Analogs of the RSK inhibitor SL0101: Optimization of in vitro biological stability

Article abstract of DOI:10.1016/j.bmcl.2012.03.033

The Ser/Thr protein kinase, RSK, is important in the etiology of tumor progression including invasion and motility. The natural product kaempferol-3-O-(3″,4″-di-O-acetyl-α-l-rhamnopyranoside), called SL0101, is a highly specific RSK inhibitor. Acylation o

Full text of DOI:10.1016/j.bmcl.2012.03.033

Bioorganic & Medicinal Chemistry Letters 22 (2012) 3244–3247
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Analogs of the RSK inhibitor SL0101: Optimization of in vitro biological stability
⇑
Michael K. Hilinski , Roman M. Mrozowski, David E. Clark, Deborah A. Lannigan
Center for Cell Signaling, University of Virginia, Charlottesville, VA 22908, USA
Department of Microbiology, Immunology, and Cancer Biology, University of Virginia, Charlottesville, VA 22908, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The Ser/Thr protein kinase, RSK, is important in the etiology of tumor progression including invasion and
motility. The natural product kaempferol-3-O-(300,400-di-O-acetyl-
a-L-rhamnopyranoside), called SL0101,
Received 18 January 2012
Revised 5 March 2012
Accepted 7 March 2012
Available online 13 March 2012
is a highly specific RSK inhibitor. Acylation of the rhamnose moiety is necessary for high affinity binding
and selectivity. However, the acetyl groups can be cleaved by esterases, which accounts for the poor
in vitro biological stability of SL0101. To address this problem a series of analogs containing acetyl group
replacements were synthesized and their in vitro stability evaluated. Monosubstituted carbamate ana-
logs of SL0101 showed improved in vitro biological stability while maintaining specificity for RSK. These
results should facilitate the development of RSK inhibitors derived from SL0101 as anticancer agents.
Ó 2012 Elsevier Ltd. All rights reserved.
Keywords:
SL0101
RSK inhibitor
RSK-specific
Breast cancer
Protein kinase
The members of the p90 ribosomal S6 kinase (RSK) family of
Ser/Thr protein kinases have been shown to play a role in a number
of different cancers as key drivers of proliferation and metastasis.1–
growth of cancer cell lines, likely due to poor membrane perme-
ability.13 These results indicate that SL0101 is not a suitable candi-
date for in vivo evaluation, as hydrolysis of the acetates by
esterases would generate a less potent inhibitor.
8
These discoveries have been enabled in part by our report of the
identification and isolation of the RSK inhibitor SL0101 (1, Fig. 1).9
SL0101 is a flavonoid glycoside (kaempferol 3-O-(300,400-di-O-acet-
An analog that replaces these acetates with ethyl ethers (3)
inhibits RSK with potency roughly equivalent to SL0101.13 We pre-
viously determined that the specificity of SL0101 and its analogs
for RSK could be evaluated by their preferential ability to inhibit
the growth of the human breast cancer line, MCF7, compared to
the normal human breast line, MCF-10A.12 Unexpectedly, we ob-
served that the ethyl ether analog 3 inhibited both lines to a similar
extent, which indicates that it has a decreased specificity for RSK.13
These results demonstrate that the acetates are a key modulator of
specificity and thus a more carefully considered approach is neces-
sary to identify suitable replacements. Accordingly, we have fo-
cused our efforts on identifying analogs bearing replacements for
the acetates that confer greater biological stability without
decreasing potency or specificity for RSK. Herein we present our
approach, which has led to the identification of SL0101 analogs
that are both specific for RSK and more biologically stable
in vitro than the parent compound.
The only structural difference between the diethyl analog 3 and
the diacetyl parent compound is the replacement of two methyl-
enes with two carbonyl groups. It is surprising that such a seem-
ingly small structural feature can regulate specificity for RSK. To
recover this specificity, in the design of new analogs we sought
to better mimic the acetates and particularly the acetate carbonyls,
sterically and electronically, in a way that would confer a greater
resistance to metabolism by esterases. In one approach we investi-
gated the dependence of potency and specificity on the relative
yl-a-L-rhamnopyranoside)) isolated from Forsteronia refracta, a
variety of dogbane found in the South American rainforest.
SL0101 is highly specific for RSK, inhibiting RSK1/2 but not unre-
lated kinases nor the closely related kinases MSK1 and
p70S6K1.2,9,10 SL0101 inhibits the proliferation of breast and pros-
tate cancer lines but not their normal counterparts even though it
inhibits RSK activity in all the lines.1,5,9 Thus it appears that some
cancer cells have become addicted to RSK, which suggests that
RSK may be a potential new target for cancer therapeutics.
SL0101, owing to its exquisite specificity, is a compelling lead com-
pound from which to begin the process of identifying drug-like RSK
inhibitors.
We and others have reported the total synthesis and biological
evaluation of SL0101 and a number of analogs, with the ultimate
goal of developing an anticancer drug that targets RSK.11–15 These
analogs have provided key information about the SAR of both
the aglycone and carbohydrate portions of the natural product. In
the course of this work we discovered that the 300 and 400 acetyl
groups of the carbohydrate are critical for potency and specificity
for RSK.13 TriOH-SL0101 (2), lacking these acetyl groups, is 12-fold
less potent for inhibition of RSK in vitro and does not inhibit the
⇑
Corresponding author. Tel.: +1 434 924 1152; fax: +1 434 924 1236.
E-mail address: mh6cu@virginia.edu (M.K. Hilinski).
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.03.033

M. K. Hilinski et al. / Bioorg. Med. Chem. Lett. 22 (2012) 3244–3247
3245
Figure 1. The RSK inhibitor SL0101 and two previously reported analogs.13
position of the carbonyl group. To this end we prepared an analog 9
in which the acetates are replaced by alkoxyacetones (Scheme 1),
moving the carbonyl group one carbon further from the carbohy-
drate ring and replacing the labile ester bond with an ether bond.
The desired functionality could be installed at a late stage in the
synthesis of the analog. Alkylation of known diol 613 with propar-
gyl bromide provided bis-alkyne 7, which after mercury-catalyzed
hydration provided bis-ketone 8. Removal of the benzyl ether
protecting groups by hydrogenolysis using Pearlman’s catalyst
provided the completed analog 9.
In a second approach we retained the acetate carbonyl in the
correct position but in a more biologically stable form in a series
of analogs in which we replaced the acetates with bioisosteric
mono- or disubstituted carbamates. Late-stage installation of the
carbamate was desirable for maximum synthetic efficiency. Thus,
carbamoylation of diol 6 with the appropriate isocyanate or diaklyc-
arbamoyl chloride followed by hydrogenolysis of the benzyl ethers
provided mono- or dialkylated carbamates 10–15 (Scheme 2).
The ability of all new analogs to inhibit RSK activity was deter-
mined in an in vitro kinase assay and compared with the parent
Scheme 2. General scheme for the preparation of carbamate analogs of SL0101.
Reagents and conditions: (a) R1NCO, Et3N, DMF, 45 °C, 44–66%; (b) R2R3NCOCl, NaH,
DMF, 0 °C to rt, 26–69%; (c) H2, Pd(OH)2/C, MeOH, EtOAc, rt, 46–94%. Yields are
unoptimized.
Table 1
Potency of analogs in in vitro kinase and MCF7 cell-based assays
Compound
RSK2 IC50
(l
M)
MCF7 IC50 (lM)
1
9
0.583 (0.489–0.696)
0.252 (0.189–0.336)*
1.13 (0.876–1.46)*
0.869 (0.649–1.16)
1.92 (1.29–2.86)*
0.493 (0.355–0.684)
0.356 (0.255–0.496)
1.43 (1.09–2.04)*
45.6 (42.7–48.8)
34.1 (30.1–38.5)*
77.0 (71.6–82.7)*
46.4 (43.2–50.0)
53.3 (50.6–56.2)*
PS
10
11
12
13
14
15
PS
>100
IC50 is concentration needed for 50% inhibition; the 95% CI is shown in parentheses;
n = 3 in triplicate.
*
p <0.05. PS; partially soluble.
compound 1 (Table 1). The ketone analog 9 was twofold more
potent than 1 at inhibiting RSK2. Analogs 11, 13, and 14 were as
potent as SL0101, and analogs 10, 12, and 15 were slightly (two
to threefold) less potent. Overall, we found that the ability of an
analog to inhibit RSK was not greatly influenced by the structure
of the acetate replacement, which is consistent with previous
observations.
We also determined the ability of all new analogs to inhibit
MCF7 cell proliferation (Table 1). The ketone analog 9 was again
the most potent of the new analogs. The three monosubstituted
carbamates, analogs 10–12, were similarly potent to the parent
compound, with a trend toward improved potency with increasing
lipophilicity of the carbamate substituent, presumably due to im-
proved membrane permeability. In the disubstituted carbamate
series, the dimethyl analog 13 and 1-pyrrolidinyl carbamate analog
14 exhibited poor solubility in cell culture media and therefore
their ability to inhibit cell growth was not determined. The mor-
pholino bis-carbamate 15 showed improved solubility but was un-
able to inhibit cell proliferation despite its ability to inhibit RSK in
the in vitro kinase assay, most likely due to poor membrane
permeability.
Scheme 1. Synthesis of a bis-ketone analog of SL0101. Reagents and conditions: (a)
NaH, propargyl bromide, THF, 0 °C to rt, 66%; (b) Hg(OAc)2, PPTS, water, acetone, rt,
62%; (c) H2, Pd(OH)2/C, MeOH, EtOAc, 50%. Yields are unoptimized.

3246
M. K. Hilinski et al. / Bioorg. Med. Chem. Lett. 22 (2012) 3244–3247
Analogs that inhibited MCF7 cell proliferation were evaluated
along with 1 for their stability in a MCF7 cell-based assay. The
inhibitor was added when the cells were plated and proliferation
analyzed at various time points to determine the persistence of
the inhibitory effect. SL0101 (1) was able to inhibit MCF7 prolifer-
ation for 48 h (Fig. 2). However, at longer time points the cells be-
gan to proliferate indicating that SL0101 was no longer effective,
which we hypothesize is due to degradation of the inhibitor by
esterases to the inactive triol 2. Treatment of cells with either
the bis-ketone analog 9 or the ethyl carbamate analog 10 did not
result in sustained growth inhibition, indicating poor in vitro sta-
bility of these analogs. As the 300 and 400 substituents of analog 9
are non-hydrolyzable, its poor stability was initially surprising.
However, MCF7 cells express aldo-keto reductases (AKRs), well
known to be Phase I metabolizing enzymes for a variety of drugs
bearing carbonyl groups.16,17 Thus an alternative metabolic path-
way is available to analog 9 whereby one or both ketones could
be reduced by AKRs to secondary alcohols, leading either directly
to a less potent RSK inhibitor or indirectly as the secondary alco-
hols could be further metabolized by conjugation.17
Figure 3. Persistence of RSK inhibition. MCF7 cells were treated with SL0101 or the
more stable analogs 11 and 12 (100 M). At the indicated time in hours (h) the cells
were lysed and the lysates immunoblotted. Each analog was analyzed on a single
membrane with SL0101. White space indicates sections of the membrane that were
cropped to remove unnecessary lanes.
l
Encouragingly, the more lipophilic monosubstitued carbamate
analogs 11 and 12 demonstrated improved in vitro stability, as
cells treated with these compounds did not proliferate over the full
time course (Fig. 2). We further examined the stability of analogs
11 and 12 by determining whether cyclin D1 levels were inhibited
(Fig. 3). Previously, we found that SL0101 inhibits proliferation in
breast cancer cell lines by inducing a cell cycle block in G1, which
is due to RSK regulation of cyclin D1 levels.1,18 In agreement with
the MCF7 stability results we observed that SL0101 decreased
the levels of cyclin D1 at 48 h compared to the control, but that cy-
clin D1 levels began to increase at later time points, indicating deg-
radation of the inhibitor. However, cyclin D1 levels remained low
in cells treated with 11 or 12, indicating persistent inhibition of
RSK and therefore improved biological stability of the carbamate
analogs over the parent compound. Taken together, these results
indicate that analogs 11 and 12 have improved stability over
SL0101 (1).
We then investigated whether our strategy of reintroducing the
carbonyl group improved the specificity of 11 and 12 relative to the
diethyl analog 3.13 We have previously shown that the specificity
of an analog for RSK can be evaluated by determining its antipro-
liferative activity in both MCF-10A and MCF7 cells, with the most
specific analogs showing no inhibition of MCF-10A but substantial
inhibition of MCF7 proliferation, due to the differential dependence
of the growth of these cell lines on RSK.9,12 We have also previously
shown that while SL0101 does not inhibit the growth of MCF-10A
Figure 4. Inhibition of growth of MCF-10A versus MCF7 cells by SL0101 and select
analogs. The inhibitor concentration was 50
when compared to vehicle).
l
M. (n = 3 in quadruplicate, ⁄p 6 0.05
specificity for RSK.13 We found that analogs 11 and 12, like
SL0101, significantly inhibited the growth of MCF7 cells but did
not significantly inhibit the growth MCF-10A cells (Fig. 4). These
results suggest that analogs 11 and 12, like SL0101, specifically
inhibit RSK.2,9,10 The only significant differences in biological activ-
ity between the two compounds are slightly improved potencies
for 11 versus 12 in both the kinase and MCF7 cell proliferation as-
says. As these small differences are unlikely to be physiologically
cells up to a concentration of 100
lM, the diethyl analog 3 signifi-
cantly inhibits the growth of MCF-10A cells, indicating reduced
Figure 2. In vitro determination of analog stability. The inhibitor (100
lM) was added to MCF7 cells at time 0 and percentage of growth determined for the indicated time
points. (n = 3 in quadruplicate, #p 6 0.05 at 48 h when compared to vehicle at 48 h, ⁄p 6 0.05 when compared to 48 h treatment with the same analog).

M. K. Hilinski et al. / Bioorg. Med. Chem. Lett. 22 (2012) 3244–3247
3247
important, either carbamate modification should render an analog
References and notes
suitable for in vivo evaluation.
In summary, the C300 and C400 acetates on the carbohydrate moi-
ety of SL0101 are required for both potent and specific inhibition of
RSK but we predict that they would be metabolized rapidly by
esterases in vivo, a fact which is supported by the poor biological
stability of the natural product in vitro. Thus, SL0101 is not suitable
for in vivo evaluation and analogs with improved stability are
needed. The number of suitable replacements for these acetates
that would confer greater biological stability is surprisingly limited
as a simple change from acetyl to ethyl leads to a reduction in spec-
ificity for RSK. As a solution to this problem, bioisosteric replace-
ment of the acetates by carbamates provided analogs that are
more biologically stable than SL0101 in vitro and are as specific
as SL0101 for RSK. These modifications along with others aimed
at further improving the stability and potency of SL0101 analogs
are currently being investigated in our laboratory with the goal
of identifying a RSK inhibitor that could be advanced to preclinical
testing.
1. Eisinger-Mathason, T. S.; Andrade, J.; Lannigan, D. A. Steroids 2010, 75, 191.
2. Doehn, U.; Hauge, C.; Frank, S. R.; Jensen, C. J.; Duda, K.; Nielsen, J. V.; Cohen, M.
S.; Johansen, J. V.; Winther, B. R.; Lund, L. R.; Winther, O.; Taunton, J.; Hansem,
S. H.; Frodin, M. Mol. Cell 2009, 35, 511.
3. Smolen, G. A.; Zhang, J.; Zubrowski, M. J.; Edelman, E. J.; Luo, B.; Yu, M.; Ng, L.
W.; Scherber, C. M.; Schott, B. J.; Ramaswamy, S.; Irimia, D.; Root, D. E.; Haber,
D. A. Genes Dev. 2010, 24, 2654.
4. Cho, Y. Y.; Yao, K.; Kim, H. G.; Kang, B. S.; Zheng, D.; Bode, A. M.; Dong, Z. Cancer
Res. 2007, 67, 8104.
5. Clark, D. E.; Errington, T. M.; Smith, J. A.; Frierson, H. F., Jr.; Weber, M. J.;
Lannigan, D. A. Cancer Res. 2005, 65, 3108.
6. Kang, S.; Dong, S.; Gu, T. L.; Guo, A.; Cohen, M. S.; Lonial, S.; Khoury, H. J.;
Fabbro, D.; Gilliland, D. G.; Bergsagel, P. L.; Taunton, J.; Polakiewicz, R. D.; Chen,
J. Cancer Cell 2007, 12, 201.
7. Kang, S.; Elf, S.; Lythgoe, K.; Hitosugi, T.; Taunton, J.; Zhou, W.; Xiong, L.; Wang,
D.; Muller, S.; Fan, S.; Sun, S. Y.; Marcus, A. I.; Gu, T. L.; Polakiewicz, R. D.; Chen,
Z. G.; Khuri, F. R.; Shin, D. M.; Chen, J. Clin. Invest. 2010, 120, 1165.
8. Kang, S.; Elf, S.; Dong, S.; Hitosugi, T.; Lythgoe, K.; Guo, A.; Ruan, H.; Lonial, S.;
Khoury, H. J.; Williams, I. R.; Lee, D. H.; Roesel, J. L.; Karsenty, G.; Hanauer, A.;
Taunton, J.; Boggon, T. J.; Gu, T. L.; Chen, J. Mol. Cell Biol. 2009, 29, 2105.
9. Smith, J. A.; Poteet-Smith, C. E.; Xu, Y.; Errington, T. M.; Hecht, S. M.; Lannigan,
D. A. Cancer Res. 2005, 65, 1027.
10. Bain, J.; Plater, L.; Elliott, M.; Shpiro, N.; Hastie, C. J.; McLauchlan, H.; Klevernic,
I.; Arthur, J. S.; Alessi, D. R.; Cohen, P. Biochem. J. 2007, 408, 297.
11. Maloney, D. J.; Hecht, S. M. Org. Lett. 2005, 7, 1097.
12. Smith, J. A.; Maloney, D. J.; Clark, D. E.; Xu, Y.; Hecht, S. M.; Lannigan, D. A.
Bioorg. Med. Chem. 2006, 14, 6034.
Acknowledgments
This work was supported by the Department of Defense
#W81XWH-11-1-0068 to M.K.H. and GM084386 to D.A.L.
13. Smith, J. A.; Maloney, D. J.; Hecht, S. M.; Lannigan, D. A. Bioorg. Med. Chem.
2007, 15, 5018.
14. Shan, M.; O’Doherty, G. A. Org. Lett. 2006, 8, 5149.
15. Shan, M.; O’Doherty, G. A. Org. Lett. 2010, 12, 2986.
16. Ruiz, F. X.; Porté, S.; Gallego, O.; Moro, A.; Ardèvol, A.; Del Río-Espínola, A.;
Rovira, C.; Farrés, J.; Parés, X. Biochem. J. 2011, 440, 335.
Supplementary data
17. Jin, Y.; Penning, T. M. Annu. Rev. Pharmacol. Toxicol. 2007, 47, 263.
18. Eisinger-Mathason, T. S.; Andrade, J.; Groehler, A. L.; Clark, D. E.; Muratore-
Schroeder, T. L.; Pasic, L.; Smith, J. A.; Shabanowitz, J.; Hunt, D. F.; Macara, I. G.;
Lannigan, D. A. Mol. Cell. 2008, 31, 722.
Supplementary data (experimental procedures and compound
characterization for all new compounds) associated with this arti-
cle can be found, in the online version, at http://dx.doi.org/
10.1016/j.bmcl.2012.03.033.