Synthesis and evaluation of biotinylated sansalvamide A analogs and their modulation of Hsp90

Article abstract of DOI:10.1016/j.bmcl.2011.06.083

Described are the syntheses of three sansalvamide A derivatives that contain biotinylated tags at individual positions around the macrocycle. The tagged derivatives indicated in protein pull-down assays that they bind to Hsp90 at the same binding site (N-Middle domain) as the San A-amide peptide. Further, these compounds inhibit binding between Hsp90 and multiple C-terminal client proteins. This interaction is unique to the San A analogs indicating they can be tuned for selectivity against Hsp90 client/co-chaperone proteins.

Full text of DOI:10.1016/j.bmcl.2011.06.083

Bioorganic & Medicinal Chemistry Letters 21 (2011) 4716–4719
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and evaluation of biotinylated sansalvamide A analogs and their
modulation of Hsp90
Joseph B. Kunicki b, Mark N. Petersen b, Leslie D. Alexander b, Veronica C. Ardi b, Jeanette R. McConnell b,
Shelli R. McAlpine a,
⇑
a Department of Chemistry, University of New South Wales, Sydney, NSW 2052, Australia
b Department of Chemistry and Biochemistry, 5500 Campanile Drive, San Diego State University, San Diego, CA 92182-1030, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Described are the syntheses of three sansalvamide A derivatives that contain biotinylated tags at individ-
ual positions around the macrocycle. The tagged derivatives indicated in protein pull-down assays that
they bind to Hsp90 at the same binding site (N-Middle domain) as the San A-amide peptide. Further,
these compounds inhibit binding between Hsp90 and multiple C-terminal client proteins. This interac-
tion is unique to the San A analogs indicating they can be tuned for selectivity against Hsp90 client/
co-chaperone proteins.
Received 20 May 2011
Revised 17 June 2011
Accepted 17 June 2011
Available online 25 June 2011
Keywords:
Sansalvamide A
Hsp90
Ó 2011 Elsevier Ltd. All rights reserved.
N-Middle domain
Recent work describing the synthesis and biological activity of
pentapeptide derivatives, based on the natural product sansalva-
mide A (San A), has brought attention to this compound class.1–5
San A is a penta-depsipeptide that was discovered by Fenical et al.
from a marine fungus of the genus Fusarium and exhibits anti-tumor
activity.1 The pentapeptide structure (San A-amide, Fig. 1, com-
pound 1) has recently been reported to bind to and inhibit Heat
shock protein 90 (Hsp90).4,5 Hsp90 is a well-established oncogenic
target that modulates client proteins involved in cellular growth,
angiogenesis, and apoptosis.6 The redundancy of pathways involved
in cancergrowthmean thattargeting multiplemechanisms simulta-
neously is key to developing a successful therapy. Recent evidence
shows that Hsp90 controls ꢀ100 client proteins and co-chaperones,
many of which are involved in multiple cancer-related cell signaling
proteins, this makes it an excellent oncogenic target.7 Further,
Hsp90 is up-regulated in most cancers, and cancer cells are more
susceptible to Hsp90 inhibitors than normal cells because Hsp90
plays a vital role in maintaining the functionality of these pathways
during cancer cell growth.6 There are currently 15 Hsp90 inhibitors
in development, with two of these in phase III clinical trials.8 Hsp90
interacts with client proteins at one or more of its three domains: N,
Middle, or C (N, M, and C, respectively). All compounds currently in
clinical development bind to the ATP binding pocket in the N-
domain, and most are structurally related to a single compound, Gel-
danamycin, including 17-AAG, which is currently in phases II and III
clinical trials. Of the three nonGeldanamycin analogs in clinical
trials, none modulate C-terminal client proteins.8 We have reported
that San A-amide (1, Fig. 1) is a cytotoxic molecule that modulates
the activity of multiple client proteins and co-chaperones, acting
via an allosteric effect.5 Indeed, we have published data showing
that San A-amide (1) allosterically modulates C-terminal client pro-
teins FKBP52 and IP6K2, unlike other Hsp90 inhibitors. We have also
recently published the synthesis of analog 2 (Fig. 1), where 2 exhibits
greater cytotoxicity than compound 1 against HCT-116 colon cancer
cells.4
In order to evaluate compound 2’s mechanism of action, we re-
port here the synthesis of biotinylated analogs of compound 2. In
addition we describe Hsp90 pull-down assay results using these bio-
tinylated analogs and binding assay data using compound 2. We
show that compound 2 has an enhanced ability to inhibit binding
⇑
Corresponding author. Tel.: +61 4 1672 8896; fax: +61 2 9385 1111.
E-mail address: s.mcalpine@unsw.edu.au (S.R. McAlpine).
Figure 1. Sansalvamide A molecules 1 and 2.4
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.06.083

J. B. Kunicki et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4716–4719
4717
O
between Hsp90 and four of its client/co-chaperone proteins (Her-2,
NRFmoc
2 (3 eq)
1. HOBT (3 eq)
DIC (6 eq),
DMF 0.2M
2
HOP, FKBP52, and IP6K2) over compounds 1 and 17-AAG. These
proof of principle experiments show that structural variations of
San A allow us to selectively ‘tune’ client/co-chaperone contacts
with Hsp90, thereby controlling unique subsets of Hsp90-protein
interactions and subsequent pathways with each structural variant.
Our previous SAR data indicated that placing a tag at positions
I–IV are all reasonable possibilities and it was not clear if any single
position would be an ideal choice (although position V had been
ruled out).4,9 Data from compound 1-tagged at position IV (1-T-
IV) pulled down Hsp90 N-Middle domain, and we were interested
in evaluating how the placement of the tag might alter compound
2’s ability to bind to its protein target. Thus, we designed and syn-
thesized compounds 2-T-I, 2-T-III, and 2-T-IV (Fig. 2).10
These compounds were synthesized using the same solid-phase,
protocol previously published (Scheme 1).4,11 2-T-I and 2-T-III were
synthesized using a pre-loaded 2-chlorotrityl-leucine resin while 2-
T-IV utilized pre-loaded 2-chlorotrityl-phenylalanine resin. Subse-
quent coupling of Fmoc-protected amino acids and deprotection
was performed until the desired linear pentapeptide was reached.
A Boc-protected lysine was incorporated at the tagged position.
After cleaving the peptide from the resin with 50% trifluoroethanol
in dichloromethane, the linear pentapeptide was cyclized using our
standard macrocyclization conditions.9 The macrocycle was then
subjected to 20% TFA to remove the Boc from the lysine, and biotin-
ylated using NHS-peg-biotin and 8 equiv of DIPEA.
HRN
RN
HO
R2
R
NRH
R1
O
R1
O
O
2. 20% piperidine/DMF
1
1-2
R = H or Me
O
O
R3
NRFmoc
(3 eq)
NRFmoc
3
4
HO
HO
R3
1. HOBT (3 eq)
DIC (6 eq),
DMF 0.2M
R4
(3 eq)
O
O
HRN
O
RN
1. HOBT (3 eq)
DIC (6 eq)
DMF 0.2M
R2
RN
R1
2. 20% piperidine/DMF
2. 20% piperidine/DMF
1-2-3
R3
O
R3
O
O
R4
O
R4
O
NRFmoc
5
N
R
N
R
HO
NR
NHR
RN
RN
R2
O
R5 (3 eq)
NHR
RN
RN
R2
O
O
1. HOBT (3 eq)
DIC (6 eq)
DMF 0.2 M
R5
R1
R1
2. 20% piperidine/DMF
O
O
Linear Pentapeptide*
1-2-3-4
1-2-3-4-5
O
R3
1. 1:1 (v/v) TFE:CH2Cl2, 24 hr
(yields: 80-95%)
R4
R5
O
N
R
Compound 2 tagged at all three positions was then run in pro-
tein pull-down assays using purified N, Middle, C, N-Middle, and
Middle-C domains of mammalian Hsp90 (Fig. 3). Although all three
tagged compound 2 molecules pulled down the anticipated N-Mid-
dle domain, it is noted that compounds 2-T-I and 2-T-III are most
effective at pulling down this domain. These data indicate that
the tag placement is important and affects the ability of the mole-
cule to bind to its target. Further, it suggests that the preferred
binding mode of compound 2 to Hsp90 does not involve residues
I or III and most likely involves residues IV and V.
NR
RN
R2
O
2. TBTU (0.7 eq), HATU (0.7 eq)
DEPBT (0.7 eq), DIPEA (6 eq)
10:1 CH2Cl2:CH3CN
NR RN
R1
O
(0.007-0.0007M) (yields: 5-76%)
O
San A derivative
1. 20% TFA,
CH2Cl2 0.1M,
2eq anisole
Biotinylated
2. 1.2 eq NHS-peg-biotin
8eq DIPEA, CH2Cl2 0.1M
(yields: 5-40%)
Sansalvamide A derivative
*A Boc-protected lysine was incorporated at the tagging position.
Given our previous data on compound 1, we anticipated that the
cytotoxic effect of San A might be a direct result of its ability to in-
hibit the interaction between Hsp90 and client proteins that are
integral to cell signaling events. Thus, we probed the effect of 1
and 2 on the binding interaction between Hsp90 and four client/
co-chaperone proteins: Her2, HOP, FKBP52, and Inositol hexakis-
phosphate kinase-2 (IP6K2) (Fig. 4).5,12 Her2 is a well-established
client protein that binds to the Middle domain, and is inhibited
Scheme 1. General solid-phase synthesis of tagged derivative references.
from binding to Hsp90 by 17-AAG.13 Binding between Her2 and
Hsp90 was inhibited by both 2 and 17-AAG (Fig. 4). HOP (Hsp orga-
nizing protein) binds to Hsp90’s C-domain and facilitates the trans-
fer of unfolded protein from Hsp70 to Hsp90. (Fig. 5). Both 1 and 2
Figure 2. Tagged compound 2 with tags at positions I–IV.

4718
J. B. Kunicki et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4716–4719
Figure 3. Pull-down data for compounds 2-T-I, 2-T-III, and 2-T-IV using mammalian Hsp90 domains: N, Middle, C, N-Middle, and Middle-C domains, respectively.
block this interaction, however, 17-AAG does not. Since HOP is vital
to the Hsp70–Hsp90 interaction, compounds 1 and 2 will be
excellent tools for investigating the mechanism of this complex.
FKBP52 is a co-chaperone that binds to the C-domain of Hsp90
and assists with protein folding and trafficking.14 Both compounds
1 and 2 inhibit the binding between Hsp90 and FKBP52. As ex-
pected, 17-AAG does not inhibit this binding event. Finally, IP6K2,
which binds to the C-terminus of Hsp90, and is known to be in-
volved in an apoptotic pathway, is also modulated by 1 and 2, while
17-AAG does not affect this binding event. Thus, these data show
that 17-AAG gave traditional N-terminal inhibition effects, blocking
Her2 interactions with Hsp90, but not inhibiting Hsp90’s interac-
tion with three C-terminal binders: HOP, FKBP52, and IP6K2. Com-
pounds 1 and 2 inhibited binding between Hsp90 and all three
C-terminal binding proteins. Further, compound 2 inhibited bind-
ing between Hsp90 and Her-2, suggesting that the compounds
can be ‘tuned’ via modification to San A side chains to allosterically
modulate interactions between Hsp90 and its proteins. These com-
pounds can be used as tools to investigate the cell signaling path-
ways by effectively inhibiting C-terminal binders to Hsp90.
via HOP and transfer unfolded proteins. Thus, it appears that San
A derivatives may control the ability of protein transfer between
these two oncogenic targets: Hsp70 and Hsp90.
In summary, we have described the synthesis of three new
biotinylated Sansalvamide A analogs. These tags were placed at
multiple positions around the macrocycle in order to determine
which position would be optimal for binding to the protein target:
Hsp90. Although all three tagged compound 2 molecules pulled
down the expected N-Middle domain, it was discovered that a
tag at positions I or III were optimal for compound 2 to bind to
Hsp90. Further, we ran client-binding assays with compounds 1,
2, and 17-AAG and found that both San A amide compounds mod-
ulate, via an allosteric interaction, binding between all three C-ter-
minal binding client proteins and Hsp90. Further, the ability to
block HOP from binding to Hsp90 would likely affect the key inter-
action between Hsp70 and Hsp90. Data on additional mechanistic
aspects of these compounds is ongoing and will be published in the
near future.
Acknowledgments
Our proposed model for how analogs 1 and 2 play a role in the
binding of C-terminal client proteins to Hsp90 is outlined in Figure
5. Based on the pull-down results, our molecules bind between the
N-Middle domain and inhibit the binding of C-terminal client pro-
tein IP6K2 and co-chaperones FKBP52 and HOP. San A analogs do
this via induction of conformational change when binding to
Hsp90, that is, translated from the N-Middle domain to the C-ter-
minal domain of Hsp90. This induced conformational change
makes the C-terminus of Hsp90 inaccessible FKBP52, HOP, and
IP6K2. Interestingly, inhibiting the binding between HOP and
Hsp90 is likely to inhibit the ability of Hsp70 to dock to Hsp90
We thank the University of New South Wales Sydney for sup-
port of SRM, the Frasch Foundation (658-HF07) (support of J.B.K.,
J.R.M., and L.D.A.), NIH 1R01CA137873 (support of S.R.M. and
V.C.A.), NIH MIRT (support of J.B.K., L.D.A., and J.R.M.).
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2011.06.083.

J. B. Kunicki et al. / Bioorg. Med. Chem. Lett. 21 (2011) 4716–4719
4719
References and notes
San A-amide: 1
120
110
100
90
80
70
60
50
40
30
20
10
0
1. Belofsky, G. N.; Jensen, P. R.; Fenical, W. Tetrahedron Lett. 1999, 40, 2913.
2. Cueto, M.; Jensen, P. R.; Fenical, W. Phytochemistry 2000, 55, 223.
3. Liu, S.; Gu, W.; Lo, D.; Ding, X.-Z.; Ujiki, M.; Adrian, T. E.; Soff, G. A.; Silverman,
R. B. J. Med. Chem. 2005, 48, 3630.
4. Sellers, R. P.; Alexander, L. D.; Johnson, V. A.; Lin, C.-C.; Savage, J.; Corral, R.;
Moss, J.; Slugocki, T. S.; Singh, E. K.; Davis, M. R.; Ravula, S.; Spicer, J. E.; Oelrich,
J. L.; Thornquist, A.; Pan, C.-M.; McAlpine, S. R. Bioorg. Med. Chem. 2010, 18,
6822.
5. Vasko, R. C.; Rodriguez, R. A.; Cunningham, C. N.; Ardi, V. C.; Agard, D. A.;
McAlpine, S. R. ACS Med. Chem. Lett. 2010, 1, 4.
6. (a) Neckers, L. Trends Mol. Med. 2002, 8, S55; (b) Chiosis, G.; JHuezo, H.; Rosen,
N.; Mimgaugh, E.; Whitesell, L.; Neckers, L. Mol. Cancer Ther. 2003, 2, 123; (c)
Hollingshead, M. G.; Alley, M.; Burger, A. M.; Borgel, S.; Pacula-Cox, C.; Fiebig,
H.-H.; Sausville, E. A. Cancer Chemother. Pharmacol. 2005, 56, 115; (d) Senju, M.;
Sueoka, N.; Sato, A.; Iwanaga, K.; Sakao, Y.; Tomimitsu, S.; Tominaga, M.; Irie,
K.; Hayashi, S.; Sueoka, E. J. Cancer Res. Clin. Oncol. 2006, 132, 150; (e) Chang, Y.-
S.; Lee, L.-C.; Sun, F.-C.; Chao, C.-C.; Fu, H.-W.; Lai, Y.-K. J. Cell. Biochem. 2006, 97,
156; (f) Matei, D.; Satpathy, M.; Cao, L.; Lai, Y.-K.; Nakshatri, H.; Donner, D. B. J.
Biol. Chem. 2007, 282, 445.
7. (a) Usmani, S. Z.; Bona, R.; Li, Z. H. Curr. Mol. Med. 2009, 9, 654; (b) Koga, F.;
Kihara, K.; Neckers, L. Anticancer Res. 2009, 29, 797; (c) Barginear, M. F.; Van
Poznak, C.; Rosen, N.; Miodi, S.; Hudis, C. A.; Budman, D. R. Curr. Cancer Drug
Targets 2008, 8, 522.
8. (a) Banerji, U. Proc. Am. Assoc. Cancer Res. 2003, 44, 677; (b) Sausville, E. A. Curr.
Cancer Drug Targets 2003, 3, 377; (c) Pearl, L. H.; Prodromou, C. Curr. Opin.
Struct. Bio. 2000, 10, 46; (d) Dehner, A.; Furrer, J.; Richter, K.; Schuster, I.;
Buchner, J.; Kessler, H. ChemBioChem 2003, 4, 870; (e) Hagn, F. X.; Richter, K.;
Buchner, J.; Kessler, H. Proc. Exp. Nucl. Mag. Res. Conf. 2005, 211; (f) Jez, J. M.;
Chen, J. C.; Rastelli, G.; Stroud, R. M.; Santi, D. V. Chem. Biol. 2003, 10, 361.
9. (a) Otrubova, K.; Lushington, G. H.; Vander Velde, D.; McGuire, K. L.; McAlpine,
S. R. J. Med. Chem. 2008, 51, 530; (b) Otrubova, K.; McGuire, K. L.; McAlpine, S. R.
J. Med. Chem. 2007, 50, 1999; (c) Rodriguez, R. A.; Pan, P.-S.; Pan, C.-M.; Ravula,
S.; Lapera, S. A.; Singh, E. K.; Styers, T. J.; Brown, J. D.; Cajica, J.; Parry, E.;
Otrubova, K.; McAlpine, S. R. J. Org. Chem. 2007, 72, 1980.
Her2
Hop
FKBP52
IP6K2
0
0
0
1
2
3
4
5
6
7
8
9
10 11
San A 1 (µM)
San A 2
120
110
100
90
80
70
60
50
40
30
20
10
0
Her2
Hop
FKBP52
IP6K2
1
2
3
4
5
6
µ
7
M)
8
9
10 11
San A 2 (
10. Compound 1-T-I was not made because it was known that the phenylalanine
residue at position I in San A-amide was important for binding.
11. It should be noted that 2-T-II was not synthesized as we have shown the
methyl-phenylalanine is critical for biological activity.
D-N-
17-AAG
120
110
100
90
80
70
60
50
40
30
20
10
0
12. See Supplementary data pS32 for experimental binding methods.
13. (a) Citri, A.; Gan, J.; Mosesson, Y.; Vereb, G.; Szollosi, J.; Yarden, Y. EMBO Rep.
2004, 1165; (b) Xu, W.; Mimnaugh, E.; Rosser, M. F.; Nicchitta, C.; Marcu, M.;
Yarden, Y.; Neckers, L. J. Biol. Chem. 2001, 276, 3702.
14. Chen, S. Y.; Sullivan, W. P.; Toft, D. O.; Smith, D. F. Cell Stress Chaperones 1998, 3,
118.
Her2
Hop
FKBP52
IP6K2
1
2
3
17-AAG (µM)
4
5
6
Figure 4. Compound 2 and client protein binding data.
Hsp90 Dimer
N-domain
unfolded
San
A
San
A
protein
M-domain
Hsp70
Complex
C-domain
inhibits
HOP
IP6K2,
FKBP52
Figure 5. Model of how San A analogs interrupt binding between Hsp90 and key
proteins.