Rapid identification of the pharmacophore in a peptoid inhibitor of the proteasome regulatory particle

Article abstract of DOI:10.1039/b717861a

Here we report a simple and effective method to identify the minimal pharmacophore in the first peptoid inhibitor of the 19S proteasome regulatory particle, which has led to the development of a derivative that exhibits improved cellular activity, presumably due to a reduction in mass of about two-fold and the elimination of positively charged lysine-like residues. The Royal Society of Chemistry.

Full text of DOI:10.1039/b717861a

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Rapid identification of the pharmacophore in a peptoid inhibitor of the
proteasome regulatory particlew
Hyun-Suk Lim, Chase T. Archer, Young-Chan Kim, Troy Hutchens and
Thomas Kodadek*
Received (in Austin, TX, USA) 19th November 2007, Accepted 3rd January 2008
First published as an Advance Article on the web 24th January 2008
DOI: 10.1039/b717861a
Here we report a simple and effective method to identify the
minimal pharmacophore in the first peptoid inhibitor of the 19S
proteasome regulatory particle, which has led to the development
of a derivative that exhibits improved cellular activity, presum-
ably due to a reduction in mass of about two-fold and the
elimination of positively charged lysine-like residues.
develop a facile and rapid ‘glycine scanning’ method to
identify the minimal pharmacophore in the molecule.
The data derived from this approach facilitated the devel-
opment of a ‘minimal’ RIP-1 derivative with a molecular mass
that is less than half that of RIP-1 and which exhibits E5-fold
greater potency as a proteasome inhibitor in cultured cells,
presumably due to enhanced cell permeability.
To identify residues in RIP-1 critical for activity, a series of
derivatives was synthesized in which various parts of the mole-
cule were deleted and each derivative was tested for activity. The
assay employed in these experiments was to monitor the effect of
the peptoid on 26S proteasome-mediated peptidolysis. Specifi-
cally, hydrolysis of a fluorogenic peptide, N-succinyl-Leu-Leu-
Val-Tyr 7-amino-4-methylcoumarin (Suc-LLVY-amc), can be
monitored conveniently via a spectroscopic assay9 and is thus
well-suited to evaluate the activity of significant numbers of
compounds. In order for this reaction to occur, a ‘flap’, com-
prised of the N-terminal tails of the 20S a proteins, that
otherwise blocks substrate access to the internal cavity of the
proteasome (Fig. 2A), must be held open by the ATPases of the
19S RP.10 This activity of the ATPases does not require ATP
hydrolysis, in contrast to their protein unfolding activity.
The 26S proteasome is a 2.5 MDa multi-catalytic protease
complex that degrades polyubiquitinylated and damaged pro-
teins.1 It is composed of a barrel-like catalytic 20S core particle
(CP) that contains four stacked hetero-heptameric rings, two a
and two b (Fig. 1).2 The openings at the top and bottom of the
barrel are small enough that folded proteins are precluded
sterically from accessing the internal proteolytic active sites.
Substrate introduction into the catalytic maw of the 20S CP is
achieved through the activity of a 19S regulatory particle (RP)
that caps each end of the 20S CP. The 19S RP contains a
putative hetero-hexameric ring of ATPases (Rpt1-6) as well as
several other proteins (Fig. 1). It binds polyubiquitinylated
proteins, unfolds them, and feeds the resultant chain into the
interior of the barrel where the proteins are degraded into
small peptides.3
There is considerable interest in pharmacological inhibitors
of proteasome function both as research tools4 and as ther-
apeutic agents.5 However, while several compounds are
known that target the proteolytic active sites in the 20S core,
there were no chemical inhibitors of the 19S RP until our
recent report6 of the isolation of such a compound from a
combinatorial library of peptoids (N-substituted oligogly-
cines).7 This compound, which we called Regulatory Parti-
cle-Inhibiting Peptoid-1 (RIP-1) (Fig. 1) was shown to target
one of the six ATPases in the 19S RP, Rpt4,8 and blocks 19S
RP-mediated protein unfolding in vitro with an IC50 of
E3 mM.6 RIP-1 also inhibits proteasome-mediated turnover
of proteins in living cells, but with an IC50 of E50 mM. This
differential between the in vitro and in vivo activity of the
compound may reflect modest cell permeability. In this report,
we take advantage of the modular structure of peptoids to
Fig. 1 RIP-1 is an inhibitor of the 26S proteasome that targets Rpt4.
The ultrastructure of the 26S proteasome is shown at the top of the
figure. It is comprised of a barrel-shaped central 20S core particle (CP)
(gray) capped on each end with a 19S regulatory particle (RP) (pink).
The 19S RP is, in turn, comprised of a lid sub-unit and a base sub-
complex that includes six AAA class ATPases,14 including Rpt4, the
target of RIP-1.8 The ATPases are responsible for ATP-dependent
unfolding of substrate proteins.
Division of Translational Research, Departments of Internal Medicine
and Molecular Biology, University of Texas Southwestern Medical
Center, 5323 Harry Hines Boulevard, Dallas, Texas 75390-9185,
USA. E-mail: thomas.kodadek@utsouthwestern.edu;
Fax: +1-214-648-4156; Tel: +1-214-648-1239
w Electronic supplementary information (ESI) available: Experimental
details for synthesis and biological assays; structures of glycine-
scanned compounds and C2. See DOI: 10.1039/b717861a
ꢀc
This journal is The Royal Society of Chemistry 2008
1064 | Chem. Commun., 2008, 1064–1066

View Article Online
Fig. 3 Identification of the pharmacophore of RIP-1. (A) Structure
of RIP-1 indicating the pieces that were eliminated in the various
derivatives analyzed. ICT ¼ invariant C-terminus; Pur ¼ purine. In
each derivative, the atoms included in the yellow region were replaced
with hydrogen. (B) Activities of the various derivatives in the pepti-
dolysis assay.
Fig. 2 RIP-1 stimulates 26S proteasome-mediated peptidolysis. (A)
Hydrolysis of an unfolded peptide substrate by the 26S proteasome is
inhibited by a ‘flap’ structure comprised of the N-terminal tails of the
20S CP a sub-units that occlude the opening into the catalytic interior
of the complex (right). The ATPases of the 19S CP (not shown here)
must hold these peptides in an ‘open’ conformation in order for the
substrate to diffuse into the barrel. (B) Effect of RIP-1, MG132 and a
control peptoid, C2, on this activity. Vehicle denotes the solvent in
which the peptoids were dissolved.
tein–protein interactions or other functions, except that the
‘point mutants’ are made synthetically rather than being
encoded at the DNA level. Only the Nlys residue near the
N-terminal end of the molecule (R5 in Fig. 3) was found to be
dispensable for activity. Substitution of any of the other four
groups with hydrogen essentially abolished the activity of the
molecule.
As expected, MG132, a peptide aldehyde inhibitor of the
chymotrypsin-like activity of the proteasome, inhibits pepti-
dolysis. In contrast, RIP-1 stimulates the peptidase activity of
the 26S proteasome with an EC50 of E 1 mM, whereas a
control peptoid, C2, not selected to bind the proteasome had
no effect (Fig. 2B). As will be discussed in a full paper on the
mechanism of action of RIP-1 (H.-S.L. and T.K., in prepara-
tion), we believe that this stimulation is a result of the peptoid
‘freezing’ Rpt4 in a conformation suitable for flap opening.
The EC50 of RIP-1 in this peptidolysis assay was essentially
identical to the IC50 of the peptoid in blocking 19S RP-
mediated protein unfolding,6 arguing that the same peptoid–
Rpt4 binding event is responsible for both activities.
The data shown in Fig. 3 led us to hypothesize that the
minimal pharmacophore in RIP-1 is the tetrapeptoid shown in
Fig. 4A. To test this idea, this molecule, called RIP-1.4, was
synthesized and its activity in several biochemical assays was
analyzed. As shown in Fig. 4B, RIP-1.4 stimulated 26S
proteasome-mediated peptidolysis (EC50 E5 mM), though
about 5-fold less potently than RIP-1. We then carried out
an in vitro assay for proteasome-mediated protein unfolding
developed in our laboratory.11 This involves exposure of an
immobilized Gal4-VP16ꢁDNA complex to the proteasome.
Binding of the 19S ATPases to the VP16 domain12 of this
chimeric protein allows the ATPases to engage Gal4-VP16 as a
substrate for ATP-dependent unfolding, which is scored via
monitoring the time-dependent loss of the protein from the
DNA.11 As shown in Fig. 4C, RIP-1.4 inhibited this unfolding
process in a dose-dependent fashion with an IC50 of E5–10
mM, a potency that is again 30–50% that exhibited by RIP-1.6
We conclude from these data that RIP-1.4 indeed represents
the minimal pharmacophore of RIP-1, with in vitro activities
no worse than 2- to 3-fold off those of the parent compound.
Since RIP-1.4 is less than half the mass of RIP-1 (604 vs.
1347 Da) and lacks the positively charged Nlys residues
present in RIP-1, we speculated that it might be significantly
more cell permeable than the parent compound. To test this
point, the ability of RIP-1.4 to inhibit both p27 and p53
Fig. 3A shows schematically the deletion strategy employed
to identify the minimal pharmacophore of RIP-1. First, a
derivative lacking the two lysine-like C-terminal residues
(Nlys) was synthesized since these residues were not rando-
mized in the library from which RIP-1 was isolated.6 As shown
in Fig. 3B, this had little effect on the peptidolysis stimulation
activity. We then created a derivative lacking the purine-
containing appendage. Surprisingly, this compound was only
slightly less potent than RIP-1, showing that the nucleoside
contributes only minimally to the activity of the molecule.
Finally, in the context of the molecule lacking the two
C-terminal Nlys residues, five additional derivatives were
synthesized in which each side chain was, in turn, replaced
with a hydrogen atom. This ‘glycine scanning’ approach is
similar to the alanine scanning technology often used in
biochemistry to identify amino acids important for pro-
ꢀc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 1064–1066 | 1065

View Article Online
RIP-1 on proteasome-mediated p27 turnover. As shown in
Fig. 4D (white bars), azide-RIP-1 is almost as potent as RIP-
1.4,13 indicating that the elimination of charge is the dominant
contributor to increased permeability,14 though the reduced
mass probably contributes somewhat as well.
This study highlights the simplicity with which structure–ac-
tivity relationships (SARs) can be probed using peptoid lead
compounds. The modular structure and simple ‘sub-mono-
mer’ synthesis of peptoids15 lend themselves to the rapid
synthesis of a modest number of derivatives of a lead com-
pound that allow for the rapid identification of the pharma-
cophore. In this case, this exercise suggested that the core
pharmacophore of RIP-1 is a simple tetrapeptoid lacking the
two C-terminal Nlys residues, the N-terminal Nlys residue
and, surprisingly, the purine moiety that was employed to cap
all of the molecules in the library.6 Resynthesis of this much
smaller and uncharged compound, called RIP-1.4, revealed
that, indeed, it is nearly as potent as RIP-1 in vitro in the
stimulation of 26S proteasome-mediated peptidolysis and the
inhibition of protein unfolding (Fig. 3B and C) and is slightly
more potent than RIP-1 in blocking proteasome-mediated
protein turnover in living cells.
The next step in this work will be to create large numbers of
derivatives of the minimal pharmacophore RIP-1.4 and to
assay this new library for compounds with significantly in-
creased potency. The discovery that this compound stimulates
the peptidolysis activity of the 26S proteasome, an assay that is
readily adaptable to high-throughput screening, should facil-
itate these efforts.
This work was funded by grants from the NIH (GM 71833),
the Welch Foundation (I-1299) and a contract from the
National Heart, Lung and Blood Institute for the UT South-
western Center for Proteomics Research (NO1-HV-28185).
Fig. 4 A biologically active ‘minimal’ RIP-1 derivative. (A) Chemical
structure of RIP-1.4, the proposed minimal pharmacophore of RIP-1.
(B) Dose-dependent stimulation of 26S proteasome-mediated peptido-
lysis by RIP-1.4. (C) Dose-dependent inhibition of ATP-dependent
unfolding of Gal4-VP16 by RIP-1.4.11 (D) Dose-dependent inhibition
of p27 (black bars) and p53 (gray bars) in C2C12 myoblasts by RIP-1.4.
White bars indicate dose-dependent inhibition of p27 by azide-RIP-1.
Notes and references
1 G. N. DeMartino and T. G. Gillette, Cell, 2007, 129, 659.
2 J. Lowe, D. Stock, B. Jap, P. Zwickle, W. Baumeister and R.
Huber, Science, 1995, 28, 533.
3 G. N. DeMartino and C. A. Slaughter, J. Biol. Chem., 1999, 274,
22123.
4 A. F. Kisselev and A. L. Goldberg, Chem. Biol., 2001, 8, 739.
5 J. Adams, Trends Mol. Med., 2002, 8, S49.
6 H. S. Lim, C. T. Archer and T. Kodadek, J. Am. Chem. Soc., 2007,
129, 7750.
7 R. J. Simon, R. S. Kania, R. N. Zuckermann, V. D. Huebner, D.
A. Jewell, S. Banville, S. Ng, L. Wang, S. Rosenberg, C. K.
Marlowe, D. C. Spellmeyer, R. Tan, A. D. Frankel, D. V. Santi,
F. E. Cohen and P. A. Bartlett, Proc. Natl. Acad. Sci. U. S. A.,
1992, 89, 9367.
8 H. S. Lim, D. Cai, C. T. Archer and T. Kodadek, J. Am. Chem.
Soc., 2007, 129, 12936.
9 A. F. Kisselev and A. L. Goldberg, Methods Enzymol., 2005, 398,
364.
turnover in cultured C2C12 muscle cells was measured. As
shown in Fig. 4D, a build-up of both proteins was observed as
increasing amounts of the peptoid were incubated with the
cells. The results were identical for both proteins, as would be
expected for a compound that targets the proteasomal AT-
Pases and thus should affect the turnover of all proteins
equally. The IC50 was approximately 10–15 mM, a value about
3- to 5-fold more potent than that measured for RIP-1 in the
same assay.6 Given the fact that RIP-1.4 is slightly more
potent than RIP-1 in vivo, but somewhat less potent in vitro,
we conclude that RIP-1.4 indeed exhibits enhanced perme-
ability, as expected.
10 A. F. Kisselev, D. Kaganovich and A. L. Goldberg, J. Biol. Chem.,
2002, 277, 22260.
11 A. Ferdous, D. Sikder, T. G. Gillette, K. Nalley, T. Kodadek and
S. A. Johnston, Genes Dev., 2007, 20, 112.
To assess the relative importance of mass reduction and
elimination of the charged residues in the increased cell
permeability of RIP-1.4, we synthesized a derivative of RIP-
1 that substitutes neutral azide moieties for the three primary
amines in RIP-1 (for the structure and synthesis, see ESIw) but
has a similar molecular mass. We tested the effect of azide-
12 C. T. Archer, L. Burdine and T. Kodadek, Mol. BioSyst., 2005, 1,
366.
13 At 50 and 100 mM of azide-RIP-1 in culture media, some pre-
cipitation was observed.
14 P. Yu, B. Liu and T. Kodadek, Nat. Biotechnol., 2005, 23, 746.
15 R. N. Zuckermann, J. M. Kerr, S. B. Kent and W. H. Moos, J.
Am. Chem. Soc., 1992, 114, 10646.
ꢀc
This journal is The Royal Society of Chemistry 2008
1066 | Chem. Commun., 2008, 1064–1066