Vibralactone as a tool to study the activity and structure of the ClpP1P2 complex from listeria monocytogenes

Article abstract of DOI:10.1002/anie.201104391

The Clp proteolytic machinery has important functions in many bacteria such as L. monocytogenes. Some organisms encode for two uncharacterized ClpP isoforms. Vibralactone was used to study the activity and assembly of ClpP1 and ClpP2 subunits in a hetero-

Full text of DOI:10.1002/anie.201104391

DOI: 10.1002/anie.201104391
Heterooligomeric Complexes
Vibralactone as a Tool to Study the Activity and Structure of the
ClpP1P2 Complex from Listeria monocytogenes**
Evelyn Zeiler, Nathalie Braun, Thomas Bçttcher, Andreas Kastenmꢀller, Sevil Weinkauf, and
Stephan A. Sieber*
Nature provides a rich source of bioactive compounds
comprising a diverse set of electrophilic core structures that
are poised to react with corresponding nucleophilic residues
such as serine and cysteine in enzyme active sites.[1–3] These
residues are usually relevant for catalysis and therefore
display fine-tuned reactivity towards their dedicated sub-
strates.[4] We and others previously investigated the dedicated
targets of monocyclic b-lactones which turned out to be
potent and selective inhibitors of diverse disease-associated
enzyme classes.[2,3,5–8] Covalent inhibition of the caseinolytic
peptidase ClpP, for instance, resulted in a dramatic attenu-
ation of bacterial virulence.[3] ClpP is an important, highly
conserved heat shock protein with additional regulatory
functions in many pathogens.[9–11] Some organisms such as
Listeria monocytogenes genetically encode for two function-
ally and structurally uncharacterized ClpP isoforms (ClpP1
and ClpP2). So far, all b-lactones were reported to target
solely ClpP2 and not ClpP1,[2] raising the question whether
monocyclic lactones lack suitable reactivity to interact with
the ClpP1 active-site nucleophile.
We herein expand the scope of natural-product-derived b-
lactones to strained bicyclic ring systems which may exhibit
enhanced reactivity profiles. The natural products omuralide,
salinosporamide, and vibralactone (VL) represent such
desired scaffolds and have been reported to be potent
proteasome or lipase inhibitors.[12–14] We utilized a chemical
proteomic strategy termed “activity-based protein profiling
(ABPP)”[15–17] to demonstrate that vibralactone (VL), con-
trary to monocyclic b-lactones, binds to both ClpP1 and ClpP2
in L. monocytogenes. Moreover, by combining transmission
electron microscopy (TEM) and homology modeling/struc-
ture predictions, we were able to determine the quaternary
structure of the hetero-oligomeric complex (Figure 1).
VL was synthesized as described by Zhou and Snider[18]
(Scheme 1 in the Supporting Information) and modified with
an alkyne handle in the final step for target discovery by
ABPP (Figure 1).[19] Target analysis started by the incubation
of this vibralactone probe (VLP) with intact living cells of
L. welshimeri and its pathogenic counterpart L. monocyto-
genes. Upon cell lysis, the proteome was treated under click
chemistry (CC)[20–22] conditions with rhodamine azide and the
targets were visualized by fluorescent SDS-PAGE analysis
(Figure 1 in the Supporting Information). Two strong fluo-
rescent bands for approximately 20 kDa proteins were
present at comparable intensities in L. welshimeri as well as
in L. monocytogenes (Figure 2A) down to a VLP concen-
tration of 3.4 mm (Figure 1C in the Supporting Information).
Pre-incubation with various concentrations of unmodified VL
gradually abolished the labeling of these bands, demonstrat-
ing that the natural product exhibits comparable target
selectivity (Figure 2B). Mass spectrometric (MS) analysis
revealed that the lower band corresponds to ClpP2 which has
been labeled by monocyclic b-lactones before.[2] Interestingly,
the upper band corresponds to ClpP1 which could not be
addressed by any other b-lactone probe (Figure 2A, Table 1
in the Supporting Information). While ClpP2 orthologues
from various organisms exhibit a high sequence homology
(77% identity between L. monocytogenes and S. aureus) with
a tetradecameric barrel-shaped assembly in crystal struc-
tures,[23–26] ClpP1 shares only 41% identity with ClpP2
(Figure 2 in the Supporting Information). This raises the
question whether ClpP1 exhibits a different fold and function
and assembles with ClpP2 in mixed complexes, as previously
suggested for hetero-oligomeric complexes of different ClpP
isoforms.[27]
[*] E. Zeiler, Prof. Dr. S. A. Sieber
Department Chemie, Center for Integrated Protein Science CIPSM,
Institute of Advanced Studies IAS, Technische Universitꢀt Mꢁnchen
Lichtenbergstrasse 4, 85747 Garching (Germany)
E-mail: stephan.sieber@tum.de
Dr. N. Braun, A. Kastenmꢁller, Prof. Dr. S. Weinkauf
Center for Integrated Protein Science Munich CIPSM
Department of Chemistry, Technische Universitꢀt Mꢁnchen
Lichtenbergstrasse 4, 85747 Garching (Germany)
Dr. T. Bçttcher[+]
AVIRU, EXIST Transfer of Research, OC II
Lichtenbergstrasse 4, 85747 Garching (Germany)
[+] Current address:
Department of Biological Chemistry and Molecular Pharmacology,
Harvard Medical School, Boston, MA (USA)
[**] We thank Mona Wolff for excellent scientific support and Wolfgang
Steglich for helpful discussions. E.Z. was supported by the SFB749
and by the TUM-GS. E.Z. thanks Quan Zhou for helpful discussions.
T.B. was supported by the German National Academic Foundation
and by an EXIST technology transfer grant of the Federal Ministry of
Economics and Technology (BMWi); S.A.S. was supported by the
Deutsche Forschungsgemeinschaft (Emmy Noether), SFB749,
FOR1406, an ERC starting grant, and the Center for Integrated
Protein Science Munich CIPSM. S.W. and N.B. were funded by the
DFG (SFB594) and CIPSM.
Supporting information for this article (including details on the
synthesis and characterization of compounds, bioassays, electron
microscopy, image processing/3D reconstruction, structure pre-
diction, homology modeling as well as proteome preparation and
labeling) is available on the WWW under http://dx.doi.org/10.1002/
anie.201104391.
To address these questions, we recombinantly overex-
pressed ClpP1 and ClpP2 independently as well as by means
of a co-expression vector system in Escherichia coli. Co-
expressed ClpP1P2 was isolated through a C-terminal strep
Angew. Chem. Int. Ed. 2011, 50, 11001 –11004
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
11001

Communications
Figure 1. Discovery platform
for vibralactone (VL) and the
structural organization of
ClpP1P2. The vibralactone
probe (VLP) was used in
living cells, and its dedicated
targets were identified by
SDS gel analysis and MS.
Functional studies with the
two prominent targets ClpP1
and ClpP2 reveal a hetero-
oligomeric assembly that was
further investigated by elec-
tron microscopy (EM).
tag at ClpP2 by affinity chromatography,
whereby enrichment of ClpP1 is only possible
if both enzymes form a stable hetero-oligo-
meric complex. Indeed, the existence of a
hetero-oligomeric ClpP1P2 complex was con-
firmed by labeling with VLP (Figure 3 in the
Supporting Information). Vice versa, equimo-
lar mixing of purified ClpP1 with ClpP2 for
different durations showed resulted in equal
labeling of both enzymes between 24 h and
three days (Figure 4 in the Supporting Infor-
mation), indicating that ClpP2 triggered acti-
vation of ClpP1. While co-expressed ClpP1P2
revealed the same labeling pattern with VLP as
that observed in the native Listeria proteomes,
ClpP1 alone did not interact with VLP and
selected monocyclic lactones (U1P, D3), in
contrast ClpP2 was labeled by all lactones
(Figure 3 in the Supporting Information). This
suggests that the presence of ClpP2 is crucial
for ClpP1 acylation and therefore likely for its
activation at least in vitro.
In a peptidase activity assay with a fluoro-
genic model substrate VLP, VL and the
monocyclic lactone U1P inhibited ClpP2 with
EC50 values of 27 mm, 154 mm, and 4 mm (Fig-
ure 2C) as well as an equimolar ClpP1/ClpP2
mixture with EC50 values of 41 mm, 167 mm, and
3 mm, respectively (Figure 2D). In addition,
both VL and VLP could be detected covalently
attached to the active-site residue S98 of ClpP2
by LC–MS (Figure 5 in the Supporting Infor-
Figure 2. A) Fluorescent SDS-PAGE analysis of the
L. monocytogenes proteome after incubation with VLP
and different monocyclic b-lactones. B) Competitive
ABPP experiment with VLP and different-fold excess of
VL. Activity of recombinant C) ClpP2 and D) ClpP1/
ClpP2 (1:1 mixture) after incubation with VLP, VL, and
U1P at different concentrations.
11002 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 11001 –11004

mation). These results confirm together with
heat denaturation studies (Figure 1D in the
Supporting Information) that probe labeling is
active-site directed and occurs only with the
native, folded protein. No enzyme activity was
observed for purified ClpP1 under any con-
ditions tested. The optimum of peptidase
activity was found at pH 7 (Figure 6A in the
Supporting Information), and interestingly
ClpP2 and ClpP1P2 activity increased by
more than twofold after the addition of 5%
glycerol (Figure 6B in the Supporting Infor-
mation).[28]
To correlate enzyme activity with the
quaternary organization of ClpP1, ClpP2, and
ClpP1P2 we subjected all purified proteins to
size-exclusion chromatography. While ClpP2
and co-expressed ClpP1P2 were found as
tetradecameric complexes at 5% glycerol,
lack of glycerol resulted to a large extent in
disassembly into heptamers (Figure 7A in the
Supporting Information). This suggests that
Figure 3. Three-dimensional model of the ClpP1P2 tetradecamer. Upper row: Surface
representations and density cross sections of the ClpP1P2 tetradecamer as viewed from
the side (A, central cross section) and along the sevenfold symmetry axis intercepting
the ClpP2 (B) and ClpP1 rings (C). Positions of the cross sections in (B) and (C) are
indicated by dashed lines in (A). Note the additional density on top of the ClpP2 ring
(arrow). Lower row: Views of the oligomer in the same orientations as in the upper row
with the docked homology model of the ClpP1P2 tetradecamer (ribbon representation)
superimposed. The ribbon models of ClpP2 and ClpP1 are shown in red and blue,
respectively. The N-terminal loop of ClpP2 is colored green, strep tag of ClpP2 yellow.
glycerol, a stabilizing agent of the compact Scale bar:
5 nm.
native form of flexible proteins, supports
oligomerization into tetradecamers as reported
previously.[29] Interestingly, ClpP1 eluted as a heptameric
complex, independent of the glycerol content, emphasizing
that ClpP1/ClpP2 activity depends on the oligomeric complex
assembly with tetradecamers as active and heptamers as
inactive forms. This is further supported by VLP incubation in
the absence of glycerol with ClpP2 and co-expressed ClpP1P2
which shows less labeling for ClpP2 and no ClpP1 labeling in
the co-expressed complex (Figure 7B in the Supporting
Information).
To elucidate the quaternary structure of the ClpP1P2
complex in more detail, the samples were visualized by
negative-stain electron microscopy (Figure 8 in the Support-
ing Information) which, revealed side and top views of barrel-
shaped oligomers of 11.5 nm in height and 11.0 nm in
diameter. The three-dimensional (3D) model at 15 ꢀ reso-
lution, obtained by single-particle reconstruction (Figure 3)
revealed a tetradecameric assembly with a central pore,
resembling the structures of ClpP from other species.[26] Most
strikingly, however, the two heptameric rings of the tetrade-
cameric assembly turned out to be nonidentical as one of the
rings contained an additional mass at one end of the barrel
(Figure 3A). Upon homology modeling of the tetradecamer
using the predicted structures of ClpP1 and ClpP2, and the
structure of ClpP from E. coli (pdb ID 1YG6) as a template,
we were able to correlate this mass to the flexible N-terminal
loop region which contains six additional amino acids in the
sequence of ClpP2 that are not present in ClpP1 (Figures 2
and 9 in the Supporting Information). Although the predicted
secondary-structure elements as well as tertiary structures of
ClpP1 and ClpP2 monomers correlated extremely well,
except for the N-terminal loop region (Figure 9A in the
Supporting Information), the corresponding homoheptamer
models exhibited distinct differences (Figure 4). The diameter
of the pore in ClpP2 (2 nm) appeared, as a result of the
Figure 4. Distribution of charged residues on the surface of homology-
modeled ClpP1 (A) and ClpP2 (B) homoheptamers. Residue color
coding: negatively charged, red; neutral, white; positively charged,
blue. Scale bars: 10 nm. C) Transmission electron micrographs of
negatively stained ClpP1, ClpP2, and ClpP1P2 complexes (0.2 mgmLÀ1
protein, 1.5% (w/v) uranyl acetate) with either 5% glycerol (left) or
diluted to a very low (<0.5%) glycerol concentration (right). Insets:
Characteristic class averages. Top: Side views, bottom: top views after
sevenfold symmetrization. For each data set, five class averages were
calculated and compared from approximately 1500 top views along the
sevenfold axis. Color coding of frames: ClpP1, blue; ClpP2, red. Scale
bars: 100 nm, box sizes of class averages: 19 nm.
Angew. Chem. Int. Ed. 2011, 50, 11001 –11004
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 11003

Communications
localization of the N-terminal loops at the entrance, narrower
[1] D. S. Johnson, E. Weerapana, B. F. Cravatt, Future Med. Chem.
2010, 2, 949.
[2] T. Bçttcher, S. A. Sieber, Angew. Chem. 2008, 120, 4677; Angew.
Chem. Int. Ed. 2008, 47, 4600.
[3] T. Bçttcher, S. A. Sieber, J. Am. Chem. Soc. 2008, 130, 14400.
[4] C. Drahl, B. F. Cravatt, E. J. Sorensen, Angew. Chem. 2005, 117,
5936; Angew. Chem. Int. Ed. 2005, 44, 5788.
[5] T. Bçttcher, S. A. Sieber, ChemMedChem 2009, 4, 1260.
[6] Z. Wang, C. Gu, T. Colby, T. Shindo, R. Balamurugan, H.
Waldmann, M. Kaiser, R. A. van der Hoorn, Nat. Chem. Biol.
2008, 4, 557.
[7] M. H. Ngai, P. Y. Yang, K. Liu, Y. Shen, M. R. Wenk, S. Q. Yao,
M. J. Lear, Chem. Commun. 2010, 46, 8335.
[8] P. Y. Yang, K. Liu, M. H. Ngai, M. J. Lear, M. R. Wenk, S. Q.
Yao, J. Am. Chem. Soc. 2010, 132, 656.
[9] D. Frees, K. Sorensen, H. Ingmer, Infect. Immun. 2005, 73, 8100.
[10] A. Michel, F. Agerer, C. R. Hauck, M. Herrmann, J. Ullrich, J.
Hacker, K. Ohlsen, J. Bacteriol. 2006, 188, 5783.
[11] D. Frees, K. Savijoki, P. Varmanen, H. Ingmer, Mol. Microbiol.
2007, 63, 1285.
than in ClpP1 (3 nm; Figure 4). Moreover, whereas the
distribution of the charged residues on the surface of ClpP1
seemed to be random (Figure 4A and Figure 10 in the
Supporting Information), the ClpP2 surface exhibited signifi-
cant patches of high negative charge (Figure 4B and Figure 10
in the Supporting Information). Indeed, these different
characteristics were reflected by distinct appearances of the
ClpP1 and ClpP2 homoheptamers in negative-stain micros-
copy. In line with the surface charge distribution, ClpP2
heptamers showed strong stain accumulation, masking the
slight depressions present in the (predicted) structure and
giving rise to a less-contoured class average (Figure 4C). In
contrast ClpP1 heptamers were only faintly stained and
resembled a symmetric seven-pointed star. Also in agreement
with the predicted structure, the central pore of ClpP2
appeared narrower in the negative stain image. In line with
the results from size-exclusion chromatography (Figure 7A in
the Supporting Information), in co-expressed ClpP1P2 prep-
arations containing 5% glycerol mainly tetradecameric
oligomers were observed whose top views strongly resembled
ClpP1 and ClpP2 homoheptameric rings (Figure 4C). Upon
dilution of glycerol to concentrations below 0.5%, we again
found only class averages correlating to either ClpP1 or ClpP2
homoheptamers. As these homoheptameric rings were
derived from the dissociation of the tetradecameric species,
we conclude that the ClpP1P2 tetradecamers are mainly
composed of homoheptameric ClpP1 and ClpP2 rings.
Surprisingly, a different structural assembly was reported
and proposed for distantly related cyanobacterial ClpP iso-
forms composed of a tetradecamer of two heteroheptameric
rings.[27,30] This indicates, consistent with our results, a func-
tional specialization of ClpP isoforms with yet unexplored,
presumably regulatory functions.
In conclusion, we have reported on the bicyclic b-lactone
vibralactone as an unprecedented probe for labeling two
isoforms of the important ClpP protease from L. monocyto-
genes. We expanded the scope of the probe as a tool to study
the activity and assembly of ClpP1 and ClpP2 subunits in a
hetero-oligomeric composition. Our results suggest that
ClpP1 is activated by hetero-oligomerization with ClpP2,
and the tetradecameric assembly enhances the catalytic
activity. Finally, electron microscopic images indicate that
the tetradecameric assembly is constituted by two homohep-
tameric ClpP1 and ClpP2 rings that are stacked on top of each
other. These results provide the first insight into a novel
complex assembly of an important class of bacterial enzymes.
[12] D. Z. Liu, F. Wang, T. G. Liao, J. G. Tang, W. Steglich, H. J. Zhu,
J. K. Liu, Org. Lett. 2006, 8, 5749.
[13] E. J. Corey, W. D. Li, Chem. Pharm. Bull. 1999, 47, 1.
[14] R. H. Feling, G. O. Buchanan, T. J. Mincer, C. A. Kauffman,
P. R. Jensen, W. Fenical, Angew. Chem. 2003, 115, 369; Angew.
Chem. Int. Ed. 2003, 42, 355.
[15] T. Bçttcher, M. Pitscheider, S. A. Sieber, Angew. Chem. 2010,
122, 2740; Angew. Chem. Int. Ed. 2010, 49, 2680.
´
[16] M. Fonovic, M. Bogyo, Expert Rev. Proteomics 2008, 5, 721.
[17] M. J. Evans, B. F. Cravatt, Chem. Rev. 2006, 106, 3279.
[18] Q. Zhou, B. B. Snider, Org. Lett. 2008, 10, 1401.
[19] A. E. Speers, G. C. Adam, B. F. Cravatt, J. Am. Chem. Soc. 2003,
125, 4686.
[20] R. Huisgen, 1,3 Dipolar Cylcoaddition Chemistry, Wiley, New
York, 1984.
[21] V. V. Rostovtsev, J. G. Green, V. V. Fokin, K. B. Sharpless,
Angew. Chem. 2002, 114, 2708; Angew. Chem. Int. Ed. 2002,
41, 2596.
[22] C. W. Tornøe, C. Christensen, M. Meldal, J. Org. Chem. 2002, 67,
3057.
[23] A. Szyk, M. R. Maurizi, J. Struct. Biol. 2006, 156, 165.
[24] A. Y. Yu, W. A. Houry, FEBS Lett. 2007, 581, 3749.
[25] A. Gribun, M. S. Kimber, R. Ching, R. Sprangers, K. M. Fiebig,
W. A. Houry, J. Biol. Chem. 2005, 280, 16185.
[26] S. R. Geiger, T. Bçttcher, S. A. Sieber, P. Cramer, Angew. Chem.
2011, 123, 5867; Angew. Chem. Int. Ed. 2011, 50, 5749.
[27] T. M. Stanne, E. Pojidaeva, F. I. Andersson, A. K. Clarke, J. Biol.
Chem. 2007, 282, 14394.
[28] M. W. Thompson, M. R. Maurizi, J. Biol. Chem. 1994, 269,
18201.
[29] Z. Maglica, K. Kolygo, E. Weber-Ban, Structure 2009, 17, 508.
[30] F. I. Andersson, A. Tryggvesson, M. Sharon, A. V. Diemand, M.
Classen, C. Best, R. Schmidt, J. Schelin, T. M. Stanne, B. Bukau,
C. V. Robinson, S. Witt, A. Mogk, A. K. Clarke, J. Biol. Chem.
2009, 284, 13519.
[31] The EM 3D structure of the tetradecameric ClpP1P2 complex of
L. monocytogenes has been deposited in the EMBD data bank
with the accession code EMD-1913.
Received: June 24, 2011
Published online: September 22, 2011
Keywords: electron microscopy · natural products ·
.
protein structures · proteomics · vibralactone
11004 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 11001 –11004