Identification and structure-activity relationship studies of 3-methylene-2-norbornanone as potent anti-proliferative agents presumably working through p53 mediated apoptosis

Article abstract of DOI:10.1016/j.bmcl.2004.08.048

Methylene-2-norbornanone (3) showed potent activity (LC₅₀ = 3-8 μM) and selectivity for mutant p53 cell types in anti-proliferation assays. structure-activity relationship (SAR) studies of compound 3 led to the generation of a series of compounds with high levels of anti-proliferative activities. We have identified a novel series of α-methylene carbonyl compounds through structure-activity relationship (SAR) studies with high levels of anti-proliferative activities. The lead molecule, 3-methylene-2-norbornanone (3) showed potent activity (LC₅₀ = 3-8 μM) against mutant p53 cell types and many fold selectivity (>13-29) over wild-type p53 cells. Further, compound 3 and its analogs showed refolding of mutant p53 protein comparable to their anti-proliferative activities suggesting possible interaction with mutant p53 protein.

Full text of DOI:10.1016/j.bmcl.2004.08.048

Bioorganic & Medicinal Chemistry Letters 14 (2004) 5645–5649
Identification and structure–activity relationship studies of
3-methylene-2-norbornanone as potent anti-proliferative
agents presumably working through p53 mediated apoptosis
N. Laxma Reddy,* Jeanette Hill, Long Ye, Prabhavathi B. Fernandes and David M. Stout
Ricerca Biosciences, 7528 Auburn Road, Concord, OH 44077, USA
Received 21 July 2004; revised 19 August 2004; accepted 19 August 2004
Available online 16 September 2004
Abstract—We have identified a novel series of a-methylene carbonyl compounds through structure–activity relationship (SAR)
studies with high levels of anti-proliferative activities. The lead molecule, 3-methylene-2-norbornanone (3) showed potent activity
(LC50 = 3–8lM) against mutant p53 cell types and many fold selectivity (>13–29) over wild-type p53 cells. Further, compound 3
and its analogs showed refolding of mutant p53 protein comparable to their anti-proliferative activities suggesting possible inter-
action with mutant p53 protein.
ꢀ 2004 Elsevier Ltd. All rights reserved.
1. Introduction
very short thus its concentration must be strictly regu-
lated because too little p53 will cause tumor prolifera-
tion and too much p53 will result in death of normal
cells. Furthermore, even a single point mutation in
TP53 causes considerable conformational change in
the resulting protein, thereby rendering this Ômutant
p53Õ protein inactive toward the induction of apoptosis.
Indeed, more than half of all human cancer tumor cells
are known to express high levels of mutant p53. The mu-
tant p53 acts as a dominant negative inhibitor of wild-
type p53, which functions as a tetramer.3 In addition,
some of the tumor-associated mutant p53 also acquires
new transforming functions that contribute to tumor
development.4 The activity of wild-type p53 is also
dependent on the level of MDM2 protein that destroys
p53.5 MDM2 is essential for normal cell proliferation,
otherwise an uncontrolled level of wild-type p53 causes
apoptosis of normal cells.
One of the most promising, but challenging approaches
for the treatment of cancer involves the selective induc-
tion of apoptosis in tumor cells. It is well recognized that
one of the causes of cancer is the perturbation of the
intricate balance (homeostasis) between cell growth
and cell death and that the tumor suppressor gene,
TP53, is intimately connected with this process. The
mutation of the TP53 gene is linked to the proliferation
of various types of tumors, such as colon, glioblastoma,
breast, lung, and osteosarcoma. The gene product of
TP53 is the tumor suppressor protein p53 (also referred
to as Ôwild-typeÕ p53), which is directly involved in
inducing apoptotic cell death in response to various
stress signals encountered during the development and
progression of malignant tumors.1 In addition to having
tumor suppression properties, p53 is also implicated in
neurodegenerative disorders.2
There has been considerable effort to develop small mole-
cules that either enhance the stability of wild-type p53
or restore the function of mutant p53. Recently, it was
reported (Fig. 1) that an amino pyrimidine derivative
(1) stabilizes wild-type p53 and restores the active con-
formation to mutant p53.6 More recently, it was shown
that a quinuclidinone derivative (2) and certain male-
imide derivatives, identified in screening a library of
low-molecular weight compounds, restores the DNA
binding activity of mutant p53 in tumor cells, reportedly
through restoration of the active conformation to
The activity of p53 protein is dependent not only on its
concentration and stability, but also on the maintenance
of its intact native conformation. The half-life of p53 is
Keywords: 3-Methylene-2-norbornanone; Mutant p53 protein; Anti-
proliferative activity; Apoptosis.
*
Corresponding author. Tel.: +1 440 357 3369; fax: +1 440 354
6276; e-mail: reddy_l@ricerca.com
Present address: DarPharma, Inc., 215 Cloister Ct., Chapel Hill,
NC 27514, USA.
0960-894X/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.08.048

5646
N. L. Reddy et al. / Bioorg. Med. Chem. Lett. 14 (2004) 5645–5649
at 37ꢁC, 5% CO2. The number of viable cells was meas-
HN
N
ured by mitochondrial reduction of Alamar Blue to a
fluorescent dye. LC50 values for each compound were
determined and the activity for mutant versus wild-type
p53 was compared. Compound 2 was used as a positive
control and vehicle as a negative control. Desirable re-
sults of a compound would have low LC50 concentra-
tion (sub-lM) and high selectivity for mutant p53 cells.
CH2OH
CH2OH
N
N
N
O
1
2
OMe
Figure 1.
Several simple a-methylene carbonyl-containing com-
pounds were identified in a primary-round of anti-pro-
liferative assays. One of the molecules, 3-methylene-2-
norbornanone (3) showed potent activity and selectivity
for mutant p53 cell types over wild-type p53. It was re-
ported earlier that natural products containing sesquit-
erpene lactones or compounds containing the a-
methylene carbonyl group are cytotoxic and potential
anti-tumor agents.9 These reports supported our find-
ings and led us to investigate further the structure–activ-
ity studies based on 3-methylene-2-norbornanone (3).
mutant p53.7 Compound 2 also induced apoptosis in a
range of cell lines containing mutations in the p53
DNA binding domain, but was less toxic to cells con-
taining wild-type p53. It has also been shown that 2-
methoxy estradiol facilitates wild-type p53-mediated
apoptosis.8 The effectiveness of known compounds that
target mutant p53 is sub-optimal, as compound 1 stabi-
lizes only the newly synthesized p53 protein, while com-
pound 2 does not reactivate all the known mutant p53
proteins. Thus, there continues to exist the need to de-
velop novel small molecules that increase the concentra-
tion of wild-type p53 in tumors or restores the activity of
mutant p53 or prevents the degradation of wild-type
p53, or any combination of these processes. Therefore,
we have undertaken a study to screen small-molecule
compound libraries against mutant and wild-type p53
cell based anti-proliferative assays to identity novel
chemical entities.
In a first attempt to understand the relative importance
of each of the functional group of compound 3 for anti-
proliferative activity, we have prepared several direct
analogs (4, 5, 6) of compound 3 through simple chemical
transformations and/or obtained compounds from com-
mercial sources (Fig. 2).10 For example, compounds 4
and 5 were obtained commercially, while compound 6
was prepared as mixture of exo and endo isomers from
reduction of 3 with NaBH4 in ethanol. Compound 7
was obtained as a mixture of E and Z isomers by direct
condensation of 3 with hydroxylamine hydrochloride.
Compounds 8–11 were prepared by reacting the corre-
sponding tetralones and indanones with formalin/acetic
acid followed by diethyl amine at 100ꢁC.11 Similarly,
compound 14 and camphor analogs 15 and 16 were pre-
pared from commercially available tricyclodecane-8-one
and D- and L-camphor, respectively. Compound 17 re-
sulted during the introduction of an a-methylene group
All compounds were tested for anti-proliferative activity
(Tables 1 and 2) using two cell lines with single point
mutations in the DNA binding region: C33A (human
cervical cells with a mutation at Cys 273) and Ramos
RA1 (human lymphocyte with a mutation at Asp 254).
Caki-1, human renal carcinoma cells with wild-type
p53, were used as a negative control. The cells were pla-
ted at sub-confluent levels (1 · 105 cells/mL) in RPMI
media with 10% fbs and incubated with the compounds
at concentrations between 0.001 and 100lg/mL for 72h
Table 1. Structure–activity relationships of 3-methylene-2-norbornanone
LC50 (lM)a
Selective activity against mutant p53 (fold increase)e
Compound #
Caki-1b
C33Ac
Ramos RA1d
Caki/C33A
Caki/Ramos RA1
1
9.3
162
13.4
25.3
3.67
>618
>924
206
14.3
17.2
8.28
>618
>924
805
0.69
6.40
29.70
1.0
0.65
9.42
13.16
1.0
2
3
109
4f
5f
6
>618
>924
805
1.0
1.0
1.0
3.90
1.0
7
>728
30.6
16.2
15.9
30.5
55.4
657
>728
27.7
24.8
22.8
21.1
49.2
359
>728
12.9
1.80
4.9
1.0
8
1.10
0.65
0.70
1.45
1.13
1.83
2.37
9.0
9
10
11
12f
13f
3.24
9.53
1.41
1.29
3.2
39.3
509
a Concentration required to kill 50% of the cells.
b Caki-1 cells are human renal carcinoma with wt p53 and are used as a negative control.
c C33A are human cervical cells with a mutation at Cys 273.
d Ramos RA1 are human lymphocyte with a mutation at Asp 254.
e Values are means of three experiments.
f Obtained compounds commerciallly.

N. L. Reddy et al. / Bioorg. Med. Chem. Lett. 14 (2004) 5645–5649
5647
Table 2. Structure–activity relationships of 3-methylene-2-norbornanone
LC50 (lM)a
Caki-1b
Selective activity against mutant p53 (fold increase)e
Compound #
C33Ac
Ramos RA1d
Caki/C33A
Caki/Ramos RA1
3
109
271
609
609
3.67
26.7
185
8.28
35.0
609
29.7
10.15
3.29
1.0
13.16
7.74
1.0
14
15
16
17
18
19
20
21
22
23
24
25
609
609
1.0
15.9
515
36.0
2.5
2.1
6.36
1.15
1.54
8.07
7.11
7.31
1.95
8.51
1.93
7.57
1.46
1.36
7.95
7.55
7.55
2.13
8.98
2.72
447
353
23.4
21.2
15.2
16.4
171
26.4
21.5
14.3
15.9
157
171
108
120
334
114
190
13.4
98.2
12.7
69.9
a Concentration required to kill 50% of the cells.
b Caki-1 cells are human renal carcinoma with wt p53 and are used as a negative control.
c C33A are human cervical cells with a mutation at Cys 273.
d Ramos RA1 are human lymphocyte with a mutation at Asp 254.
e Values are means of three experiments.
O
R1
R1
R2
R3
O
n
R2
O
R4
3
O
8 R1 = R2 = H, n = 2
4 R1= R2 = H; R3, R4 = O
5 R1, R2 = CH2, R3, = R4 = H
6 R1, R2 = CH2, R3 , R4 = H, OH
7 R1, R2 = CH2, R3, R4 = NOH
12
9 R1 = H, R2 = OMe, n = 2
10 R1= R2 = OMe, n = 2
11 R1 = H, R2 = OMe, n = 1
O
R
O
O
13
14
18 R = CO2H
19 R = CO2Me
20 R = CONHCH2CH(Me)2
21 R = CONHC6H4-4-OMe
R1
22 R = CONHCH2C6H4-4-OMe
R2
N
S
O
17
O
23 R = CH2SO2NHCH2CH(Me)2
24 R = CH2SO2NHC6H4-4-OMe
25 R = CH2SO2NHCH2C6H4-4-OMe
O
15 R1 = H, R2 = Me
16 R1 = Me, R2 = H
Figure 2.
on camphor sulfonamide. Several compounds analo-
gous to 3 (18–25) were prepared using either ketopinic
acid or camphor sulfonyl chloride as starting materials.
Ketopinic acid was first converted to the acid chloride
(SOCl2) before condensation with the appropriate
amines. The resulting amides/sulfonamides were finally
converted to the a-methylene analogs 18–25 using the
above reaction method.
ing biological activity. This observation was further
supported when the alcohol (6) and oxime (7) were
found to be less active and selective compared to com-
pound 3. The loss of activity and selectivity of these
compounds for mutant and wild-type p53 indicate the
importance of the a-methylene carbonyl functionality
for retaining biological activity.
Next we focused on modifying the carbocyclic part of
compound 3 while keeping the a-methylene carbonyl
moiety intact. First we investigated the extension of
the carbocyclic frame with an aromatic ring and also
eliminating the methylene bridge. The addition of an
aromatic ring would allow us to introduce additional
groups on to it for optimization studies and further it
Initially we focused on the relative importance of the
methylene and ketone functionalities of compound 3
for anti-proliferative activity (Table 1). Simple norbor-
nanone (4) or 2-methylene norbornane (5) were found
to be far less active and selective compared to 3, indicat-
ing that both functional groups are required for retain-

5648
N. L. Reddy et al. / Bioorg. Med. Chem. Lett. 14 (2004) 5645–5649
may improve the physico-chemical properties of the
compounds as 3-methylene-2-norbornanone (3) is a
volatile liquid and difficult to handle in the biological
assays. The first compound of this series, an a-methyl-
ene-1-tetralone (8), showed moderate activity and selec-
tivity against mutant p53 cell types over wild-type p53
(Table 1). The 6-methoxy analog (9) showed improved
activity and selectivity for mutant p53 cell types and
more importantly showed a >10-fold improvement in
activity for mutant p53 (Ramos-RA-1) cell types com-
pared to parent tetralone (8). However, the 6,7-dimeth-
oxy analog (10) showed decreased selectivity in spite of
retaining activity for mutant p53 cell types. In order to
explore the effect of the ring size on activity and selectiv-
ity for mutant p53 cell types, we have tested the 5-meth-
oxy-a-methylene-1-indanone (11). Its activity and
selectivity for mutant p53 cell types was found to be
comparable to the homologue, compound 9. As earlier
reports indicate that a-methylene carbocyclic ketones
and lactones are cytotoxic and potential anti-cancer
agents, we tested compounds 12 and 13 to see whether
they showed any anti-proliferative activity.9 Biological
data for these compounds show reduced activity and
selectivity for mutant p53 cell lines compared to com-
pound 3.
120
100
80
60
40
20
0
Blank
2
3
9
11
17
19
Figure 3.
and mutant p53 protein was measured by Mutant p53
ELISA (Oncogene, Boston, MA) according to the manu-
facturers instructions; this ELISA utilizes an antibody
that is specific for the common conformation of mis-
folded p53 with mutations in the DNA binding region.
The antibody does not recognize wild-type p53. Com-
pound 2 was used as a positive control and vehicle
and compound 19, a partially active derivative, as a
negative control. The results are expressed as percent
loss of mutant p53 content/total protein content (Fig.
3) and show a correlation between loss of mutant p53
protein and anti-proliferation activity.
Next we turned our focus to direct modifications to the
norbornanone carbocyclic frame while keeping the a-
methylene carbonyl moiety intact. Extension of the
carbocyclic frame with the cyclopentane ring (14) did
not significantly affect its anti-proliferative activity for
mutant p53 cell types, while the a-methylene analogs
of camphor (D and L; 15, 16) showed poor activity
and selectivity for mutant p53 cells. However, the excel-
lent activity and selectivity of the cyclic sulfonamide (17)
for mutant p53 cell lines indicate that the carbocyclic
frame will tolerate certain types of substitutions. This re-
sult led us to investigate further the SAR of this partic-
ular chemical series.
As compounds containing a-methylene-c-lactones or 2-
cyclopentenone are known to react with thiol-rich pro-
teins,12 the whole cell glutathione content was measured
by monochlorobimane.9 Briefly, plated cells were incu-
bated for 4h with compounds or vehicle. The cells were
washed with PBSand incubated at 37 ꢁC for 30min in
the dark with 2mM monochlorobimane (Sigma). The
fluorescence resulting from reduced glutathione reaction
with monochlorobimane was measured at ex 390nm, em
520nm. An IC50 value was calculated for compounds
exhibiting a decrease in glutathione content. Compound
3 decreased the glutathione content in mutant p53 C33A
cells at an IC50 concentration approximately 10 times
higher than for anti-proliferation activity. For non-mu-
tant p53 Caki-1 cells, it inhibited glutathione content at
approximately the same concentration as anti-prolifera-
tion activity indicating that the anti-proliferative activity
for mutant p53 cells observed for compound 3 is not due
to interaction with glutathione. Compound 3 was incu-
bated with WI-38 human fibroblast (non-cancer) cells
under conditions similar to the anti-proliferation assay.
The calculated LC50 for inhibition of WI-38 growth was
approximately 17 and 12 times higher than for inhibi-
tion of mutant p53 cells Ramos and C33A, respectively,
suggesting that toxicity is selective for cancer cells with
mutated p53.
We have first tested the a-methylene analogs of simple
acid, the ketopinic acid (18) and its methyl ester (19)
(Table 2). The acid analog showed poor activity and
selectivity, however the methyl ester showed about 8-
fold and 23-fold improved activity for mutant p53
C33A and Ramos-RA-1 cell lines, respectively, com-
pared to the parent camphor analog (15). We next inves-
tigated several amide (20, 21, 22) and sulfonamide (23,
24, 25) analogs of a-methylene camphor. In general
these compounds showed good activity and selectivity
against mutant p53 cell lines over wild-type p53. In
particular, the 4-methoxyaniline analogs of ketopinic
and camphor sulfonic acids (21 and 24) showed excellent
activity and selectivity for the mutant p53 cell types. The
first lead compound of this series, compound 3 was
further investigated in biochemical, metabolic, and toxi-
cological assays.
Refolding of mutant p53 into the active conformation
and a corresponding loss of misfolded protein is the
likely mechanism of compound 3 and derivatives. To
measure the amount of misfolded p53 protein, Ramos
cells were treated for 4h with compounds at 100lM
2. Conclusions
We have identified a novel series of small molecules based
on a structure–activity relationship (SAR) study of 3-

N. L. Reddy et al. / Bioorg. Med. Chem. Lett. 14 (2004) 5645–5649
5649
3. de Vries, A.; Flores, E. R.; Miranda, B.; Hsieh, H.-M.; van
Oostrom, C.; Sage, J.; Jacks, T. Proc. Natl. Acad. Sci.
U.S.A. 2002, 99, 2948.
4. Sigal, A.; Rooter, V. Cancer Res. 2000, 60, 6788.
5. Chene, P. Nat. Rev. 2003, 3, 102.
6. (a) Forster, B. A.; Coffey, H. A.; Morin, M. J.; Rastinejad,
F. Science 1999, 286, 2507; (b) Rippin, T. M.; Bykov, V.
J.; Freund, S. M.; Selivanova, G.; Wilman, K. G.; Fersht,
A. R. Oncogene 2002, 21, 2119.
methylene-2-norbornanone (3). This series showed
potent activity and selectivity for mutant p53 cell types
in anti-proliferation assays and specific interaction to
mutant p53 protein refolding in an ELISA assay. Fur-
ther, several compounds of this series showed desirable
metabolic and tox profiles and are currently being inves-
tigated in additional biological studies including animal
models.
7. (a) Bykov, V. J.; Issaeva, N.; Shilov, A.; Hultcrantz, M.;
Pugacheva, E.; Chumakov, P.; Bergman, J.; Wilman, K.
G.; Selivanova, G. Nat. Med. 2002, 8, 282; (b) Bykov, V.;
Selivanova, G.; Wilman, K. PCT WO 02/24962 A1, 2002.
8. Mukhopadhyay, T.; Roth, J. A. U.S. Patent 6,410,029,
2002.
Acknowledgements
We thank scientists at gvk bioSciences, Hyderabad,
India for synthesizing some of the compounds.
9. (a) Kupchan, S. M.; Eakin, M. A. J. Med. Chem. 1971, 14,
1147; (b) Cassady, J. M.; Byrn, S. R.; Stamos, I. K.;
Evans, S. M.; Mckenzie, A. J. Med. Chem. 1978, 21, 815;
(c) Lee, K.-H.; Mar, E.-C.; Okamoto, M.; Hall, I. H. J.
Med. Chem. 1978, 21, 819.
References and notes
10. All new compounds were characterized by 1H NMR, LC–
MS, and HPLC.
1. (a) Vousden, K. H.; Lu, X. Nat. Rev. 2002, 2, 594; (b)
Bullock, A. N.; Fersht, A. R. Nat. Rev. 2001, 1, 68.
2. Zhu, X.; Yu, Q.-S.; Cutler, R. G.; Culmsee, C. W.;
Holloway, H. W.; Lahiri, D. K.; Mattson, M. P.; Greig,
N. H. J. Med. Chem. 2002, 45, 5090.
11. Komatsubara, K.; Sano, K.; Tabuchi, T.; Iwazawa, J.;
Kishi, H. JP 10265415, 1998.
12. Troyano, A.; Fernandez, C.; Sancho, P.; de Blas, E.; Aller,
P. J. Biol. Chem. 2001, 276, 47107.