J.-H. Kang et al. / Bioorg. Med. Chem. Lett. 20 (2010) 1008–1012
1011
Table 2
Antitumor activities of DAG-lactones (GI50
)
Compd #
PKCa
Leukemia
CCRF-CEM
Breast
Colon
Melanoma
SK-MEL-5
Lung
K562
MOLT 4F
HL-60
HS 578T
Colo #205
NCI-H322
Ki (nM)
GI50
(lg/ml)
ADR
HK-434
1
4
5
6
7
8
9
15
17
18
21
22
25
27
28
30
31
0.375
0.118
<0.1
0.277
1.098
0.227
2.384
0.498
1.639
1.029
2.139
1.842
1.343
1.374
7.910
>10
2.022
>10
1.546
0.605
3.167
6.352
0.181
<0.1
0.140
0.701
4.261
<0.1
0.877
0.304
>10
0.650
6.630
5.715
0.933
>10
0.413
3.469
2.398
5.217
4.653
3.967
2.960
4.525
>10
>10
6.389
>10
>10
>10
0.596
3.304
2.266
7.337
5.449
6.770
3.002
4.637
>10
>10
>10
>10
>10
0.523
0.138
0.253
0.541
0.118
0.178
0.491
0.264
2.008
2.487
2.337
8.908
>10
5.144
3.861
2.099
0.597
1.535
4.735
0.481
2.153
0.779
2.808
1.753
2.134
2.994
1.431
8.326
5.813
5.336
>10
0.615
4.236
>10
>10
>10
>10
9.520
>10
>10
>10
>10
>10
>10
>10
>10
>10
>10
8.428
>10
2.9
7.83
8.2
4.4
5.0
0.516
0.189
0.138
1.149
0.156
3.694
1.090
0.872
7.303
5.106
2.579
0.856
0.959
0.352
1.518
3.384
8.1
3.2
198
14.5
8.24
118
6.38
11.4
6.83
9.3
>10
>10
>10
>10
>10
5.238
>10
0.691
7.55
>10
2.820
1.076
>10
>10
7.783
>10
>10
7.796
5.618
3.456
5.104
>10
10.5
36.4
8.25
>10
group was critical. The hydroxyl in the middle of the side chain
caused a dramatic reduction in binding affinity (compound 16 vs
18). The aldehyde analogues (19–22) showed an SAR pattern sim-
ilar to that of the hydroxyl analogues. In particular, compound 21
exhibited better binding affinity despite of lower log P value com-
pared to compound 1. The carboxylic acid analogue (23) showed
moderately reduced binding affinity compared to the parent com-
pound; however, the ester analogue (25) had a similar affinity. The
acyloxy analogues (26–28) showed activities similar to those of the
other polar compounds. However, a higher log P rather caused a
lower binding affinity. Whereas the azide (29) and amino (30)
analogues showed moderate decreases in binding affinities, the
amide analogue (31) displayed binding affinity even better than
that of the corresponding ester (27).
We conclude that incorporation of polar groups at the terminus
of the linear alkyl chain had little effect on binding affinity in most
cases. However, charged polar groups, such as carboxylate and
amino, caused a moderate reduction in binding affinity. The posi-
tion of the polar groups is also critical. Whereas the polar groups
are tolerated at the terminus of the chain, there was a dramatic
reduction in binding affinity when they were present in the middle
of the chain. These results suggest that whereas the middle part of
side chain needs to be inserted into the hydrophobic interior of the
membrane, the terminus may be able to fold back to the region of
the hydrophilic lipid head groups.
affinities (Ki = 36–198 nM) displayed moderate to low activities.
The antitumor activity of the DAG-lactones was selective for the
colon cancer and several of the leukemia cell lines compared to
adriamycin. In particular, compounds 5, 6 and 8 demonstrated
excellent tumor inhibition against the colon cancer line which
was several times more potent than adriamycin.
Interestingly, although compounds 21 and 25 had similar PKC
binding affinities and lipophilicities compared to compound 5, they
showed weak cell growth inhibition. The role of intracellular local-
ization in determining functional activity is critical, and the ability
of PKCs to phosphorylate their substrates depends not only on
their intrinsic level of catalytic activity but also on their proximity
to their potential substrates. Typically, PKCs translocate to differ-
ent cellular compartments in response to ligand binding to the
C1 domains. Stability under the conditions of the cellular assays
may also be a factor.
In summary, in order to find less lipophilic DAG-lactones with
high affinity binding, a series of polar 3-alkylidene DAG-lactones
containing ether, hydroxyl, aldehyde, acid, ester, acyloxy, azido,
amino or amide functionalities have been synthesized and evalu-
ated as PKC-a ligands and as antitumor agents. Most polar groups
except the charged groups were tolerated and found to be compa-
rable to the corresponding linear alkyl ones in terms of binding
affinity. However, positioning of the polar groups and the overall
log P of the compounds were critical for binding to the enzyme.
The etheric DAG-lactones 5, 6 and 8 showed not only high binding
affinity but also significant antitumor activities, particularly
against the leukemia and colon cancer cell lines.
The comparison between c log P and binding affinity indicated
that the overall log P seems to be critical for binding affinity. The
binding affinity of DAG-lactones was correlated with the log P of
the compounds and a value of 6–7 appeared to be optimal for par-
titioning between enzyme and membrane. Importantly, the E-iso-
mer in this series was more potent than the Z-isomer, a result
different from our previous findings in the alkylidene series. It is
of substantial practical significance because the E-isomer is more
synthetically accessible and can be obtained exclusively during
the alkylation step in the DAG-lactone synthesis.
Acknowledgments
This work was supported by a Korea Research Foundation Grant
(KRF-2008-313-E00763), the ERC program of MOST/KOSEF (R11-
2007-107-02001-0) and by the Intramural Research Program of
the NIH, National Cancer Institute, Center for Cancer Research.
The antitumor activities of the PKC ligands were evaluated in
eight different cell lines and compared with adriamycin and HK-
434 as reference compounds. Compound 1 showed significant anti-
tumor activity in a broad range of cell lines with the exception of
the lung cancer cell line. In general, the DAG-lactones (compounds
5, 6, and 8) with high affinities (Ki = 3.2–5) showed significant anti-
tumor activities as expected whereas ligands (9, 18, 30) with low
References and notes
1. Protein Kinase C. Current Concepts and Future Perspectives; Lester, D. S., Epand, R.
M., Eds.; Ellis Horwood: New York, 1992.
2. Protein Kinase C; Kuo, J. F., Ed.; Oxford University: New York, 1994.
3. Nishizuka, Y. Science 1992, 258, 607.
4. Newton, A. C. Chem. Rev. 2001, 101, 2353.