unit (RLU)/μg protein.
Compound 3i selectively inhibited CREB-mediated
gene
transcription. HEK 293T cells were transfected with indicated combinations
of plasmids. Then the cells were treated with different concentrations
of 3i before RLuc activity measurement. Forskolin (Fsk,
10 μM) was added to CRE-Rluc only transfected cells at 30 min
after drug treatment to stimulate CREB’s activity. The RLuc
activity was normalized to the protein concentration and presented
as relative luciferase unit (RLU)/μg protein.
Compound 3i Selectively Inhibited
the Growth of
Cancer Cells but Not Normal Cells
With a potent and specific
CREB inhibitor 3i in hand, we tested if it was toxic
to normal cells. Previous genetic studies have shown that normal cells
tolerate well with reduced levels of CREB.[34,35] As shown in Table 2 and Figure 6A,B, 3i potently inhibited growth of MDA-MB-231
and MDA-MB-468 cells with GI50 in the midnanomolar concentration
range. On the other hand, no significant inhibition of growth was
observed in two different normal cell lines, human mammary epithelial
cells (HMEC) and human foreskin fibroblasts (HFF), up to 1 μM
concentration, which is more than 10-fold higher than its GI50 in MDA-MB-231 and MDA-MB-468breast cancer cells. This selective
toxicity is in strong contrast to conventional chemotherapeutics like
doxorubicin, which did not show differential toxicity between cancer
and normal cells under the same assay conditions.[36] Therefore, pharmacological inhibition of CREB’s
transcription activity is well tolerated in normal cells, which is
consistent with the idea of cancer cells’ addiction to CREB.[3,37,38]
Figure 6
Compound 3i selectively inhibited
tumor cell growth.
Shown are antiproliferative dose–response curves of 3i in breast cancer MDA-MB-468 (A) and MDA-MB-231 (B) cells as well
as normal HMEC (C) and HFF (D) cells. The cells were incubated with 3i for 72 h, and then the remaining live cells were quantified
by the MTT assay.
Compound 3i selectively inhibited
tumor cell growth.
Shown are antiproliferative dose–response curves of 3i in breast cancerMDA-MB-468 (A) and MDA-MB-231 (B) cells as well
as normal HMEC (C) and HFF (D) cells. The cells were incubated with 3i for 72 h, and then the remaining live cells were quantified
by the MTT assay.
Compound 3i Completely Suppressed the Tumor Growth
in Vivo
The selective in vitro toxicity of 3i against cancer cells versus normal cells prompted us to investigate
its in vivo antitumor activity. Preliminary toxicity studies showed
that intraperitoneal (ip) injection of 10 mg/kg of 3i is well tolerated in mice (Figure S3).
This dose was chosen for in vivo antitumor efficacy studies in the
MDA-MB-468 xenografts. The MDA-MB-468tumor was allowed to grow to
an average size of 100 mm3 in nude mice. Then the mice
were randomized to receive either vehicle or 3i at 10
mg/kg once a day, 5 days per week for 5 weeks by ip injection. The
tumor volumes and body weights were measured 2–3 times/week.
The data in Figure 7A showed the tumor growth
in the mice treated with 3i was efficaciously inhibited
with complete tumor stasis. During the same period, the tumor volume
in the vehicle-treated group was more than tripled (Figure 7A). The body weights of 3i-treated
animals and vehicle-treated ones were indistinguishable from each
other during the entire treatment period (Figure 7B), indicating no overt toxicity with this compound treatment.
These results are consistent with in vitro studies with compound 3i (Figure 6C,D) where normal cells
could tolerate much higher concentrations of 3i than
cancer cells. These data further support the notion that pharmacologically
targeting CREB is a promising strategy for development of novel cancer
therapeutics.
Figure 7
Compound 3i suppressed MDA-MB-468 tumor growth
in
vivo. MDA-MB-468 tumor-bearing mice were treated either with vehicle
or with 3i at 10 mg/kg once a day for 5 days a week.
The duration of treatment was 5 weeks. The relative tumor volume (A)
and body weight (B) of the treated mice are shown: (∗) P < 0.05 by Student’s t test.
Compound 3i suppressed MDA-MB-468tumor growth
in
vivo. MDA-MB-468tumor-bearing mice were treated either with vehicle
or with 3i at 10 mg/kg once a day for 5 days a week.
The duration of treatment was 5 weeks. The relative tumor volume (A)
and body weight (B) of the treated mice are shown: (∗) P < 0.05 by Student’s t test.
Conclusion
In
an important extension of previous work,[17] we have prepared a series of naphthamide derivatives based
on the structure of 3a. Overall, the observed antiproliferative
activities of these naphthamides correlated well with their CREB inhibition
activity. Structure–activity relationships observed for members
of this series revealed that many structural elements present in 3a are crucial for maintaining CREB inhibition and cancer
cell growth inhibition activity. The phenol in the chlorophenyl ring,
the primary amino group in the side chain, and two naphthyl rings
in 3a are all important for maintaining 3a’s bioactivity. Importantly, the carbon chain length of the
linker between the two naphthyl rings, and the carbon chain length
of the side chain are absolutely critical for optimal activities.
We identified compound 3i, which we named as 666-15, as a potent and efficacious inhibitor toward CREB-mediated gene
transcription. 666-15 also displayed potent and efficacious
growth inhibition activity against cancer cells in vitro and in vivo. 666-15 should give us a new tool to further investigate CREB
signaling.[39]
Experiment
Section
Chemistry. General
Glass Contout solvent purification
system was used to purify all the anhydrous solvents to be used for
reactions. Melting points were determined in capillary tubes using
Mel-Temp and are uncorrected. All 1H and 13C
NMR spectra were obtained in a Bruker Avance 400 MHz spectrometer
using CDCl3 or DMSO-d6 as the
solvent, and the chemical shifts of the residual CHCl3 (δ
7.24) or DMSO (δ 2.50) were taken as references. Chemical shifts
(δ) are reported in parts per million (ppm), and the signals
are described as brs (broad singlet), d (doublet), dd (doublet of
doublet), td (triplet of doublet), m (multiplet), q (quartet), s (singlet),
and t (triplet). Coupling constants (J values) are
given in Hz. Silica gel flash chromatography was performed using 230–400
mesh silica gel (EMD). All reactions were monitored using thin-layer
chromatography (TLC) on silica gel plates (EMD). Yields were of purified
compounds. All final compounds for biological evaluations were confirmed
to be of >95% purity based on reverse phase HPLC (Waters, Milford,
MA) analysis using an XBridge C18 column (4.6 mm × 150 mm) and
detected at 254 nm. The mobile phases for HPLC are water and acetonitrile,
both of which contained 0.1% TFA. The mass spectra were obtained from
a Thermo Electron LTQ-Orbitrap Discovery high resolution mass spectrometer
(Thermo Scientific) with electrospray operated in either positive
or negative mode.
General Procedure A: Mitsunobu Reaction
To a solution
of phenol (1 equiv), alcohol (1.2–1.5 equiv), and PPh3 (1.2–1.5 equiv) in THF (1.5–2 mL/mmol) was added DEAD
(1.2–1.5 equiv) in THF (0.2–0.3 mL/mmol) dropwise at
0 °C. The reaction mixture was stirred at room temperature overnight.
The solvent was removed under reduced pressure and the residue was
purified by silica gel flash column chromatography to give the corresponding
product.
General Procedure B: Saponification of the Methyl Esters with
LiOH
To a solution of methyl ester (1 equiv) in MeOH–THF–water
(1:1:1, 9 mL/mmol) was added LiOH·H2O (5 equiv) at
room temperature. The resulting mixture was stirred at room temperature
overnight. The organic solvents were removed under reduced pressure,
and the residue was acidified with 2 N HCl at 0 °C to pH ∼2
(pH ∼7 for 5e). The reaction mixture was extracted
with ethyl acetate or THF for 5e. The organic layer was
separated and washed with brine and dried over Na2SO4. The solution was filtered and the solvent was evaporated
to give the corresponding acid.
General Procedure C: Amide
Formation by MsCl and TEA
To a stirred solution of 5 (1 equiv) and TEA (1.0 equiv)
in THF (3 mL/mmol) was added MsCl (1 equiv) dropwise at 0 °C.
The reaction mixture was stirred at 0 °C for 30 min, when the
corresponding ammonium salt 7 or aniline 9 (1 equiv) was added. The reaction mixture was stirred at room temperature
overnight. Another portion of TEA (1.0 equiv) was added if the salt 7 was used. The reaction mixture was diluted with 5% NaHCO3 and extracted with ethyl acetate. The organic layer was separated,
washed with brine, and dried over Na2SO4. The
solution was filtered and the solvent was removed to give a residue,
which was purified by silica gel flash column chromatography to yield
the corresponding amide.
General Procedure D: Removal of the Boc and
MOM with 2 N HCl
An HCl solution in Et2O (2 M,
2–10 equiv) was
added to a stirred solution of 6 or 8 (1.0
equiv) in CHCl3–MeOH (1:1, 6–10 mL/mmol).
The resulting mixture was stirred at room temperature overnight. The
solvents were removed under reduced pressure, and the solid was treated
with acetone or ethyl ether. The solid was collected by filtration
to give the corresponding product.
HEK
293T, HFF and A549 cell lines were
obtained from American Tissue Culture Collection (ATCC, Manassas,
VA). MDA-MB-231, MDA-MB-468, and MCF-7 cell lines were obtained from
Developmental Therapeutics Program at the National Cancer Institute
(NCI). The cells were maintained in DMEM (Life Technologies, Grand
Island, NY) with nonessential amino acids (Life Technologies) and
10% (v/v) HyClone fetal bovine serum (FBS, GE Healthcare Life Science,
Logan, UT) at 37 °C with 5% CO2. HMEC cells were obtained
from Lonza (Walkersville, MA) and were cultured in MEGM complete media
(Lonza) supplemented with 10 μg/mL penicillin and 10 μg/mL
streptomycin (Life Technologies) at 37 °C under 5% CO2.
Inhibition of CREB-Mediated Gene Transcription
HEK
293T cells in a 10 cm plate were transfected with pCRE-RLuc (6 μg)
with Lipofectamine2000 (Life Technologies) following the
manufacturer’s instructions. Three hours after transfection,
the cells were collected and replated into 96-well plates at ∼10 000
cells/well. The cells were allowed to attach to the bottom of the
plates overnight. The cells were then treated with different concentrations
of different compounds for 30 min, when forskolin (10 μM) was
added to each well. The cells were incubated for further 5 h before
cell lysis using 1× 30 μL Renilla luciferase
lysis buffer (Promega, Madison, WI). An amount of 5 μL of the
lysate was combined with 30 μL of benzyl-coelenterazine (Nanolight,
Pinetop, AZ) solution in PBS (pH 7.4, 10 μg/mL). The protein
concentration in each well was determined by Dye Reagent Concentrate
(Bio-Rad, Hercules, CA). The Renilla luciferase activity
was normalized to protein content in each well and expressed as relative
luciferase unit/μg protein (RLU/μg protein). The IC50 was derived from nonlinear regression analysis of the RLU/μg
protein–concentration curve in Prism 5.0 (La Jolla, CA).
Cell Growth Inhibition Assay
The growth inhibition
of different cell types was assessed by MTT assay using MTT reagent
(Sigma, St. Louis, MO). Briefly, the cells were plated into 96-well
plates and the cells were allowed to attach to the bottom of the plates
overnight. Then the cells were treated with different concentrations
of different drugs for 72 h. The media were removed, and MTT reagent
in complete tissue culture media (0.5 mg/mL) was added to each well
and incubated at 37 °C for 3 h. The incubation media were removed
and 100 μL of DMSO was added to each well. The absorbance of
the formed purple formazan solution was read at 570 nm using Packard
Fusion plate reader. The percent of growth is defined as 100 ×
(Atreated – Ainitial)/(Acontrol – Ainitial), where Atreated represents absorbance in wells treated with a compound, Ainitial represents the absorbance at time 0,
and Acontrol denotes media-treated cells.
The GI50 was derived from nonlinear regression analysis
of the percent of growth–concentration curve in Prism 5.0.
qRT-PCR
The qRT-PCR assay was carried out essentially
in the same way as described before.[17] Briefly,
HEK 293T cells were treated with different compounds for 1 h followed
by treatment with DMSO or forskolin (10 μM) for 45 min. Then
total RNA was isolated and treated with DNase I using NucleoSpin RNA
kit (Clontech). The first-strand cDNA was synthesized using SuperScript
III First-Strand Synthesis System for RT-PCR (Invitrogen). Quantitative
real-time PCR was performed on QuantStudio 7 Flex using SYBR Advantage
qPCR Premix (Clontech). The 2–ΔΔCT method
was used to analyze the relative changes in gene expression[16,40] with hypoxanthine phosphoribosyltransferase 1 (HPRT) as the reference gene. The primers used were the same as before.[17]
In Vivo Xenograft Study
All the
procedures for animal
handling, care, and the treatment in this study were performed according
to the guidelines approved by the Institutional Animal Care and Use
Committee (IACUC) of Oregon Health & Science University following
the guidelines of the Association for Assessment and Accreditation
of Laboratory Animal Care (AAALAC). Each 6- to 8-week old BALB/c nude
mouse (Charles River Laboratories) was inoculated subcutaneously at
the right flank with MDA-MB-468 cells (5 × 106) in
0.1 mL of HBSS with Matrigel (1:1) for tumor development. When the
tumor volume reached approximately 100 mm3, the mice were
randomized to be treated with either vehicle or 3i at
10 mg/kg. 3i was dissolved in 1% N-methylpyrrolidone
(NMP), 5% Tween-80 in H2O. The dosing solution was prepared
weekly. The mice were treated once a day for 5 consecutive days a
week, and the treatment lasted for 5 weeks. During the treatment,
the tumor size and body weight were measured 2–3 times a week.
The tumors were measured in two dimensions using a digital caliper,
and the volume was expressed in mm3 using the formula V = 0.5ab2, where a and b represent the long and short diameters of
the tumor, respectively. The tumor volume was normalized to the initial
tumor volume at the time of the first treatment. Student t-test was used for statistical analysis.
Molecular Modeling
All the molecular modeling studies
were conducted in Schrodinger Small Molecule Drug Discovery Suite
(Portland, OR). All the structures were first optimized using MacroModel
with MMFFs force field in the absence of any solvent. The charges
were from the force field. Powell–Reeves conjugate gradient
(PRCG) minimization algorithm was used, and all minimizations were
converged to 0.05 kJ mol–1 Å–1. To identify the global conformational minimum for each compound,
the minimized structure was subjected to a conformational search by
rotating all the rotatable bonds in each structure. MMFFs force field
was used. The global energy minimum was taken for cLogP and PSA calculation
using QikProp module.
Authors: D Wu; H E Zhau; W-C Huang; S Iqbal; F K Habib; O Sartor; L Cvitanovic; F F Marshall; Z Xu; L W K Chung Journal: Oncogene Date: 2007-02-19 Impact factor: 9.867
Authors: Laura Rodón; Alba Gonzàlez-Juncà; María del Mar Inda; Ada Sala-Hojman; Elena Martínez-Sáez; Joan Seoane Journal: Cancer Discov Date: 2014-08-01 Impact factor: 39.397
Authors: B E Wadzinski; W H Wheat; S Jaspers; L F Peruski; R L Lickteig; G L Johnson; D J Klemm Journal: Mol Cell Biol Date: 1993-05 Impact factor: 4.272
Authors: Shu Z Wiley; Krishna Sriram; Wenjing Liang; Sarah E Chang; Randall French; Thalia McCann; Jason Sicklick; Hiroshi Nishihara; Andrew M Lowy; Paul A Insel Journal: FASEB J Date: 2018-01-03 Impact factor: 5.191
Authors: Paul M Daniel; Gulay Filiz; Daniel V Brown; Michael Christie; Paul M Waring; Yi Zhang; John M Haynes; Colin Pouton; Dustin Flanagan; Elizabeth Vincan; Terrance G Johns; Karen Montgomery; Wayne A Phillips; Theo Mantamadiotis Journal: Neuro Oncol Date: 2018-09-03 Impact factor: 12.300