Small-molecules,peptides,natural
products:gettingagripon14-3-3
protein-proteinmodulation
Authors
Maria Bartel1, Anja Schäfer1, Loes Stevers1, Christian Ottmann1,*
1
Laboratory of Chemical Biology, Department of Biomedical Engineering, Technische Universiteit
Eindhoven, Den Dolech, 5612 AZ Eindhoven, The Netherlands
*
corresponding author
For readability, the general term 14-3-3 will be used to name the protein, although it is well known
that this protein occurs in seven isoforms. For most of the studies, only one isoform was used and
described in the results of the studies that are presented here.
Abstract
One of the proteins that is found in a diverse range of eukaryotic protein-protein interaction is the
adaptor protein 14-3-3 . It was already discovered in the late sixties and then interest in 14-3-3 rose
again in the nineties of the twentieth century. The first protein crystal structures were solved at that
time and gave molecular insights for further research. The plant analog of this protein binds to a plant
plasma membrane H+-ATPase and this protein complex is stabilized by the fungal phytotoxin
Fusicccocin A. As a cause of that the proton transport becomes overactive and the stomata are
permanently opened, which finally leads to wilting of the plant and delivers metabolites for the
survival of the fungus Phomopsis amygdali. As 14-3-3 proteins are also found in humans, the
knowledge gained from the plant process was transferred to and applied in human models to find
stabilizers or inhibitors of 14-3-3 interaction in human cellular pathways. Studying disease related
proteins is one of the research tasks in chemical biology. The overall aim is to find valuable targets
and address interaction on a molecular level to in the long run find useful modulators of diseases in
humans. But as the interactions of proteins are very diverse, also various studies provide molecular
insights into 14-3-3 protein-protein interactions. A diverse range of classes of molecules, peptides,
small-molecules or natural products, is used to modify the protein interactions, providing both,
stabilization or inhibition of the interaction of 14-3-3 with its binding partner.
Keywords:
14-3-3 proteins, protein-protein interaction, protein-protein inhibition, protein-protein stabilization,
small molecules, natural products, peptides
1
Abbreviations:
chronic myeloid leukemia (CML)
high throughput screening (HTS)
plant plasma membrane H+-ATPase (PMA)
Fusicoccin A (FC)
Fusicoccin THF (FC-THF)
Pyrrolidone1 (Pyr1)
small molecules microarray (SMM)
chronic myeloid leukemia (CML)
green fluorescent protein (GFP)
structure-activity relationship (SAR)
protein protein interaction (PPI)
time-resolved fluorescence resonance energy transfer (TR-FRET)
2
TableofContents
Authors.................................................................................................................................................... 1
Abstract ................................................................................................................................................... 1
Key words:............................................................................................................................................... 1
Abbreviations: ......................................................................................................................................... 2
Table of figures ....................................................................................................................................... 4
Introduction ............................................................................................................................................ 6
Protein protein interaction (PPI) ......................................................................................................... 6
Introduction 14-3-3 ............................................................................................................................. 6
Inhibitors of 14-3-3 interaction ............................................................................................................... 8
Peptides binding in the amphipathic groove. ..................................................................................... 8
On the way to small molecules: peptide-small molecule hybrids. ..................................................... 8
Small molecules: Struggling to overcome the phosphate dependency ............................................. 9
Natural products: The solution? ....................................................................................................... 11
14-3-3 PPI stabilizers ............................................................................................................................. 12
Natural product PPI stabilizers.......................................................................................................... 12
Non-natural product 14-3-3 PPI stabilizers....................................................................................... 13
PPI stabilizers in drug discovery ........................................................................................................ 13
Chemical induced dimerization by a 14-3-3 PPI stabilizer ................................................................ 14
Different approaches lead to goals ....................................................................................................... 16
Phage display for finding the first peptide based inhibitor .............................................................. 16
Discovery of small-molecule inhibitors with Virtual Screening ........................................................ 17
Identifying a chemistry starting point by HTS ................................................................................... 17
Future perspective and executive summary......................................................................................... 18
References ............................................................................................................................................ 19
Figures ................................................................................................................................................... 24
3
Tableoffigures
Figure 1: Publication entries in PubMed of the last 20 years on “inhibitors of protein protein
interaction” and on “14-3-3 ” (until Dec 2013). Publications on PPI inhibitors started to appear in the
60s, trend is rising still, though including the keyword 14-3-3 to the search gave only 144 hits in total.
Although 14-3-3 proteins were first reported by Moore and Perez in 1968, [12] the publications on
14-3-3 are constantly accelerating since the middle of the 90s, when the first crystallographic
structures were published. (Figure made using GraphPad Prism version 6.02 for Windows, GraphPad
Software, San Diego California USA, www.graphpad.com) ------------------------------------------------- 24
Figure 2: Conserved residues of 14-3-3 : Amino acid sequence alignment of the seven human
isoforms of 14-3-3 performed with ClustalOmega. (Sievers et al. 2011) Residue conservation was
determined with ConSurf [67] Corresponding protein entries (UniProt): 14-3-3 ε-P62258, 14-3-3 σP31947, 14-3-3 η-Q04917, 14-3-3 γ-P61981, 14-3-3 τ-P27348, 14-3-3 β-P31946, 14-3-3 ζ-P63104
[68] -------------------------------------------------------------------------------------------------------------------- 25
Figure 3: Overview of known small molecule inhibitors of 14-3-3 ----------------------------------------- 27
Figure 4: The phosphorylation independent peptidic inhibitor of 14-3-3 : R18 (chocolate colored
surface), the aspartate of the short sequence WLDLE is coordinated into the amphipathic groove of
14-3-3 ζ (gray surface). Binding pocket of 14-3-3 ζ (K49, R56, R127, Y128) surface colored by
element green C, gray H, blue N, red O [24] (PDB ID: 1A38 ----------------------------------------------- 28
Figure 5: Peptide staples (lightgreen sticks) improving the binding of the wild type sequence of ExoS
to 14-3-3 : Wild type sequence peptide of ExoS: ESp (lightorange surface) and the binding enhancing
stapled peptides: βSS12 staple (sulfur yellow colored surface), βRS8 staple (brightorange colored
surface). Binding pocket of 14-3-3 ζ (K49, R56, R127, Y128) surface colored by element green C,
gray H, blue N, red O [23] (PDB ID: 4N7G, 4N7Y, 4N84) -------------------------------------------------- 28
Figure 6: The FOurteen three three BInding INhibitor 101 (FOBISIN101, (skyblue sticks) bound to
Lys120 (element colored sticks/surface) of 14-3-3 (gray surface). Binding pocket of 14-3-3 ζ (K49,
R56, R127, Y128) surface colored by element green C, gray H, blue N, red O [69] (PDB ID: 3RDH)
-------------------------------------------------------------------------------------------------------------------------- 29
Figure 7: Phosphonate inhibitors A) overlay of all hits which shows that the phosphate moiety
coordinates the orientation of the molecules into the pocket B) A1 (lightblue), C) A2 (paleyellow)
with alternative conformation in crystal structure, D) B1 wheat colored, E) B2 palegreen binding to
14-3-3 (gray surface) found in virtual screening: Binding pocket of 14-3-3 σ (K49, R56, R129, Y130)
surface colored by element green C, gray H, blue N, red O [35] (PDB ID: 4DHV, 4DHT, 3T0L,
3T0M) ----------------------------------------------------------------------------------------------------------------- 30
Figure 8: Molecular tweezer (cyan colored sticks) binding to Lys214 (green sticks/light gray surface)
of 14-3-3 σ (gray surface) and blocking the amphipathic groove. Binding pocket of 14-3-3 σ (K49,
R56, R129, Y130) surface colored by element green C, gray H, blue N, red O [32] (PDB ID: 4HQW)
-------------------------------------------------------------------------------------------------------------------------- 31
Figure 9: Overview of known small molecule stabilizers of 14 3 3 interactions -------------------------- 32
Figure 10: Fusiccocin A (smudge colored sticks), the first known natural product stabilizer of the
14-3-3 interaction with the PMA2 (tv_blue colored surface). Binding pocket of T14-3-3 c (K56, R63,
R136, Y137) surface colored by element green C, gray H, blue N, red O [51] (PDB ID: 1O9F) ------ 33
4
Figure 11: Peptide sequence of Raf (yelloworange surface) binding to 14-3-3 (gray surface) stabilized
by Cotylenin A (deepteal sticks): The binding of the peptide to 14-3-3 ζ is coordinated via its
phosphorylated serine. Binding pocket of 14-3-3 ζ (K49, R56, R127, Y128) surface colored by
element: green C, gray H, blue N, red O [57] (PDB ID: 4IHL) ---------------------------------------------- 33
Figure 12: FC-THF (black sticks), a modified Fusicoccane, to exclusively stabilize mode-3
interactions binding to 14-3-3 and the phosphopeptide TASK3 (sand colored surface) . Binding pocket
of 14-3-3 σ (K49, R56, R129, Y130) surface colored by element: green C, gray H, blue N, red O [55]
(PDB ID: 3SPR) ----------------------------------------------------------------------------------------------------- 34
Figure 13: A) Pyrrolidone1 (yellow) and B) Epibestatin (orange): First small-molecule non-natural
product stabilizers of T14-3-3 c (gray surface) and the C-terminal PMA2-CT30 (tv_blue colored
surface). Binding pocket of T14-3-3 c (K56, R63, R136, Y137) surface colored by element: green C,
gray H, blue N, red O [20] (PDB ID: 3M51 and 3M50) ------------------------------------------------------ 35
Figure 14: Optimized stabilizer molecule Compound 34 (olive colored sticks) for the interaction of
PMA-CT30 (tv_blue colored surface) and T14-3-3 e (gray surface). Binding pocket of T14-3-3 e
(K54, R61, R134, Y135) surface colored by element green C, gray H, blue N, red O [58] (PDB ID:
4DX0) ----------------------------------------------------------------------------------------------------------------- 36
5
Introduction
Protein protein interaction (PPI)
Cellular processes are regulated through interactions between proteins. These interactions are based
on a direct, non-covalent contact between at least two proteins. Biological networks are organized
through protein-protein interactions [1,2].
A general definition of protein-protein interactions is that they share physical contacts in a living
organism. Certainly these physical contacts are specific, not all proteins just interact with each other
when they are in close proximity. The contacts may not be permanent or static as a cell itself
undergoes permanent changes. Some cellular tasks are performed by protein assemblies. These
interact as a complex having different subunits, for example the RNA polymerase complex [3].
The protein-protein interaction can be of a different nature and a differentiation can be made between
permanent and transient. Furthermore, the protein-protein interaction can be described by its
dissociation constant. The binding affinity is inversely related to the dissociation constant.
Additionally to the binding itself, the interaction can be modulated by conformational changes in the
proteins or an effector molecule [1,4]. Small-molecules, peptides, natural products and hybrids of
these classes can be such effectors.
Most of the pharmacologically used substances function as inhibitors of enzymes or as allosteric
modulators. To date, the focus in research still lies on inhibition defined as the competitive
replacement of ligands [5].
Small-molecule stabilization of PPIs can be achieved by two general approaches. The stabilizer can
bind to an allosteric site of one of the proteins resulting in a higher binding affinity for the second
protein; or the stabilizer can bind to the interfacial surface of both proteins for a direct increase of
binding affinity. Focusing on the direct PPI stabilization, there are two possible courses of action: a
small-molecule can bind to one of the proteins to create or modify the interaction surface for the
second protein or a small-molecule can bind to an already established protein-protein interface to
increase the binding affinity of the two proteins. [6]
Upon database search (pubmed.gov) for "inhibitors of protein-protein interaction", around 38.000 hits
can be found, whereas "stabilizers of protein-protein interaction" yields less than a hundred hits (78,
Dec 2013) (Figure 1). This might indicate that exploring this subject is a worthwhile challenge for
future research.
Introduction 14-3-3
The proteins of the 14-3-3 family have been found in all eukaryotic organisms. They belong to a class
of highly conserved small adaptor proteins with no enzymatic activity. In human cells seven isoforms
of 14-3-3 proteins could be identified which are named with the greek letters β, γ, ε, ζ, η, τ and σ.
They are approximately 30 kDa acidic proteins and can form homo- and heterodimers. These proteins
have been shown to interact with other cellular proteins. All the diverse interactions these proteins can
undergo make them frequently studied proteins. The known interactions are as diverse as a cell with
all its processes itself. Just to name a few, 14-3-3 proteins interact with the Raf Kinase which is a
component of the Ras-RAF pathway. Mutations in this pathway can be found in 30 % of all human
cancer cells. Binding to 14-3-3 keeps Raf in an inactive state and the Ras-Raf interaction is disrupted
[7]. Furthermore, 14-3-3 proteins have been found to bind the regulatory C-terminal domain of the
6
tumor suppressor protein p53. Upon binding of the tumor suppressor to 14-3-3 , the degradation of
p53 through MDM2 is prevented and the p53 is kept in a tetrameric state. [8]. After phosphorylation
of CDC25 it can bind to 14-3-3 and is the sequestered in the cytoplasm. Finally, CDC25 then cannot
activate CDC2 anymore and the entry into mitosis is prevented. [9] The subcellular localization of
Bcr-Abl is also influenced by 14-3-3 . Mutations in Bcr-Abl are one reason for resistance against
drugs for chronic myeloid leukemia (CML). When binding to 14-3-3 the Bcr-Abl is kept in the
cytoplasm, but if this interaction is disrupted upon DNA damage, Abl will be translocated to the
nucleus and apoptosis will be activated. [10]
Their role in these interactions makes 14-3-3 proteins a valuable target in cancer research, in
Alzheimer’s disease or cystic fibrosis research. [9,11] These proteins were already identified in 1968
by Moore and Perez, [12] but even as they already have been known for a long time, 14-3-3 proteins
are still not fully described. They are adaptor proteins for more than 200 other proteins, and therefore
play an important role in diverse cellular processes [13].
14-3-3 proteins mainly bind phosphorylation-dependent to their partner proteins. Binding to a partner
protein of 14-3-3 is mediated by phosphorylated serines or threonines. These are often located in
intrinsically disordered domains of the binding partner protein and usually contain a linear recognition
sequence. These binding motifs occur in three different modes: mode 1: RSXpSXP, mode 2:
RXF/YXpSXP and mode 3: X-pS/T-X-COOH [14–16]. Lysin, Arginin and Tyrosine residues in the
amphipathic binding groove of 14-3-3 coordinate the phosphorylated serine or threonine mediate the
binding to its partner proteins through electrostatic interactions.
About 30 years after their discovery, the first structures of the protein were published by two different
groups simultaneously in 1995: Liu [17] and Xiao [18]. After that point, research on 14-3-3 proteins
came into focus again (Figure 1) with dozens of publications each year.
After comparison of sequence alignments between different isoforms of 14-3-3 , it becomes obvious
that the binding channel, the amphipathic groove, is highly conserved. The differences between the
isoforms are mostly in the regions of the outer surface. (Figure 2) Being involved in many cellular
processes, it is likely that 14-3-3 protein binding is related to many diseases as well. This makes
modulation of the binding by small molecules interesting. Depending on the cellular function, it might
not only be inhibition of 14-3-3 binding, but also stabilization that can be considerably interesting, for
example to attenuate signaling cascades by sequestering proteins to the cytoplasm.
The 14-3-3 proteins do not display enzymatic activity by themselves. They function as adaptor
proteins and one of their characteristics is the interaction with other proteins thus influencing the
cellular processes involved. Partner proteins are for example enzymes, signal transduction proteins,
transcription factors or tumor suppressor proteins [11].
Based on the structural knowledge of 14-3-3 proteins and the characteristics of the binding
mechanisms of the binding motifs, the adapter function of 14-3-3 proteins can be modified in either
way.
As the 14-3-3 proteins can undergo such diverse interactions, also many approaches exist to influence
the interaction of 14-3-3 to its partners, with both, inhibition and stabilization being realized [19] [20].
As diverse as the interaction is also the scope of molecules that do the job: small molecules, peptides,
natural products, hybrid molecules.
7
Inhibitors of 14-3-3 interaction
Inhibitors of 14-3-3 interaction are often peptides derived from the pSer or pThr containing the
binding motif sequence of partner proteins of 14-3-3 . These peptides are not cell permeable and not
stable due to hydrolysis of the phosphate ester. These disadvantages should be overcome to generate
valuable tool compounds for 14-3-3 interactome research.
Peptides binding in the amphipathic groove.
Most 14-3-3 binding motifs carry a phosphorylated serine or threonine. Therefore, most 14-3-3
binding peptides contain a phosphorylated residue with other amino acids decorated around it.
Interestingly, some proteins bind 14-3-3 but are not phosphorylated. One of these is exoenzyme S
from Pseudomonas aeruginosa (ExoS).[21][22] ExoS binds the amphipathic groove, but not in the
phosphate binding pocket (Figure 5A). An interesting new development in this area are the very
recently published crosslinked peptides by Grossmanns group.[23]. By stapling short binding
sequences derived from the binding motif of ExoS, binding is improved. The linkers, which were
synthesized through ring closing metathesis and subsequent reduction, were developed to restrict
conformational freedom in the peptide. The crystal structure (see Figure 5C) shows that the
hydrocarbon linker of compound βRS8 (Kd: 0.25 µM) itself is in close contact with 14-3-3 . The
overall most affine binder, βSS12, (Figure 5B) showed a Kd of 41 nM, which is a 28fold increase
compared to the original peptide ESp (Kd: 1.14 µM, all from a fluorescence polarization assay). In a
pull down assay, the peptides displayed inhibition of 14-3-3 interaction with an ExoS fragment and
could therefore be starting points for peptidomimetic drug development.
A 20mer peptide found by phage display, R18, [19] containing the sequence WLDLE is able to
coordinate the same basic cluster of three arginines and a lysine with its two acidic groups as
“normal” binders do with their phosphate group. It binds to the three investigated 14-3-3 isoforms (ζ,
β, τ) with a Kd of 70-90 nM. This peptide could be a starting point for non-phosphate containing
inhibitors of 14-3-3 . Liddington, Fu et al reported the albeit low resolution crystal structures of R18
bound to 14-3-3 . [24] Two R18 peptide linked together with a short peptide spacer generated the
14-3-3 binding peptide difopein (DImeric FOurteen-three-three PEptide INhibitor). As it was shown
before [15] that peptides containing tandem phosphoserine motifs show enhanced binding due to one
peptide binding the 14-3-3 dimer, the Fu group reasoned that linking two R18 peptides should result
in a similarly enhanced potency. The conjugate was expressed in HEK293 cell and led to apoptosis
and enhanced cisplatin toxicity. [25]
Onthewaytosmallmolecules:peptide-smallmoleculehybrids.
An elegant study of Yao et al [26] employed a small molecule microarray library which kept the
phosphorylated serine from the original optimal 14-3-3 binding motif RFRpSYPP but exchanged the
flanking residues systematically with 243 commercially available acids or 50 amines, respectively.
From these investigations, two amines and three acids emerged that bound strongly to 14-3-3 . These
were then combined and the resulting six compounds tested for inhibition of 14-3-3 action in a
fluorescence polarization assay. Their most potent compound, 2-5 (Figure 3) displayed an IC50 value
of 2.6 µM. The original peptide used as positive control showed an IC50 of 0.25 µM. The inhibition
depends strongly on the phosphate, as analogs carrying a hydroxy group instead exhibited no
8
inhibition. The compounds were found to be reasonably cell permeable (Papp 554 nm/s in MDCK
canine kidney cells) and lead to apoptosis in A549 cells as evidenced by caspase 3 activation.
To further improve cell permeability, the group of Borch made a quite elaborate, more lipophilic and
therefore more cell permeable, prodrug of these inhibitors. [27] A 16 step synthesis was developed to
prepare a non-hydrolysable difluoromethylenephosphonate prodrug (15) of the phosphoserine
compound 2-5.[27] In brief, the compound was esterified with a nitrofurane moiety and 4-chloro-Nmethylbutane-1-amine, which are intended to be cleaved reductively and by hydrolysis following
intramolecular cyclization, respectively. The developed methodology may also be used to make other
phosphorylated inhibitors more cell permeable. To prove enhanced cell permeability, cytotoxicity
studies in DG75 leukemia cells were performed. In contrast to 2-5, which showed no inhibition of cell
growth, 15 displayed an IC50 of 5µM. In addition, direct mediation via 14-3-3 was shown in a
luciferase assay with FOXO3a. When FOXO3a is bound to 14-3-3 , it is inactive, hence, no FOXO3a
dependent gene expression can be observed. Addition of 15 (Figure 3) was able to restore FOXO3a
dependent expression of genes in a concentration dependent manner.
An intriguing concept employing chemical ligation was developed by the group of Ohkanda. [28]
Fusicoccin A (Figure 9) was modified with an epoxide and incubated together with peptide QSYpTV
or QSYDC, respectively, and 14-3-3 . In the latter, the phosphorylated threonine was replaced with a
bioisostere, aspartic acid. It was investigated whether 14-3-3 could form a scaffold to generate its own
potent inhibitor in situ. Interestingly, the mode 3 binding motif QSYpTC did not react with FC-A
epoxide in solution, whereas the modified QSYDC did. The authors speculate that the highly charged
phosphate decreases nucleophilicity of the thiol, impeding the reaction with the epoxide. Their
compound 3 increased conjugate generation to 199% in presence of compared to in absence of 14-3-3
.
Smallmolecules:Strugglingtoovercomethephosphatedependency
In a project aimed at finding new strategies against chronic myeloid leukemia (CML), the group of
Botta found BV02 (Figure 3) in a structure-based design approach. [29] [30] The Brc-Abl kinase
inhibitor imatinib is used against CML, but cancer cells develop resistance to it due to point mutations
in Abl. When the interaction of 14-3-3 and Abl is disrupted, it relocates to the nucleus and leads to
apoptosis. The small molecule BV02 displayed toxicity in CD34+ and Ba/F3 cells. The former cell
line shows resistance to imatinib, whereas the latter is transduced to express either the wild type BcrAbl protein or the imatinib resistant T315I variant (LD50 of BV02 in wt and the T315I mutant 1 and
1.5 µM, respectively). Addition of BV02 to cells led to relocation of c-Abl from cytoplasma to the
nucleus and mitochondrial membranes in Ba/F3 cells. C-Abl may only relocate if interaction with
14-3-3 is interrupted. The compound is an interesting starting point for further structure-activity
relationship (SAR) as it carries no phosphate group. In a later publication, two additional structures
found by virtual screening were presented as inhibitors of 14-3-3 action (BV01 and BV101, see
Figure 3).[31]
A novel mode of inhibitory action is displayed by the molecular tweezer [32] shown in Figure 8.
according to the crystallographic data, it wraps itself around Lys214 located at the rim of the
amphipathic groove of 14-3-3 σ. In this way, it is able to inhibit interaction of 14-3-3 with FAM
labeled peptides containing the binding motive of the c-Raf (phosphorylated) and ExoS
(unphosphorylated) with IC50 values of 480 and 520 µM, respectively. Though not a drug candidate
due to its two phosphate groups which lead to at least two negative charges at neutral pH, the tweezer
teaches valuable lessons regarding the rules by which they interfere with 14-3-3 binding: The
9
interaction of tweezer and 14-3-3 is quite specific to Lys214 (out of 17 possible lysines in 14-3-3 σ),
because 1) it is well accessible as it is located in an α-helical part of the protein; 2) other hydrophilic
amino acids are located in the vicinity, so additional H bonds can be established to the phosphate
groups to enhance binding, and 3) no additional positively charged amino acids (Lys or Arg) are close
to the lysine because they would lead to binding of external ion pairs. Of all surface exposed lysins
only Lys214 fulfills all of these requirements. (Figure 8)
A small molecule PPI inhibitor of 14-3-3 was found by Fu et al [33] and aptly named fobisin101
(FOurteen-three-three BInding Small molecule INhibitor, (Figure 6). Fobisin101 blocked the binding
of PRAS40 and ExoS with IC50 ranging from 9-19 µM in COS-7 cell lysates in a concentration
dependent manner. In the crystal structure, only the phosphate part of the compound is found in the
binding groove, having formed a covalent N=N bond to Lys120. This reaction was thought to have
occurred due to x-ray radiation, as the crystal changed color from orange to colorless during data
collection. In addition, gel digestion and MS analysis showed the mass difference for modified lysine
only for samples exposed to x-rays. The authors speculate that radiation triggered drugs could be
developed from fobisin101 for cancer treatment. Alternatively, the observed adduct could be due to
the fact that the pyridoxal phosphates covalently bind to lysines. Incubation of fobisin101 with 14-3-3
and subsequent ESI-MS showed that the 14-3-3 proteins may carry up to four modifications with
fobisin101. [34]
Ottmann et al. reported the discovery of phosphonates as small molecule inhibitors of extracellular
14-3-3 interactions. These phosphonates were found in a virtual screen (A1 and A2; [35] (Figure 7).
Structure activity relationship studies revealed that the phosphonate group is important, but not
sufficient to modulate the binding. A substituent in p-position of the phenyl ring is detrimental to
inhibitory activity. The crystallographic data provides the structural basis for this as such a substituent
can simply not be accommodated in the binding pocket. Also, molecules that do not have a proton at
the amide nitrogen are inactive. This knowledge was then applied and led to the discovery of
compound B1 as the most potent inhibitor of the set (IC50 5 µM against FAM-Raf peptide). The
second ring can be quite diverse, as exemplified in molecules B1 and the second best inhibitor B2
(Figure 7). which carry a 2,3-dichlorophenyl and a cyclohexyl moiety, respectively. An advantage of
the compounds is that they can be used to study the interaction of 14-3-3 with its extracellular
receptor aminopeptidase N exclusively, whereas inhibition of cytosolic 14-3-3 interaction is prevented
due to low membrane penetration of the molecules. In a cell assay, the inhibitors showed substantial
decrease in matrix metalloprotease (MMP) activity.
Lacosamide (Figure 3) is a functionalized amino acid originally marketed against seizures and
neuropathic pain in diabetes, the mechanism of action is unclear. In 2011, Kohn et al described an
AB&CR (affinity bait and chemical reporter) approach to find moderate binders of proteins. They
modified lacosamide with the isothiocyanate AB moiety for establishing a covalent bond to the
receptor protein and a propargyl CR moiety (Figure 3) to undergo click reations with rhodamine-azide
or biotin-azide reporter probes (R)-2. They found that (R)-2 selectively binds to the 14-3-3 ζ isoform
by gel staining with the rhodamine probe. MS measurements indicated that covalently attaches itself
to Lys120 which is located near the amphipathic groove of 14-3-3 . Surprisingly, the interaction of
(R)-2 was dependent on the presence of xanthine in the buffer. ITC measurements were conducted and
it was shown that xanthine interacts with 14-3-3 with a Kd of 29 µM. When lacosamide was also
added, small changes were observed, but they could not confidently be assigned to specific protein
binding. [36]
In a high throughput screening with 58019 compounds, molecules bearing the PRLX24905
thiadiazole scaffold (Figure 3) inhibited interaction of 14-3-3 γ and a phosphorylated HSP20 peptide
10
(6-FAM-WLRRApSAP) in a fluorescence polarization assay.[37] The results for the most efficient
compounds were also verified in cell based assays monitoring cell stiffness and finally ex vivo in
intact trachealis tissue from rats. In the long run, the compounds are envisioned to be used in the
treatment of asthma. Inhibition of pHSP20 interaction with 14-3-3 is linked to relaxation of airway
and vascular smooth muscle by an incompletely understood mechanism.
Naturalproducts:Thesolution?
Recently, three new natural products were isolated from extracts of the stink beetle (Blaps
japanensis). These were named blapsin A, blapsin B and blapsamide. [38] Blapsins A and B were
found to inhibit 14-3-3 γ interaction with a TMR-Raf pS259 peptide in a fluorescence polarization
assay with IC50 values of 2 and 2.5 µM and of 9.2 and 10 µM in an ELISA assay. Importantly, these
compounds do not carry phosphate or phosphonate groups and should be investigated and developed
further.
In their search for inhibitors of cancer cell migration, Imoto et al screened more than 2000 microbial
extracts and found moverastin A, a natural product produced by from Aspergillus sp. F7720 [39]. It
inhibited migration of EC17 cells by blocking farnesyltransferase (FTase) with an IC50 of ~7µg/mL.
FTase farnesylates Ras proteins, upon which they become activated and contribute substantially to
migration of endothelial and fibroblast cells. In a study to optimize the compound, UTKO1 was
synthesized and found to inhibit cancer cell migration more potently than moverastin A, but
surprisingly showed no inhibitory activity of FTase [40]. UTKO1 was thus biotinylated (still
exhibiting the same potency in inhibition of cell migration) and incubated with avidin beads to coprecipitate interacting proteins, two of them were found to be 14-3-3 ε and ζ by LC-MS/MS. A pull
down assay with GST-tagged 14-3-3 ζ then showed interaction of the biotinylated UTKO1 with
14-3-3 ζ. [41]
11
14-3-3PPIstabilizers
The previous section gave a look on the diverse and fast growing field in modulation of protein functions
by PPI inhibition and its promising pharmaceutical applications. However, another way of modulating
PPIs, namely stabilization of the protein complex, is still an underexploited technique in drug discovery.
Nature is providing some excellent examples of PPI stabilizers that rationalize the potential of this
approach for targeted Chemical Biology approaches (e.g. Rapamycin [42], Forskolin [43], Brefeldin A
[44], Auxin [45], Jasmonate-Ile [46] or Ins(1,4,5,6)P4 [47] . However, the dedicated use of PPI
stabilizers in industry as well as in academia is remarkably low.
14-3-3 has a large amount of binding partners in the eukaryotic cell and is associated with a broad scale
of diseases. As discussed in the previous chapter, 14-3-3 is an interesting target for PPI inhibition in drug
discovery. Nonetheless, this protein family is maybe even more interesting for PPI stabilization based
drug discovery. Specificity is a difficulty in designing an active compound for a protein with that many
interaction partners as 14-3-3 . By designing a compound that binds to two protein surfaces at the same
time, PPI stabilizers can be more specific than PPI inhibitors which bind to just one of the proteins. In
addition, a small molecule binds to the protein complex in a non-competitive manner, which decreases
the necessary binding affinity to show effect. A small increase of binding affinity between two proteins by
a small molecule can shift the balance significantly to result in a strong physiological effect.
NaturalproductPPIstabilizers
An excellent example to show the possibilities of stabilizing 14-3-3 protein complexes, is the 14-3-3
/PMA2/Fusicoccin A complex. The plant plasma membrane H+-ATPase (PMA2) is a proton pump
responsible for the generation of an electrochemical force across the plasma membrane that is needed for
nutrient uptake and the maintenance of cell turgor.[48] 14-3-3 interacts with a phosphorylated mode-III
motif (QSYpTV-COOH) of the C-terminal regulatory domain of PMA2 to activate the proton
pump.[49][50] The fungal phytotoxin Fusicoccin A (FC-A) is known to stabilize this interaction,
resulting in a permanently active proton pump, osmotic swelling of the guard cells and full-time opening
of the stomatal pores, causing dehydration and death of the plant.[49] The FC-A molecule fills the cavity
in the interaction interface of 14-3-3 and PMA2, increasing the binding affinity significantly.[51][52] The
crystal structure of 14-3-3 in complex with the binding motif of PMA2 and stabilized by FC-A (Figure 9
and Figure 10) shows that FC-A interacts with 14-3-3 and the PMA2 binding motif simultaneously. The
glycosidic part of FC-A is solvent exposed and forms two hydrogen bonds (to Asn49 and Asp222) and
hydrophobic interactions with 14-3-3 . The diterpene part of FC-A is directed in the cavity and makes
two additional hydrogen bonds (to Asp222 and Lys129) and hydrophobic contacts to 14-3-3 . The
interaction between FC-A and the peptide involves the C-terminal valine of the peptide and the five- and
eight–membered carbocycles of FC-A, creating an extra hydrophobic pocket for the PMA2 peptide. The
addition of FC-A creates a ~90-fold increase of binding affinity of the PMA2 peptide to 14-3-3 .[51]
Although FC-A is originally targeting a plant protein, analyzes of its influence on 14-3-3 binding targets
showed that the use of the fungal toxin FC-A can as well be used for investigation of human 14-3-3
binding targets. For example the human protein GPIbα, a subunit of the membrane protein complex of
GPIb-IX-V (glycoprotein Ib-IX-V) which is placed on the platelet plasma membrane and plays a role in
mediating the initial adhesion of circulating platelets to high shear vessel wall matrix. [53] Upon binding
of 14-3-3 to the C-terminus of GPIbα (RYSGH(pS)L-COOH), the complex is able to bind the
subendothelial von-Willebrand-factor’ Stabilization of this interaction by FC-A promotes platelet
12
adhesion to von-Willebrand-factor’, which could lead to platelet spreading and integrin αIIbβ3 activation.
This makes FC-A a good starting point for future therapeutic use in genetic bleeding diseases. [54]
Apart from Fusicoccin FC-A, additional natural 14-3-3 PPI stabilizers have been found in the fusicoccin
family. These other analogs of the fusicoccanes like Fusicoccin J (FC-J), Fusicoccin H (FC-H) and
Cotylenin A (CN-A) are produced by fungi, bacteria, liverwort and higher plants, and have been shown
to have stabilizing properties in different 14-3-3 complexes. They are versatile tools in plant biology and
start to play a role in drug discovery. [55][56][57] (Figure 11 and Figure 12).
Non-naturalproduct14-3-3PPIstabilizers
FC-A is a complex natural product and therefore too expensive for an agrochemical agent, although the
wheat killing character would qualify FC-A as a promising herbicide candidate. Screening for new
herbicides based on the mode of action of the fungal toxin FC-A, but less complex and chemically
tractable, led to the discovery of two structurally unrelated 14-3-3 /PMA2 PPI stabilizers: Epibestatin
and Pyrrolidone1 (Pyr1).[20] Although the increase of binding affinity by the use of these stabilizers is
lower than with FC-A, which is caused by the smaller contact surface with the proteins, these small
molecules showed to be selective among a selection of other 14-3-3 PPIs (Raf1, p53, Cdc25C, RNF11,
Mlf1, AICD, Cby, and YAP) and can be an important starting point in the development of 14-3-3 PPI
stabilization compounds. Both small molecules have a different binding site in the 14-3-3 PMA2
interface (see Figure 13) and different binding kinetics.[20] Pyr1 shares most of its contact surface with
14-3-3 , but its contact interface to PMA2 is rather small. In addition, the compound binds to a site that is
easier accessible from the solvent space, which might account for the faster association and dissociation
kinetics. Epibestatin is more buried into the binding interface and shares a roughly equal contact surface
with 14-3-3 and PMA2, resulting in slower association and dissociation kinetics.[20]
After the discovery of the stabilizing effect of Pyr1 on the 14-3-3 /PMA2 interaction, different
derivatives of Pyr1 were synthesized in search of an improved stabilizer. Screening of these compounds
showed that three modifications were important for activity enhancement: 1) conversion of the Pyr1
scaffold into a pyrazole, 2) introduction of a tetrazole moiety to the phenyl ring that contacts PMA2 and
3) addition of a bromine to the phenyl ring that contacts 14-3-3 . The combination of these three
modifications resulted in Compound 37 with a threefold increase of binding affinity compared to Pyr1.
The most important reason for binding enhancement is the significantly larger contact surface to PMA2,
visible in the crystal structure of the complex with comparable compound 34 (Figure 14). A specificity
experiment with two other important 14-3-3 binding partners (C-RAF and p53) showed that compound
37, like Pyr1, is specific for 14-3-3 /PMA2 stabilization.[58] The different binding characteristics, the
selectivity and the ability to modify the binding affinity of 14-3-3 PPI stabilizers shows the variety of
possibilities in the design of new compounds in, for example, 14-3-3 PPI stabilizer based drug discovery.
PPIstabilizersindrugdiscovery
The 14-3-3 /PMA2/FC-A complex shows a good proof a principle for the use of PPI stabilization to be
applied in the field of drug discovery. This knowledge was then applied in the TWIK-related acidsensitive potassium channels (TASKs)/14-3-3 complex. TASK(1-3) are two-pore domain channels that
are known to be involved in a various of diseases as cancer development, inflammation, ischemia and
epilepsy. [59] 14-3-3 proteins play a role in the expression of the human potassium channel TASK-3 by
masking the basic endoplasmic reticulum retention signal and allowing transport to the plasma
13
membrane, resulting in an increased density of these ion channels in the plasma membrane.[59] The
interaction between 14-3-3 and the TASK-1 and TASK-3 channel is analogous to the 14-3-3 /PMA2
interaction, via a c-terminal mode-III motif (KRRK(pS)V).[13] Stabilization of this interaction would be
a valuable chemical biology tool to explore the possibility of therapeutic interventions targeting these
channels. It has been shown that FC-A can enhance the interaction between 14-3-3 and the TASK-3
binding motif approximately 30-fold. [55] In addition to this, other natural fusicoccanes (FC-H, 16-OMe-FC-H and aglycones of FC-A and FC-J) were tested for their stabilization capabilities and showed
the ability to bind and stabilize the interaction, although to a lesser extent than FC-A. [55][56] Crystal
structures of these complexes showed that all these fusicoccanes bind to the same interface of the
complex but are not specific for the 14-3-3 /TASK-3 complex. [55] To create a more exclusive stabilizer
for the mode-III binding partners of 14-3-3 , a five-membered tetrahydrofuran ring was constructed
utilizing the 12-hydroxyl group of FC-J, resulting in the semisynthetic Fusicoccin THF (FC-THF)
(Error! Reference source not found.). This group produces steric clashes with any residue at position
+2 of the phosphorylated serine, and makes this stabilizer therefore specific for C-terminal mode-III
binding motifs which lack the presence of residues at position +2. Binding experiments showed that the
interaction between 14-3-3 and TASK-3 increased 21-fold by addition of FC-THF and the compound
showed to be specific for mode-III binding motifs. Surface expression of Xenopus laevis oocytes showed
an significant increase in TASK-1 and TASK-3 channels in the cell membrane after incubation with FCTHF. [55]
Another drug target related to 14-3-3 PPI stabilization is the proto-oncogene C-RAF. C-RAF is part of the
ERK1/2 pathway that functions downstream of the RAS subfamily. One third of all human cancers harbor
mutations in the RAS gene, leading to abnormal activation of downstream RAF signaling pathways and
extensive proliferative signals. C-RAF activity is tightly regulated by serine phosphorylation and binding
with 14-3-3 . For this interaction, three binding sites have been identified. Two of them (pS233 and
pS259) are playing a inhibitory role while one (pS621) is reported to stimulate the C-RAF
pathway.[60,61] Therefore, the interaction between 14-3-3 and the pS233/pS259 binding site should be
stabilized, but not the interaction between 14-3-3 and the pS621 binding site, to prevent excessive
proliferation. The most important difference between these sites is the polar contacts between the side
chain of E622 of C-RAF and K49 and K122 of 14-3-3 . The ring system of the natural compound CN-A
is in severe steric and electrostatic conflict with the side chain of E622, but the molecule is able to
stabilize the C-RAFpS233pS259/14-3-3 /CN-A complex (Figure 11). Although CN-A showed to be
inactive in RAS mutant cancer models, the combination of CN-A and anti-EGFR antibody cetuximab
showed to be highly cytotoxic and tumor growth suppressing while the same dose of cetuximab alone
was not effective. This shows that stabilization of the 14-3-3 /C-RAF complex can be a new possible
therapeutic strategy for drug design for RAS mutant cancers. [57]
Chemicalinduceddimerizationbya14-3-3PPIstabilizer
In addition, PPI stabilizers can function as tool compound in Chemical Biology. For example, FC-A was
used to dimerize 14-3-3 fused target proteins with proteins tagged to the C-terminus of PMA2 in the
living cell.[62] The interaction surface was engineered to prevent binding to endogenous cell proteins and
facilitate FC-A-induced dimerization exclusively between the introduced protein constructs. The
advantage of this system is that FC-A can only bind to the 14-3-3 PMA complex but not to any of these
proteins separately.[62] This concept can in principle be used to induce dimerization of any pair of
proteins and was shown to succeed in dimerization of transcription factor NF-κB (fused with 14-3-3 ) and
a nucleus localization sequence (tagged with the C-terminus of PMA2). Expression of these constructs in
14
the cell and addition of FC-A led to nuclear accumulation of NF-κB and an increase in NF-κB targeted
IL-8 transcription.[62]
15
Differentapproachesleadtogoals
On the way to develop inhibitors for 14-3-3 protein-protein interactions, a range of techniques was used,
among them, state-of-the-art high-throughput screening assays and computational methods. Finally, most
of the approaches have in common to move away from peptides, which were among the first inhibitors
found. Overcoming the phosphorylation is one of the goals too, as this is characteristic for the binding of
14-3-3 to its partners. Overall, only a few small-molecule inhibitors of 14-3-3 are known. This raises the
question if it could be worth taking alternative approaches into account. A concept titled as
“underestimated in drug discovery” might lead to new insights and offer an alternative base for lead
discovery.
By having a look at examples of protein-protein modulation that are given by nature and thus
evolutionary developed over millions of years, more examples of protein-protein stabilizers are found
compared to protein-protein interaction inhibitors [35].
Phagedisplayforfindingthefirstpeptidebasedinhibitor
One of the first 14-3-3 inhibitors was the peptide R18. The 14-3-3 binding sequence
PHCVPRDLSWLDLEANMCLP was found by phage display and the binding of WLDLE to 14-3-3
verified by protein crystallography [19].
Phage-display offers a possibility to present libraries of sequences on the surface of phages [63]. A library
of random peptides was used. Those had a common sequence motif: a 20-mer (X2-C-X14-C-X2) and a
21-mer (X2-C-X18). All the binding phages were then selected by biopanning with GST-14-3-3 .
According to their study, also other peptides that have known binding motives of 14-3-3 might be good
inhibitors. Those are blocking the binding site and the amphipathic binding groove, by mimicking the
phosphorylation and thus inhibit the binding to other interaction partners [19,24].
But still this is a peptide inhibitor of 14-3-3 interactions, not a small-molecule. As described before,
peptide small-molecule hybrids were another solution on the way to small molecule protein-protein
inhibitors of 14-3-3 .
The peptide small molecule hybrid 2-5 was found by using a combination of techniques: the optimal
14-3-3 binding sequence RFRpSYPP and a small-molecule microarray (SMM) [26].
The known 14-3-3 binding peptide was further modified by small molecule fragments. The fragments
were found by an SMM. This approach combined thus the results found by high-throughput screening
with the knowledge about the peptide binding sequence to 14-3-3 .
SMM offers a tool to screen large libraries of molecules for discovery of new substances for
pharmaceutical development. The collection of molecules is immobilized by contact printing in spots on
a functionalized surface. Then a fluorescently labeled protein of interest are probed against the molecules.
This technique offers highly parallel testing of compound libraries [64] and is still widely used in ongoing
research.
16
Discoveryofsmall-moleculeinhibitorswithVirtualScreening
Moving away from the peptide inhibitors, or peptide derived inhibitors, recent approaches using virtual
screenings for finding small-molecules have been published. As a basis the well-known structure of the
14-3-3 protein and their corresponding partner molecules are used. Until now (November 2013), more
than 80 crystal structures of 14-3-3 have been deposited in the protein data bank (pdb.org). This gives a
good anchor for in silico approaches for finding inhibitors.
Virtual screening for inhibitors gives valuable tools for identification of lead compounds. The first
non-peptidic inhibitor BV02 was found by in silico virtual screening [29]. Their docking map was based
on a crystal structure of 14-3-3 (PDB ID: 1YWT) and a phosphopeptide with a mode 1 motif. Starting
with 200.000 commercially available compounds of the Asinex Gold collection, the collection was
reduced to 14 final molecules that were used for biological testing. A superimposition of their final
compound with a peptide showed good agreement with the model [30].
Another virtual screening was based on the rationale, that 14-3-3 is binding in a phosphorylation
dependent manner. This gives a starting point to screen substructures containing phosphates and
phosphonates. In a virtual screening based on about 8 Mio compounds of the ZINC all now subset
(release 11), these substructures were filtered and after applying successive filters, 512 compounds were
docked in a structure of 14-3-3 (PDB ID: 3P1N).
Two active inhibitors with a common scaffold were identified. Consequently SAR brought up additional
substances that could be confirmed as active inhibitors [35]. Those Hits could be verified as well in
biophysical test as by protein crystallography (Figure 7).
IdentifyingachemistrystartingpointbyHTS
Despite in silico methods becoming more and more valuable tools in chemical biology and medicinal
chemistry, it is still one of the most used approaches to screen molecule libraries in high-throughput to
find new leads. There is no exception for 14-3-3 : An inhibitor for 14-3-3 was the phosphonate molecule
Fobisin101 [33].
Besides this successful HTS using a fluorescence polarization assay technique, the same group validated
a time-resolved fluorescence resonance energy transfer (TR-FRET) to find further inhibitors of the 14-3-3
interaction. Whereas their first screening campaign was based on the interaction of 14-3-3 with
phosphorylated Raf-pS259, the second campaign was based on a Bad-pS136 [65].
The two small-molecule stabilizers Pyr1 and Epibestatin were identified by a high throughput screening
approach with a 37.000 compounds library. Based on a fluorescence readout of a GFP-fused 14-3-3
binding to a PMA2 construct the two molecules Pyr1 and Epibestatin were found, providing small
molecule alternative stabilizers to Fusicoccin A. Those two hits were verified by surface plasmon
resonance and protein crystallography (Figure 13) [20]. In a follow-up project the Pyr1 was further
modified by SAR (Figure 14) [58]
17
Futureperspectiveandexecutivesummary
Until now, around a dozen molecules have been published to be modifiers of 14-3-3 protein-protein
interactions. These modifiers can be either stabilizers or inhibitors of PPI. In contrast to the usual
medicinal chemistry approach to find possible inhibitors of protein-protein interactions, the 14-3-3
/PMA2/Fusicoccin A example shows that in this case there are alternatives to inhibition. For 14-3-3
protein-protein interaction with diverse targets stabilizers have been reported.
Regarding that there are about 200 to 300 interactions partners of 14-3-3 in human cells, there can be
either a rationale for designing both, stabilizers and inhibitors. As the involvement of 14-3-3 in cellular
pathways is very diverse also various approaches exist for the modulation. The first inhibitors were
peptides and modified peptides which were designed to bind to 14-3-3 directly and block the amphipathic
binding channel for other interactions. Other approaches use high-throughput- and virtual- screenings for
finding small-molecules.
To date, most of the molecules, that have been proven to be a 14-3-3 protein-protein inhibitor or stabilizer
have been used as a tool compound for further research. Using these valuable tools, progress could be
made in studying protein interaction networks.
But nevertheless it remains a huge obstacle to evolve from tool compounds to actual drugs, which poses
the next great challenge in 14-3-3 modulation research.
18
References
Papers of special note have been highlighted as:
*of interest
**of considerable interest
1.
Perkins JR, Diboun I, Dessailly BH, Lees JG, Orengo C. Transient protein-protein interactions:
structural, functional, and network properties. Structure. 18(10), 1233–43 (2010).
2.
Kastritis PL, Bonvin AMJJ. On the binding affinity of macromolecular interactions: daring to ask
why proteins interact. J. R. Soc. - Interface. 10(79), 20120835 (2013).
3.
De Las Rivas J, Fontanillo C. Protein-protein interactions essentials: key concepts to building and
analyzing interactome networks. PLoS Comput. Biol. 6(6), e1000807 (2010).
4.
Nooren IM., Thornton JM. Structural Characterisation and Functional Significance of Transient
Protein–Protein Interactions. J. Mol. Biol. 325(5), 991–1018 (2003).
5.
Block P, Weskamp N, Wolf A, Klebe G. Strategies to search and design stabilizers of proteinprotein interactions: a feasibility study. Proteins. 68(1), 170–86 (2007).
6.
Thiel P, Kaiser M, Ottmann C. Small-molecule stabilization of protein-protein interactions: an
underestimated concept in drug discovery? Angew. Chemie Int. Ed. 51(9), 2012–8 (2012).
7.
Molzan M, Schumacher B, Ottmann C, et al. Impaired Binding of 14-3-3 to C-RAF in Noonan
Syndrome Suggests New Approaches in Diseases with Increased Ras Signaling. Mol Cell Biol.
30(19), 4698–4711 (2010).
8.
Schumacher B, Mondry J, Thiel P, Weyand M, Ottmann C. Structure of the p53 C-terminus bound
to 14-3-3: implications for stabilization of the p53 tetramer. FEBS Lett. 584(8), 1443–8 (2010).
9.
Hermeking H, Benzinger A. 14-3-3 proteins in cell cycle regulation. Semin. Cancer Biol. 16(3),
183–92 (2006).
10.
Corradi V, Mancini M, Manetti F, Petta S, Santucci MA, Botta M. Identification of the first nonpeptidic small molecule inhibitor of the c-Abl/14-3-3 protein-protein interactions able to drive
sensitive and Imatinib-resistant leukemia cells to apoptosis. Bioorg. Med. Chem. Lett. 20(20),
6133–7 (2010).
11.
Morrison DK. The 14-3-3 proteins: integrators of diverse signaling cues that impact cell fate and
cancer development. Trends Cell Biol. 19(1), 16–23 (2009).
12.
Moore BW, Perez VJ, Gehring M. Assay and regional distribution of a soluble protein
characteristic of the nervous system. J. Neurochem. 15(4), 265–72 (1968).
**Survey of binding partners of 14-3-3, that gives a comprehensive overview of binding sites and
sequences of binding partners of 14-3-3
13. Johnson C, Crowther S, Stafford MJ, Campbell DG, Toth R, MacKintosh C. Bioinformatic and
experimental survey of 14-3-3-binding sites. Biochem. J. 427(1), 69–78 (2010).
19
14.
Coblitz B, Shikano S, Wu M, et al. C-terminal recognition by 14-3-3 proteins for surface
expression of membrane receptors. J. Biol. Chem. 280(43), 36263–72 (2005).
*Analysis of 14-3-3 binding proteins, confirming the 14-3-3 binding motifs
15.
Yaffe MB, Rittinger K, Volinia S, et al. The Structural Basis for 14-3-3: Phosphopeptide Binding
Specificity. Cell. 91(7), 961–971 (1997).
16.
Muslin AJ, Tanner JW, Allen PM, Shaw AS. Interaction of 14-3-3 with Signaling Proteins Is
Mediated by the Recognition of Phosphoserine. Cell. 84(6), 889–897 (1996).
*First published crystal structure of 14-3-3 at the same time as Xiao et al. 1995
17.
Liu D, Bienkowska J, Petosa C, Collier RJJ, Fu H, Liddington R. Crystal structure of the zeta
isoform of the 14-3-3 protein. Nature. 376(6536), 191–4 (1995).
*First published crystal structure of 14-3-3 at the same time as Liu et al. 1995
18.
Xiao B, Smerdon SJ, Jones DH, et al. Structure of a 14-3-3 protein and implications for
coordination of multiple signalling pathways. Nature. 376(6536), 188–91 (1995).
19.
Wang B, Yang H, Liu Y-C, et al. Isolation of High-Affinity Peptide Antagonists of 14-3-3 Proteins
by Phage Display †. Biochemistry. 38(38), 12499–12504 (1999).
20.
Rose R, Erdmann S, Bovens S, et al. Identification and structure of small-molecule stabilizers of
14-3-3 protein-protein interactions. Angew. Chemie Int. Ed. 49(24), 4129–4132 (2010).
21.
Yang X, Lee WH, Sobott F, et al. Structural basis for protein-protein interactions in the 14-3-3
protein family. Proc. Natl. Acad. Sci. 103(46), 17237–42 (2006).
22.
Ottmann C, Yasmin L, Weyand M, et al. Phosphorylation-independent interaction between 14-3-3
and exoenzyme S: from structure to pathogenesis. EMBO J. 26(3), 902–13 (2007).
23.
Glas A, Bier D, Hahne G, Rademacher C, Ottmann C, Grossmann TN. Constrained Peptides with
Target-Adapted Crosslinks as Inhibitors of a Pathogenic Protein-Protein Interaction. Angew.
Chemie Int. Ed. (2013). DOI: 10.1002/anie.201310082
24.
Petosa C. 14-3-3zeta Binds a Phosphorylated Raf Peptide and an Unphosphorylated Peptide via Its
Conserved Amphipathic Groove. J. Biol. Chem. 273(26), 16305–16310 (1998).
25.
Masters SC, Fu H. 14-3-3 Proteins Mediate an Essential Anti-Apoptotic Signal. J. Biol. Chem.
276(48), 45193–200 (2001).
26.
Wu H, Ge J, Yao SQ. Microarray-assisted high-throughput identification of a cell-permeable
small-molecule binder of 14-3-3 proteins. Angew. Chemie Int. Ed. 49(37), 6528–32 (2010).
27.
Arrendale A, Kim K, Choi JY, Li W, Geahlen RL, Borch RF. Synthesis of a phosphoserine
mimetic prodrug with potent 14-3-3 protein inhibitory activity. Chem. Biol. 19(6), 764–71 (2012).
28.
Maki T, Kawamura A, Kato N, Ohkanda J. Chemical ligation of epoxide-containing fusicoccins
and peptide fragments guided by 14-3-3 protein. Mol. Biosyst. 9(5), 940–3 (2013).
29.
Corradi V, Mancini M, Manetti F, Petta S, Santucci MA, Botta M. Identification of the first nonpeptidic small molecule inhibitor of the c-Abl/14-3-3 protein-protein interactions able to drive
20
sensitive and Imatinib-resistant leukemia cells to apoptosis. Bioorg. Med. Chem. Lett. 20(20),
6133–7 (2010).
30.
Mancini M, Corradi V, Petta S, et al. A new nonpeptidic inhibitor of 14-3-3 induces apoptotic cell
death in chronic myeloid leukemia sensitive or resistant to imatinib. J. Pharmacol. Exp. Ther.
336(3), 596–604 (2011).
31.
Corradi V, Mancini M, Santucci MA, et al. Computational techniques are valuable tools for the
discovery of protein-protein interaction inhibitors: the 14-3-3σ case. Bioorg. Med. Chem. Lett.
21(22), 6867–71 (2011).
32.
Bier D, Rose R, Bravo-Rodriguez K, et al. Molecular tweezers modulate 14-3-3 protein-protein
interactions. Nat. Chem. 5(3), 234–9 (2013).
33.
Zhao J, Du Y, Horton JR, et al. Discovery and structural characterization of a small molecule 143-3 protein-protein interaction inhibitor. Proc. Natl. Acad. Sci. , 1100012108– (2011).
34.
Röglin L, Thiel P, Kohlbacher O, Ottmann C. Covalent attachment of pyridoxal-phosphate
derivatives to 14-3-3 proteins. Proc. Natl. Acad. Sci. 109(18), E1051–3; author reply E1054
(2012).
35.
Thiel P, Röglin L, Meissner N, Hennig S, Kohlbacher O, Ottmann C. Virtual screening and
experimental validation reveal novel small-molecule inhibitors of 14-3-3 protein-protein
interactions. Chem. Commun. 49(76), 8468–70 (2013).
36.
Park KD, Kim D, Reamtong O, et al. Identification of a lacosamide binding protein using an
affinity bait and chemical reporter strategy: 14-3-3 ζ. J. Am. Chem. Soc. 133(29), 11320–30
(2011).
37.
An SS, Askovich PS, Zarembinski TI, et al. A novel small molecule target in human airway
smooth muscle for potential treatment of obstructive lung diseases: a staged high-throughput
biophysical screening. Respir. Res. 12, 8 (2011).
38.
Yan Y-M, Dai H-Q, Du Y, et al. Identification of blapsins A and B as potent small-molecule 14-33 inhibitors from the insect Blaps japanensis. Bioorg. Med. Chem. Lett. 22(12), 4179–81 (2012).
39.
Takemoto Y, Watanabe H, Uchida K. Chemistry and Biology of Moverastins, Inhibitors of Cancer
Cell Migration, Produced by Aspergillus. Chem. Biol. 12(12), 1337–47 (2005).
40.
Sawada M, Kubo S, Matsumura K. Synthesis and anti-migrative evaluation of moverastin
derivatives. Bioorg. Med. Chem. Lett. 21(5), 1385–9 (2011).
41.
Kobayashi H, Ogura Y, Sawada M, et al. Involvement of 14-3-3 proteins in the second epidermal
growth factor-induced wave of Rac1 activation in the process of cell migration. J. Biol. Chem.
286(45), 39259–68 (2011).
42.
Choi J, Chen J, Schreiber S, Clardy J. Structure of the FKBP12-rapamycin complex interacting
with binding domain of human FRAP. Science (80-. ). (1996).
43.
Tesmer JJ. Crystal Structure of the Catalytic Domains of Adenylyl Cyclase in a Complex with
Gs·GTPS. Science (80-. ). 278(5345), 1907–1916 (1997).
21
44.
Renault L, Guibert B, Cherfils J. Structural snapshots of the mechanism and inhibition of a
guanine nucleotide exchange factor. Nature. 426(6966), 525–30 (2003).
45.
Tan X, Calderon-Villalobos LI a, Sharon M, et al. Mechanism of auxin perception by the TIR1
ubiquitin ligase. Nature. 446(7136), 640–5 (2007).
46.
Sheard LB, Tan X, Mao H, et al. Jasmonate perception by inositol-phosphate-potentiated COI1JAZ co-receptor. Nature. 468(7322), 400–5 (2010).
47.
Watson PJ, Fairall L, Santos GM, Schwabe JWR. Structure of HDAC3 bound to co-repressor and
inositol tetraphosphate. Nature. 481(7381), 335–40 (2012).
48.
Morsomme P, Boutry M. The plant plasma membrane H+-ATPase: structure, function and
regulation. Biochim. Biophys. Acta - Biomembr. 1465(1-2), 1–16 (2000).
49.
Jahn T, Fuglsang a T, Olsson A, et al. The 14-3-3 protein interacts directly with the C-terminal
region of the plant plasma membrane H(+)-ATPase. Plant Cell. 9(10), 1805–14 (1997).
50.
Maudoux O, Batoko H, Oecking C, et al. A plant plasma membrane H+-ATPase expressed in
yeast is activated by phosphorylation at its penultimate residue and binding of 14-3-3 regulatory
proteins in the absence of fusicoccin. J. Biol. Chem. 275(23), 17762–70 (2000).
51.
Würtele M, Jelich-Ottmann C, Wittinghofer A, Oecking C. Structural view of a fungal toxin acting
on a 14-3-3 regulatory complex. EMBO J. 22(5), 987–94 (2003).
52.
Ottmann C, Marco S, Jaspert N, et al. Structure of a 14-3-3 coordinated hexamer of the plant
plasma membrane H+ -ATPase by combining X-ray crystallography and electron cryomicroscopy.
Mol. Cell. 25(3), 427–40 (2007).
53.
Varga-Szabo D, Pleines I, Nieswandt B. Cell adhesion mechanisms in platelets. Arterioscler.
Thromb. Vasc. Biol. 28(3), 403–12 (2008).
54.
Camoni L, Di Lucente C, Visconti S, Aducci P. The phytotoxin fusicoccin promotes platelet
aggregation via 14-3-3-glycoprotein Ib-IX-V interaction. Biochem. J. 436(2), 429–36 (2011).
55.
Anders C, Higuchi Y, Koschinsky K, et al. A semisynthetic fusicoccane stabilizes a proteinprotein interaction and enhances the expression of K+ channels at the cell surface. Chem. Biol.
20(4), 583–93 (2013).
56.
De Boer AH, de Vries-van Leeuwen IJ. Fusicoccanes: diterpenes with surprising biological
functions. Trends Plant Sci. 17(6), 360–8 (2012).
57.
Molzan M, Kasper S, Röglin L, et al. Stabilization of physical RAF/14-3-3 interaction by
cotylenin A as treatment strategy for RAS mutant cancers. ACS Chem. Biol. 8(9), 1869–75 (2013).
58.
Richter A, Rose R, Hedberg C, Waldmann H, Ottmann C. An optimised small-molecule stabiliser
of the 14-3-3-PMA2 protein-protein interaction. Chem. - A Eur. J. 18(21), 6520–7 (2012).
59.
Bittner S, Budde T, Wiendl H, Meuth SG. From the background to the spotlight: TASK channels
in pathological conditions. Brain Pathol. 20(6), 999–1009 (2010).
22
60.
Tzivion G, Luo Z, Avruch J. A dimeric 14-3-3 protein is an essential cofactor for Raf kinase
activity. Nature. 394(6688), 88–92 (1998).
61.
Dumaz N, Marais R. Protein kinase A blocks Raf-1 activity by stimulating 14-3-3 binding and
blocking Raf-1 interaction with Ras. J. Biol. Chem. 278(32), 29819–23 (2003).
62.
Skwarczynska M, Molzan M, Ottmann C. Activation of NF-κB signalling by fusicoccin-induced
dimerization. Proc. Natl. Acad. Sci. 110(5), E377–86 (2013).
63.
Abelson JN, editor. Combinatorial Chemistry. In: Methods in Enzymology. Elsevier Inc., 3–493
(1996).
64.
MacBeath G, Koehler AN, Schreiber SL. Printing Small Molecules as Microarrays and Detecting
Protein−Ligand Interactions en Masse. J. Am. Chem. Soc. 121(34), 7967–7968 (1999).
65.
Du Y, Fu RW, Lou B, et al. A time-resolved fluorescence resonance energy transfer assay for highthroughput screening of 14-3-3 protein-protein interaction inhibitors. Assay Drug Dev. Technol.
11(6), 367–81 (2013).
66.
Sievers F, Wilm A, Dineen D, et al. Fast, scalable generation of high-quality protein multiple
sequence alignments using Clustal Omega. Mol. Syst. Biol. 7, 539 (2011).
67.
Ashkenazy H, Erez E, Martz E, Pupko T, Ben-Tal N. ConSurf 2010: calculating evolutionary
conservation in sequence and structure of proteins and nucleic acids. Nucleic Acids Res. 38(Web
Server issue), 529–533 (2010).
68.
The UniProt Consortium. Update on activities at the Universal Protein Resource (UniProt) in
2013. Nucleic Acids Res. 41(Database issue), 43–47 (2013).
69.
Zhao J, Du Y, Horton JR, et al. Discovery and structural characterization of a small molecule 143-3 protein-protein interaction inhibitor. Proc. Natl. Acad. Sci. 108(39), 16212–6 (2011).
23
Figures
Figure 1: Publication entries in PubMed of the last 20 years on “inhibitors of protein protein interaction” and on “14-3-3 ”
(until Dec 2013). Publications on PPI inhibitors started to appear in the 60s, trend is rising still, though including the
keyword 14-3-3 to the search gave only 144 hits in total. Although 14-3-3 proteins were first reported by Moore and Perez in
1968, [12] the publications on 14-3-3 are constantly accelerating since the middle of the 90s, when the first crystallographic
structures were published. (Figure made using GraphPad Prism version 6.02 for Windows, GraphPad Software, San Diego
California USA, www.graphpad.com)
24
Figure 2: Conserved residues of 14-3-3 : Amino acid sequence alignment of the seven human isoforms of 14-3-3 performed
with ClustalOmega. (Sievers et al. 2011) Residue conservation was determined with ConSurf [67]
Corresponding protein entries (UniProt): 14-3-3 ε-P62258, 14-3-3 σ-P31947, 14-3-3 η-Q04917, 14-3-3 γ-P61981, 14-3-3 τP27348, 14-3-3 β-P31946, 14-3-3 ζ-P63104 [68]
25
Table 1: Crystal structures of known 14-3-3 PPI modulators
PDB ID
Small molecule ligand
Peptide ligand
Reference
14-3-3 PPI inhibitors
1A38
R18
Petosa et al., 1998 [24]; Wang et al.,
1999 [19]
2C23
ExoS
Yang et al., 2006 [21]
2O02
ExoS 416- 430
Ottmann et al., 2007 [22]
3RDH
Fobisin101
Zhao et al. 2011 [69]
4HQW
Molecular Tweezers
Bier et al. 2013 [32]
4DHV,
4DHT,
3T0L,
3T0M
Phosphonate inhibitors:
Thiel et al. 2013 [35]
B1, B2, A1, B2
4N7G,
4N7Y,
4N84
ExoS-WT,
ExoS-βRS8,
ExoS-βSS12
Glas et al. 2013 [23]
14-3-3 PPI stabilizers
1O9F
Fusicoccin
PMA2
Würtele et al. 2003 [51]
3M51,
3M50
Pyrrolidone1,
Epibestatin
PMA2-CT30
Rose et al., 2010 [20]
4DX0
Compound 34
PMA2-CT30
Richter et al. 2012 [58]
4IHL
Cotylenin
Raf 229-264 pS259
Molzan et al., 2013 [57]
3SPR
FC-THF
TASK3 370-374 pS373
Anders et al., 2013 [55]
26
Figure 3: Overview of known small molecule inhibitors of 14-3-3
27
Figure 4: The phosphorylation independent peptidic inhibitor of 14-3-3 : R18 (chocolate colored surface), the aspartate of the
short sequence WLDLE is coordinated into the amphipathic groove of 14-3-3 ζ (gray surface). Binding pocket of 14-3-3 ζ
(K49, R56, R127, Y128) surface colored by element green C, gray H, blue N, red O [24] (PDB ID: 1A38
Figure 5: Peptide staples (lightgreen sticks) improving the binding of the wild type sequence of ExoS to 14-3-3 : Wild type
sequence peptide of ExoS: ESp (lightorange surface) and the binding enhancing stapled peptides: βSS12 staple (sulfur yellow
colored surface), βRS8 staple (brightorange colored surface). Binding pocket of 14-3-3 ζ (K49, R56, R127, Y128) surface
colored by element green C, gray H, blue N, red O [23] (PDB ID: 4N7G, 4N7Y, 4N84)
28
Figure 6: The FOurteen three three BInding INhibitor 101 (FOBISIN101, (skyblue sticks) bound to Lys120 (element
colored sticks/surface) of 14-3-3 (gray surface). Binding pocket of 14-3-3 ζ (K49, R56, R127, Y128) surface colored by
element green C, gray H, blue N, red O [69] (PDB ID: 3RDH)
29
Figure 7: Phosphonate inhibitors A) overlay of all hits which shows that the phosphate moiety coordinates the orientation of
the molecules into the pocket B) A1 (lightblue), C) A2 (paleyellow) with alternative conformation in crystal structure, D) B1
wheat colored, E) B2 palegreen binding to 14-3-3 (gray surface) found in virtual screening: Binding pocket of 14-3-3 σ
(K49, R56, R129, Y130) surface colored by element green C, gray H, blue N, red O [35] (PDB ID: 4DHV, 4DHT, 3T0L,
3T0M)
30
Figure 8: Molecular tweezer (cyan colored sticks) binding to Lys214 (green sticks/light gray surface) of 14-3-3 σ (gray
surface) and blocking the amphipathic groove. Binding pocket of 14-3-3 σ (K49, R56, R129, Y130) surface colored by
element green C, gray H, blue N, red O [32] (PDB ID: 4HQW)
31
Figure 9: Overview of known small molecule stabilizers of 14 3 3 interactions
32
Figure 10: Fusiccocin A (smudge colored sticks), the first known natural product stabilizer of the 14-3-3 interaction with the
PMA2 (tv_blue colored surface). Binding pocket of T14-3-3 c (K56, R63, R136, Y137) surface colored by element green C,
gray H, blue N, red O [51] (PDB ID: 1O9F)
Figure 11: Peptide sequence of Raf (yelloworange surface) binding to 14-3-3 (gray surface) stabilized by Cotylenin A
(deepteal sticks): The binding of the peptide to 14-3-3 ζ is coordinated via its phosphorylated serine. Binding pocket of
14-3-3 ζ (K49, R56, R127, Y128) surface colored by element: green C, gray H, blue N, red O [57] (PDB ID: 4IHL)
33
Figure 12: FC-THF (black sticks), a modified Fusicoccane, to exclusively stabilize mode-3 interactions binding to 14-3-3
and the phosphopeptide TASK3 (sand colored surface) . Binding pocket of 14-3-3 σ (K49, R56, R129, Y130) surface colored
by element: green C, gray H, blue N, red O [55] (PDB ID: 3SPR)
34
Figure 13: A) Pyrrolidone1 (yellow) and B) Epibestatin (orange): First small-molecule non-natural product stabilizers of
T14-3-3 c (gray surface) and the C-terminal PMA2-CT30 (tv_blue colored surface). Binding pocket of T14-3-3 c (K56, R63,
R136, Y137) surface colored by element: green C, gray H, blue N, red O [20] (PDB ID: 3M51 and 3M50)
35
Figure 14: Optimized stabilizer molecule Compound 34 (olive colored sticks) for the interaction of PMA-CT30 (tv_blue
colored surface) and T14-3-3 e (gray surface). Binding pocket of T14-3-3 e (K54, R61, R134, Y135) surface colored by
element green C, gray H, blue N, red O [58] (PDB ID: 4DX0)
36