Phosphorylation of Capsaicinoid Derivatives Provides Highly Potent and Selective Inhibitors of the Transcription Factor STAT5b

Article abstract of DOI:10.1021/acschembio.5b00817

Design approaches for inhibitors of protein-protein interactions are rare, but highly sought after. Here, we report that O-phosphorylation of simple derivatives of the natural products dihydrocapsaicin and N-vanillylnonanamide leads to inhibitors of the SH2 domain of the transcription factor STAT5b. The most potent molecule is obtained from dihydrocapsaicin in only three synthetic steps. It has submicromolar affinity for the SH2 domain of STAT5b (K_i = 0.34 μM), while displaying 35-fold selectivity over the highly homologous STAT5a (K_i = 13.0 μM). The corresponding pivaloyloxymethyl ester inhibits STAT5b with selectivity over STAT5a in human tumor cells. Importantly, it inhibits cell viability and induces apoptosis in human tumor cells in a STAT5-dependent manner. Our data validate O-phosphorylation of appropriately preselected natural products or natural product derivatives as a semirational design approach for small molecules that selectively inhibit phosphorylation-dependent protein-protein interaction domains in cultured human tumor cells.

Full text of DOI:10.1021/acschembio.5b00817

Articles
pubs.acs.org/acschemicalbiology
Phosphorylation of Capsaicinoid Derivatives Provides Highly Potent
and Selective Inhibitors of the Transcription Factor STAT5b
Nagarajan Elumalai,† Angela Berg,† Stefan Rubner, and Thorsten Berg*
Institute of Organic Chemistry, University of Leipzig, Johannisallee 29, 04103 Leipzig, Germany
S
* Supporting Information
ABSTRACT: Design approaches for inhibitors of protein−protein interactions are rare, but highly sought after. Here, we report
that O-phosphorylation of simple derivatives of the natural products dihydrocapsaicin and N-vanillylnonanamide leads to
inhibitors of the SH2 domain of the transcription factor STAT5b. The most potent molecule is obtained from dihydrocapsaicin
in only three synthetic steps. It has submicromolar affinity for the SH2 domain of STAT5b (Ki = 0.34 μM), while displaying 35-
fold selectivity over the highly homologous STAT5a (Ki = 13.0 μM). The corresponding pivaloyloxymethyl ester inhibits
STAT5b with selectivity over STAT5a in human tumor cells. Importantly, it inhibits cell viability and induces apoptosis in human
tumor cells in a STAT5-dependent manner. Our data validate O-phosphorylation of appropriately preselected natural products or
natural product derivatives as a semirational design approach for small molecules that selectively inhibit phosphorylation-
dependent protein−protein interaction domains in cultured human tumor cells.
rotein−protein interactions mediate the functions of most
respective peptide motifs only in their phosphorylated state.
These interactions, collectively referred to as phosphorylation-
dependent protein−protein interactions, require a different
strategy.
1−3
Pproteins.
Selective targeting of protein interaction
domains with cell-permeable organic molecules opens the
door to functional modulation of the vast majority of proteins
in living organisms. Unlike most current chemical biology
approaches, inhibition of protein−protein interactions is not
restricted to more established drug targets such as cell surface
receptors and enzymes.4 This enables the functional analysis of
the vast majority of proteins for basic research studies and
allows the development of new therapeutic strategies to treat
unmet medical needs. However, generally applicable strategies
for the design of functional modulators of intracellular protein−
protein interactions are rare.5 For interactions mediated by α-
helices binding to hydrophobic grooves, numerous proteomi-
metic scaffolds have been developed,6,7 allowing rational design
of competitive inhibitors of the respective protein−protein
interaction. Another design approach targeting this class of
protein−protein interactions involves conformational stabiliza-
tion of the α-helical peptide motifs by introduction of
hydrophobic cross-links.8,9 However, these approaches remain
restricted to interactions mediated by α-helices, and new
approaches need to be developed which address the specific
requirements of other individual types of protein−protein
interactions.
By screening chemical libraries consisting of known bioactive
compounds, we recently identified O-phosphorylated salicylic
acid (Fosfosal, Figure 1A) and dexamethasone-21-phosphate as
inhibitors of two phosphorylation dependent protein−protein
interactions.10 The protein target of Fosfosal is the
phosphotyrosine-binding Src homology 2 (SH2) domain of
STAT5b, a transcription factor which is constitutively activated
in many human tumors. STAT5b has been validated as a target
for cancer therapy.11,12 Fosfosal was subsequently modified to
give Stafib-1 (Figure 1A), which displayed nanomolar affinity
against the STAT5b SH2 domain and which displays high
selectivity over STAT5a in biochemical and cell-based assays.13
Retrospective analysis of the composition of the screening
library and the profile of compound activities displayed in the
screen in which Fosfosal was identified10 revealed a high hit rate
from a subset of compounds which can be described as O-
phosphorylated natural product derivatives.10 Consequently, we
proposed O-phosphorylation of preselected natural products or
natural product derivatives as a straightforward approach for
the design of nonpeptidic ligands of phosphorylation-depend-
Cellular signaling via protein kinases and phosphatases is
mediated, to a large extent, by interactions between short
peptide motifs containing phosphorylated serine, threonine, or
tyrosine residues and protein domains which recognize the
Received: May 15, 2015
Accepted: October 15, 2015
© XXXX American Chemical Society
A
DOI: 10.1021/acschembio.5b00817
ACS Chem. Biol. XXXX, XXX, XXX−XXX

ACS Chemical Biology
Articles
Figure 1. Natural product based inhibitors of STAT5b. (A) Structures of known salicylic acid-derived inhibitors.10,13 (B) Synthesis of STAT5b
inhibitors based on the natural products dihydrocapsaicin and N-vanillylnonanamide. (i) Dibenzyl phosphite, CCl4, DIEA, DMAP, CH3CN, 0 °C, 1
h. (ii) Pd/C, H2, EtOH, 1 h. (iii) 1 M BBr3 in DCM, 16 h.
ent protein−protein interaction domains.10 The rationale for
this approach is that the combination of the intrinsic propensity
of natural products to interact with proteins,14 together with
the presence of a part of the molecule reminiscent of
phosphorylated amino acid side chains, should result in a
larger chance of binding between the small molecule and the
phosphorylation-dependent protein−protein interaction do-
main.
STAT5a and STAT5b in competitive binding assays based on
fluorescence polarization13,18 revealed their inhibitory activity
against the SH2 domains of both STAT5 proteins (Table 1).
The highest activity was observed for 3b, with an inhibitory
constant (Ki) of 49 4 μM against STAT5a and a Ki of 96
4
μM against STAT5b. In contrast, the nonphosphorylated
natural products 1a and 1b were completely inactive against
either STAT5 protein (Table 1). Thus, although the activities
of 3a/b against STAT5a/b are relatively weak, O-phosphor-
ylation of natural products was indeed an effective measure by
which to create starting points for inhibitor development.
Since the catechol bisphosphate structure had been essential
for high activity against STAT5b in the previous study,13 we
also aimed to convert the natural products to the 1,2-
dihydroxyphenyl (catechol) derivatives, followed by phosphor-
ylation of both hydroxyl groups. To this end, the catechol
moieties of dihydrocapsaicin (1a) and N-vanillylnonanamide
(1b) were liberated by demethylation using boron tribromide
(Figure 1B). Subsequently, the hydroxyl groups of diols 4a and
4b were converted to dibenzylphosphates 5a and 5b by
reaction with dibenzylphosphite. Finally, Pd/C-catalyzed
hydrogenation resulted in the bisphosphates 6a and 6b.
RESULTS AND DISCUSSION
■
In order to validate our design concept for inhibitors of
phosphorylation-dependent protein−protein interactions, we
turned our attention to the natural product class of the
capsaicinoids. Most peppers contain capsaicinoids such as
dihydrocapsaicin (1a) and N-vanillylnonanamide (1b),15−17
which are responsible for the fruits’ pungency (Figure 1B).
Both capsaicinoids possess a masked catechol moiety bearing
an N-acylated aminomethyl group, reminiscent of structural
features contained in Stafib-1, and are therefore promising
starting points for the development of STAT5 inhibitors.
Conversion of the two natural products 1a/b to the benzyl
phosphates 2a/b and their subsequent hydrogenation afforded
O-phosphorylated dihydrocapsaicin (3a) and N-vanillylnona-
namide (3b) in good yields. Analysis of their activities against
Biochemical analysis of the bisphosphates 6a and 6b revealed
both compounds to be submicromolar inhibitors of the
B
DOI: 10.1021/acschembio.5b00817
ACS Chem. Biol. XXXX, XXX, XXX−XXX

ACS Chemical Biology
Articles
Table 1. Activity of Natural Products and Phosphorylated Derivatives in a Fluorescence Polarization Assay against the SH2
a
Domain of STAT5b and STAT5a
a
N/A = not applicable. Conversion of IC50 data to Ki values was carried out using the published equation.19
STAT5b domain (Table 1).18 6a (Ki = 0.34
0.01 μM) is
moiety of compound 6a into the phosphotyrosine pocket of
STAT5b (Figure 2B).13 Binding of the catechol bisphosphate
moiety to the phosphotyrosine binding pocket of the STAT5b
SH2 domain had previously been experimentally confirmed
using point mutants in a direct binding assay.13 The amide
bond is assumed to be involved in hydrogen bonds with the
backbone of Asn642 (Figure S1), and the hydrophobic tail of
the molecule is placed along the indole ring of Trp641,
reaching toward Tyr665.
In order to analyze the ability of 6a to selectively inhibit the
STAT5b SH2 domain in cultured cells, 6a was converted to the
pivaloyloxymethylester 7 (Figure 3A). This class of prodrugs
has been thoroughly demonstrated to mediate cell permeability
of phosphonates24−26 and phosphates13,27 and liberates the
active agent after intracellular cleavage of the pivaloyloxymethyl
groups by esterases. Since the function of the SH2 domain is
required for activation of STAT5 proteins by phosphorylation
of the conserved tyrosine residue (STAT5b, Tyr699; STAT5a,
Tyr694) by tyrosine kinases such as Bcr-Abl,28 inhibition of the
SH2 domain leads to reduced tyrosine phosphorylation (Figure
3B). This can be monitored by Western blot analysis using
slightly more potent than 6b (Ki = 0.39 0.03 μM). Thus, 6a
is almost 3 times as potent as catechol bisphosphate (Ki = 0.93
0.07 μM).13 Both 6a and 6b displayed high selectivity over
the SH2 domain of the close homologue STAT5a, which is
93% identical on the amino acid level20 (6a, Ki (STAT5a) =
13.0 0.7 μM; 6b, Ki (STAT5a) = 12.3 0.7 μM).
In-depth analysis of the selectivity profile of the best
compound 6a against the SH2 domains of other STAT
proteins and of the tyrosine kinase Lck21 confirmed the high
degree of selectivity for STAT5b (Figure 2A and Table S1). 6a
displayed more than 35-fold selectivity for STAT5b over
STAT5a, more than 20-fold selectivity over STAT6, and more
than 45-fold selectivity over STAT4. The SH2 domains of
STAT1, STAT3, and Lck were inhibited to an even lesser
extent. These data join our previous results13 in challenging the
dogma that SH2 domains cannot be targeted selectively with
small organic molecules.
Molecular docking with AutoDock Vina22 using a homology
model of the STAT5b SH2 domain10 (based on the crystal
structure of STAT5a)23 placed the catechol bisphosphate
C
DOI: 10.1021/acschembio.5b00817
ACS Chem. Biol. XXXX, XXX, XXX−XXX

ACS Chemical Biology
Articles
phosphates in the cellular milieu, we conducted a time-course
experiment using 10 μM of 7 on STAT5b-GFP-transfected
K562 cells. The compound was found to exert its strongest
effect between 1 and 4 h (Figure S2), with inhibitory activity
partially reduced after 8 h of exposure.
Although leukemia cells transformed by the Bcr-Abl
oncoprotein display constitutive activation of both STAT5a
and STAT5b,29 there is strong indication that the activation of
STAT5b is of larger functional relevance than the activation of
STAT5a. Inhibition of Bcr-Abl mediated signaling by a
dominant negative STAT5b mutant, which lacks the ability to
dimerize via reciprocal phosphotyrosine-SH2 domain inter-
actions, strongly inhibited cell growth, and viability of Bcr-Abl
expressing cells.30 Proliferation of K562 cells is inhibited by
shRNA-mediated depletion of STAT5b to a larger extent than
by depletion of STAT5a.31 In addition, short hairpin RNA
(shRNA)-mediated depletion of STAT5b, but not shRNA-
mediated depletion of STAT5a, sensitizes K562 cells to the
Bcr-Abl inhibitor Imatinib.31 Sh-mediated depletion of STAT5b
caused a strong increase in the apoptotic rate in Bcr-Abl
expressing leukemia cells, while depletion of STAT5a was
ineffective.32 Consequently, selective inhibition of STAT5b has
been proposed as a therapeutic strategy to combat tumori-
genesis driven by Bcr-Abl.31
Cell viability assays conducted on K562 cells indicated a
dose-dependent effect of 7 (Figure 4A). To exclude the
possibility that the reduced cell viability was caused by an effect
of 7 on targets other than STAT5 proteins, we also tested the
effect of 7 on MDA-MB-231 cells, which do not harbor
constitutively active STAT5a/b (Figure S3) and thus do not
proliferate via STAT5 signaling. We observed virtually no
inhibition of the cell viability in MDA-MB-231 cells in the
presence of 7 (Figure 4B), indicating that the reduction of cell
viability in K562 cells was caused by a STAT5-dependent effect.
In order to analyze whether the reduced cell viability of K562
cells in the presence of 7 was caused by an increased rate of
apoptosis, we analyzed K562 cells treated with 7 for exposure of
phosphatidylserine on the outside of the cell membrane by
annexin V staining, followed by flow cytometry. K562 cells
treated with 7 showed a dose-dependent increase in apoptotic
cells (Figure 4C and Figure S4). In contrast, there was no
significant increase in apoptotic cells in the STAT5-
independent control cell line MDA-MB-231 (Figure 4D and
Figure S5). These data strongly indicate that the induction of
apoptosis in K562 cells is caused by a STAT5-dependent effect.
Given the high selectivity of 7 against STAT5b over STAT5a
(Figure 3C−F) and the literature reports31,32 describing the
crucial relevance of STAT5b, but not STAT5a, for proliferation
and survival of Bcr-Abl expressing tumor cells, our data strongly
suggest that the effects of 7 on K562 cells are mediated via the
inhibition of STAT5b. Activity and specificity of 7 against the
individual STAT5 proteins at the time point at which cell
viability and apoptosis were analyzed (48 h) cannot be reliably
tested using STAT5a/b-transfected cells because of the
transient nature of the plasmid transfection, which would lead
to a loss of the STAT5a/b expression plasmid in the course of
the experiment. We attribute the higher concentrations
required to achieve the effects in cell viability assays and in
apoptosis assays to the longer time of exposure in these assays,
which presents a larger challenge for the stability of the active
agent against phosphatases, as indicated by the reduced activity
of 7 already after 8 h of exposure (Figure S2).
Figure 2. (A) Selectivity profile of 6a in fluorescence polarization
assays. (B) Binding mode of 6a to the STAT5b SH2 domain as
proposed by AutoDock Vina.
phospho-specific antibodies that recognize STAT5b only when
phosphorylated at Tyr699. However, these antibodies also
recognize STAT5a when phosphorylated at Tyr694, since they
are raised against a short peptide motif which is identical in
both STAT5 proteins. To enable us to differentiate between the
phosphorylated forms of the proteins, we designed fusion
constructs between STAT5a or STAT5b and GFP.13 K562
cells, in which STAT5 proteins are constitutively activated by
phosphorylation of STAT5b Tyr699 and STAT5a Tyr694 by
Bcr-Abl, were transfected with plasmids encoding either
STAT5a-GFP or STAT5b-GFP. The added molecular weight
of the GFP allows for differentiation between endogenous
STAT5, consisting of both STAT5a and STAT5b, and the
respective transfected GFP-STAT5 protein. Treatment of
STAT5b-GFP-transfected K562 cells with 7 led to a strong,
dose-dependent decrease of STAT5b Tyr699 phosphorylation
(Figure 3C, D). In contrast, phosphorylation of the STAT5a-
GFP fusion protein at STAT5a Tyr694 was reduced only to a
minor extent (Figure 3E,F). This demonstrates that 6a is able
to inhibit the phosphorylation of STAT5b in cells, and that
specificity for STAT5b over STAT5a is maintained in the
cellular environment to a major extent. Tyrosine phosphor-
ylation of endogenous STAT5 was inhibited to a lesser extent
than STAT5b-GFP and a higher extent than STAT5a-GFP
(Figure 3G,H), reflecting the presence of both STAT5 proteins
in K562 cells.
Phosphates are prone to intracellular enzymatic cleavage by
phosphatases. To investigate the stability of 7 against
D
DOI: 10.1021/acschembio.5b00817
ACS Chem. Biol. XXXX, XXX, XXX−XXX

ACS Chemical Biology
Articles
Figure 3. Western blot analysis of compound 7 on K562 cells. (A) Synthesis of compound 7. (B) In K562 leukemia cells, STAT5a/b are
constitutively phosphorylated on STAT5a Tyr694 and STAT5b Tyr699 by Bcr-Abl, leading to signal transduction via STAT5a/b. An inhibitor of the
SH2 domain (symbolized by the triangle) prevents STAT5 tyrosine phosphorylation and subsequent signal transduction. (C) 7 inhibits
phosphorylation of STAT5b-GFP in K562 cells in a dose-dependent manner. (D) Quantification of the pSTAT5b-GFP bands from C, normalized
against total STAT5b-GFP. Error bars represent the standard deviations from two independent experiments. (E) 7 has only a minor effect on
tyrosine phosphorylation of STAT5a-GFP in K562 cells. (F) Quantification of the pSTAT5a-GFP bands, normalized against total STAT5a-GFP.
Error bars represent the standard deviations from two independent experiments. (G) Effect of 7 on endogenous STAT5a/b in K562 cells. (H)
Quantification of the endogenous pSTAT5 bands, normalized against total endogenous STAT5. Error bars represent the standard deviations from
four independent experiments.
tumor cells. Our data validate O-phosphorylation of appropri-
ately preselected natural products or derivatives thereof as a
semirational design approach for small molecules that
selectively inhibit phosphorylation-dependent protein−protein
interaction domains in cultured human tumor cells. On a wider
note, since the discussed chemical transformations involved in
the conversion of the capsaicinoids (demethylation and
phosphorylation) could also be carried out by plant enzymes,
our study might stimulate natural product chemists to attempt
the isolation of phosphorylated and thus water-soluble natural
products, which may well be overlooked using current natural
product isolation protocols.33
CONCLUSIONS
■
In summary, we have demonstrated that O-phosphorylation of
the widely abundant natural products dihydrocapsaicin and N-
vanillylnonanamide, or simple derivatives thereof, creates
chemical entities with activity against STAT5a/b. The bi-
sphosphates 6a/b based on the demethylated compounds 4a/b
are nanomolar inhibitors of the STAT5b SH2 domain and
display a high degree of selectivity over the highly homologous
SH2 domain of STAT5a. Conversion of the most potent
compound 6a, dubbed Capstafin (for “capsaicinoid-based
STAT five b inhibitor”), to its pivaloyloxymethylester resulted
in compound 7, which caused selective inhibition of STAT5b
phosphorylation, STAT5-dependent reduction in cell viability,
and a STAT5-dependent increased rate of apoptosis in human
E
DOI: 10.1021/acschembio.5b00817
ACS Chem. Biol. XXXX, XXX, XXX−XXX

ACS Chemical Biology
Articles
Figure 4. Compound 7 inhibits cell viability and increases the rate of apoptosis in tumor cells in a STAT5-dependent manner. (A) Effect of 7 on cell
viability on STAT5-dependent K562 cells and (B) on STAT5-independent MDA-MB-231 cells. Cells were treated for 48 h. (C) Effect of 7 on the
rate of apoptosis in STAT5-dependent K562 cells and (D) in STAT5-independent MDA-MB-231 cells. Cells were treated for 48 h. Error bars depict
the standard deviations from three independent experiments.
service group (L. Hennig) and the mass spectrometry group
(C. Birkemeyer) of the Faculty of Chemistry and Mineralogy at
Leipzig University. This work was supported by the Core Unit
Fluorescence-Technologies of the Interdisciplinary Centre for
Clinical Research (IZKF) Leipzig, at the Faculty of Medicine of
METHODS
■
Details on chemical synthesis, spectroscopic characterization, plasmid
construction, and assay design are summarized in the Supporting
Information.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acschem-
bio.5b00817.
■
the University of Leipzig (K. Jager and A. Losche).
̈
̈
S
REFERENCES
■
(1) Wells, J. A., and McClendon, C. L. (2007) Reaching for high-
hanging fruit in drug discovery at protein-protein interfaces. Nature
450, 1001−1009.
Table S1, Figures S1−S5, plasmid construction and
protein expression, fluorescence polarization assays, cell
culture, Western blots, cell viability assay, apoptosis
assay, synthesis and characterization of compounds,
supporting references, NMR spectra (PDF)
(2) Berg, T. (2008) Small-molecule inhibitors of protein-protein
interactions. Curr. Opin. Drug Discovery Dev. 11, 666−674.
(3) Milroy, L. G., Grossmann, T. N., Hennig, S., Brunsveld, L., and
Ottmann, C. (2014) Modulators of protein-protein interactions. Chem.
Rev. 114, 4695−4748.
(4) Hopkins, A. L., and Groom, C. R. (2002) The druggable genome.
Nat. Rev. Drug Discovery 1, 727−730.
(5) Azzarito, V., Long, K., Murphy, N. S., and Wilson, A. J. (2013)
Inhibition of alpha-helix-mediated protein-protein interactions using
designed molecules. Nat. Chem. 5, 161−173.
(6) Cummings, C. G., and Hamilton, A. D. (2010) Disrupting
protein-protein interactions with non-peptidic, small molecule alpha-
helix mimetics. Curr. Opin. Chem. Biol. 14, 341−346.
(7) Jayatunga, M. K., Thompson, S., and Hamilton, A. D. (2014)
alpha-Helix mimetics: outwards and upwards. Bioorg. Med. Chem. Lett.
24, 717−724.
AUTHOR INFORMATION
Corresponding Author
*E-mail: tberg@uni-leipzig.de.
■
Author Contributions
†These authors contributed equally
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was generously supported by the Deutsche
Forschungsgemeinschaft (BE 4572/4-1 and INST 268/281-1
FUGG). We extend our thanks to K. Natarajan for
experimental support. We extend out thanks to the NMR
(8) Cromm, P. M., Spiegel, J., and Grossmann, T. N. (2015)
Hydrocarbon Stapled Peptides as Modulators of Biological Function.
ACS Chem. Biol. 10, 1362.
F
DOI: 10.1021/acschembio.5b00817
ACS Chem. Biol. XXXX, XXX, XXX−XXX

ACS Chemical Biology
Articles
(9) Walensky, L. D., and Bird, G. H. (2014) Hydrocarbon-stapled
peptides: principles, practice, and progress. J. Med. Chem. 57, 6275−
6288.
(29) de Groot, R. P., Raaijmakers, J. A., Lammers, J. W., Jove, R., and
Koenderman, L. (1999) STAT5 activation by BCR-Abl contributes to
transformation of K562 leukemia cells. Blood 94, 1108−1112.
(30) Sillaber, C., Gesbert, F., Frank, D. A., Sattler, M., and Griffin, J.
D. (2000) STAT5 activation contributes to growth and viability in
Bcr/Abl-transformed cells. Blood 95, 2118−2125.
(31) Schaller-Schonitz, M., Barzan, D., Williamson, A. J., Griffiths, J.
R., Dallmann, I., Battmer, K., Ganser, A., Whetton, A. D., Scherr, M.,
and Eder, M. (2014) BCR-ABL affects STAT5A and STAT5B
differentially. PLoS One 9, e97243.
(32) Casetti, L., Martin-Lanneree, S., Najjar, I., Plo, I., Auge, S., Roy,
L., Chomel, J. C., Lauret, E., Turhan, A. G., and Dusanter-Fourt, I.
(2013) Differential contributions of STAT5A and STAT5B to stress
protection and tyrosine kinase inhibitor resistance of chronic myeloid
leukemia stem/progenitor cells. Cancer Res. 73, 2052−2058.
(33) Shimizu, Y., and Li, B. (2005) Purification of Water-Soluble
Natural Products, in Natural Products Isolation (Sarker, S., Latif, Z., and
Gray, A., Eds.), pp 415−438, Humana Press.
(10) Graber, M., Janczyk, W., Sperl, B., Elumalai, N., Kozany, C.,
̈
Hausch, F., Holak, T. A., and Berg, T. (2011) Selective Targeting of
Disease-Relevant Protein Binding Domains by O-Phosphorylated
Natural Product Derivatives. ACS Chem. Biol. 6, 1008−1014.
(11) Darnell, J. E., Jr. (2002) Transcription factors as targets for
cancer therapy. Nat. Rev. Cancer 2, 740−749.
(12) Yu, H., and Jove, R. (2004) The STATs of cancer–new
molecular targets come of age. Nat. Rev. Cancer 4, 97−105.
(13) Elumalai, N., Berg, A., Natarajan, K., Scharow, A., and Berg, T.
(2015) Nanomolar Inhibitors of the Transcription Factor STAT5b
with High Selectivity over STAT5a. Angew. Chem., Int. Ed. 54, 4758−
4763.
(14) Breinbauer, R., Vetter, I. R., and Waldmann, H. (2002) From
protein domains to drug candidates-natural products as guiding
principles in the design and synthesis of compound libraries. Angew.
Chem., Int. Ed. 41, 2879−2890.
(15) Kozukue, N., Han, J. S., Kozukue, E., Lee, S. J., Kim, J. A., Lee,
K. R., Levin, C. E., and Friedman, M. (2005) Analysis of eight
capsaicinoids in peppers and pepper-containing foods by high-
performance liquid chromatography and liquid chromatography-mass
spectrometry. J. Agric. Food Chem. 53, 9172−9181.
(16) Maillard, M.-N., Giampaoli, P., and Richard, H. M. J. (1997)
Analysis of Eleven Capsaicinoids by Reversed-phase High Performance
Liquid Chromatography. Flavour Fragrance J. 12, 409−413.
(17) Constant, H. L., Cordell, G. A., and West, D. P. (1996)
Nonivamide, a constituent of Capsicum oleoresin. J. Nat. Prod. 59,
425−426.
(18) Muller, J., Schust, J., and Berg, T. (2008) A high-throughput
̈
assay for signal transducer and activator of transcription 5b based on
fluorescence polarization. Anal. Biochem. 375, 249−254.
(19) Nikolovska-Coleska, Z., Wang, R., Fang, X., Pan, H., Tomita, Y.,
Li, P., Roller, P. P., Krajewski, K., Saito, N. G., Stuckey, J. A., and
Wang, S. (2004) Development and optimization of a binding assay for
the XIAP BIR3 domain using fluorescence polarization. Anal. Biochem.
332, 261−273.
(20) Hennighausen, L., and Robinson, G. W. (2008) Interpretation
of cytokine signaling through the transcription factors STAT5A and
STAT5B. Genes Dev. 22, 711−721.
(21) Sperl, B., Seifert, M. H., and Berg, T. (2009) Natural product
inhibitors of protein-protein interactions mediated by Src-family SH2
domains. Bioorg. Med. Chem. Lett. 19, 3305−3309.
(22) Trott, O., and Olson, A. J. (2010) AutoDock Vina: improving
the speed and accuracy of docking with a new scoring function,
efficient optimization, and multithreading. J. Comput. Chem. 31, 455−
461.
(23) Neculai, D., Neculai, A. M., Verrier, S., Straub, K., Klumpp, K.,
Pfitzner, E., and Becker, S. (2005) Structure of the unphosphorylated
STAT5a dimer. J. Biol. Chem. 280, 40782−40787.
(24) Mandal, P. K., Gao, F., Lu, Z., Ren, Z., Ramesh, R., Birtwistle, J.
S., Kaluarachchi, K. K., Chen, X., Bast, R. C., Liao, W. S., and
McMurray, J. S. (2011) Potent and Selective Phosphopeptide Mimetic
Prodrugs Targeted to the Src Homology 2 (SH2) Domain of Signal
Transducer and Activator of Transcription 3. J. Med. Chem. 54, 3549−
3563.
(25) Mandal, P. K., Liao, W. S., and McMurray, J. S. (2009) Synthesis
of phosphatase-stable, cell-permeable peptidomimetic prodrugs that
target the SH2 domain of Stat3. Org. Lett. 11, 3394−3397.
(26) Morlacchi, P., Mandal, P. K., and McMurray, J. S. (2014)
Synthesis and in Vitro Evaluation of a Peptidomimetic Inhibitor
Targeting the Src Homology 2 (SH2) Domain of STAT6. ACS Med.
Chem. Lett. 5, 69−72.
(27) Zhao, S., and Etzkorn, F. A. (2007) A phosphorylated prodrug
for the inhibition of Pin1. Bioorg. Med. Chem. Lett. 17, 6615−6618.
(28) Berg, T. (2008) Signal Transducers and Activators of
Transcription as Targets for Small Organic Molecules. ChemBioChem
9, 2039−2044.
G
DOI: 10.1021/acschembio.5b00817
ACS Chem. Biol. XXXX, XXX, XXX−XXX