Structural analysis of phenothiazine derivatives as allosteric inhibitors of the MALT1 paracaspase

Article abstract of DOI:10.1002/anie.201304290

Second site: In the crystal structure of human MALT1_casp-Ig3 (mucosa-associated lymphoid tissue lymphoma translocation protein 1) in complex with the tricyclic phenothiazine derivative thioridazine (violet in the picture), the inhibitor is bound in a hydrophobic pocket far from the active site. This explains the action of phenothiazine derivatives as noncompetitive, reversible inhibitors. Copyright

Full text of DOI:10.1002/anie.201304290

Angewandte
Chemie
DOI: 10.1002/anie.201304290
Medicinal Chemistry
Structural Analysis of Phenothiazine Derivatives as Allosteric
Inhibitors of the MALT1 Paracaspase**
Florian Schlauderer, Katja Lammens,* Daniel Nagel, Michelle Vincendeau,
Andrea C. Eitelhuber, Steven H. L. Verhelst, Dale Kling, Al Chrusciel, Jꢀrgen Ruland,
Daniel Krappmann, and Karl-Peter Hopfner
The human mucosa-associated lymphoid tissue lymphoma
translocation protein 1 (MALT1) is responsible for the
survival, proliferation, and activation of B and T lymphocytes
upon antigen stimulation by means of the canonical NF-kB
signaling pathway.[1] MALT1 is constitutively associated with
BCL10 and assembles to form the CBM complex with
CARMA1, CARMA3, or CARD9 upon activation by their
individual receptors (CARMA1: CARD-containing mem-
brane-associated guanylate kinase protein 1; CARD9: cas-
pase-recruitment-domain-containing protein 9).[2] In the acti-
vated CBM complex MALT1 acts as a scaffolding platform
and promotes the recruitment of signaling factors like tumor
necrosis factor (TNF) receptor-associated factor 6 (TRAF6),
transforming growth factor (TGF)-b-activated kinase
(TAK1), and the regulatory subunit NEMO (NF-kB essential
modulator) of the IKK complex (inhibitor of transcription
factor NF-kB (IkB) kinase)[3] which finally leads to IKK
activation.[4] Besides its scaffolding function, MALT1 confers
proteolytic activity that is required for optimal T-cell activa-
tion.[5] Cleavage of the MALT1 substrates BCL10, A20,
CYLD, and RelB results in enhanced NF-kB as well as c-Jun
N-terminal kinase (JNK) activation and controls T-cell
adhesion.[5a,b,6] Recent studies revealed that MALT1 is
activated by ligand binding and dimerization of the para-
caspase domain.[7] However, spontaneous dimerization of
MALT1 in vitro leads to a catalytically inactive conforma-
tion.[7b] Hence, an additional structural rearrangement is
needed for MALT1 activation that is supported by mono-
ubiquitination in the C-terminal Ig3 domain of MALT1.[7b,8]
Modified activity of individual members of the CBM
signaling complex is associated with lymphomagenesis.[9] In
this context the deregulated expression of CARMA1, BCL10,
and MALT1 is critical for the survival of the activated B-cell
subtype of diffuse-large B-cell lymphoma (ABC-DLBCL)[10]
and constitutive MALT1 paracaspase activity is a common
feature of ABC-DLBCL cells. Moreover, two recent studies
demonstrated the crucial role of MALT1 in the early phase of
experimental autoimmune encephalomyelitis (EAE), the
main animal model for multiple sclerosis (MS).[11] These
results suggest that MALT1 is an important therapeutic target
to treat multiple sclerosis, the most common chronic inflam-
matory demyelinating disease of the human central nervous
system.[12] Thus, there is substantial interest in developing
small-molecule compounds that specifically inhibit MALT1,
a promising new approach for the treatment of MALT
lymphoma, ABC-DLBCL, and multiple sclerosis.[11b,c]
Recently, distinct phenothiazines have been identified as
potent small-molecule inhibitors of MALT1 that selectively
kill ABC-DLBCL in vitro and in vivo.[13] The identified drugs,
mepazine, thioridazine, and promazine, have a long clinical
history as antipsychotics.[14] Yet, it is still unclear how these
substances inhibit MALT1. A detailed understanding of the
binding mode is crucial to further optimize the efficacy and
selectivity of these compounds for clinical use.
Here, we report the crystal structure of ligand-free
dimeric human MALT1Casp-Ig3 in complex with the tricyclic
phenothiazine derivative thioridazine. Unexpectedly, the
structure reveals that the inhibitor binds in a pocket located
opposite to the caspase active site, in the interface between
[*] F. Schlauderer,[+] Dr. K. Lammens,[+] Prof. K.-P. Hopfner
Gene Center, Department of Biochemistry
Prof. Dr. J. Ruland
Institut fꢀr Klinische Chemie und Pathobiochemie
Klinikum rechts der Isar, Technische Universitꢁt Mꢀnchen
81675 Mꢀnchen (Germany)
Ludwig-Maximilians University Munich
Feodor-Lynen-Strasse 25, 80377 Munich (Germany)
E-mail: klammens@genzentrum.lmu.de
[+] These authors contributed equally to this work.
D. Nagel, Dr. M. Vincendeau, Dr. A. C. Eitelhuber, Dr. D. Krappmann
Research Unit Cellular Signal Integration
Institute of Molecular Toxicology and Pharmacology
Helmholtz Zentrum Mꢀnchen
[**] We thank C. Basquin for the CD measurements and G. Witte for help
with the fluorescence quenching assays. We thank the Max-Planck
Crystallization Screening Facility for initial crystallization trials. We
thank the Swiss Light Source (SLS, Villigen, Switzerland) and the
European Synchrotron Radiation Facility (ESRF, Grenoble, France)
for beam time and onsite support. S.V. is supported by the DFG
(Emmy Noether program); F.S., K.L., D.K., and K.P.H. are
supported by the SFB 1054.
Ingolstꢁdter Landstrasse 1, 85764 Neuherberg (Germany)
Dr. S. H. L. Verhelst
Lehrstuhl fꢀr Chemie der Biopolymere
Technische Universitꢁt Mꢀnchen
Weihenstephaner Berg 3, 85354 Freising (Germany)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201304290.
D. Kling, A. Chrusciel
Kalexsyn, Inc.
4502 Campus Drive, Kalamazoo, MI 49008 (USA)
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
Figure 1. Thioridazine binds at an allosteric position in the interface
between the paracaspase and Ig3 domain connecting helix a1Ig3.[17]
a) Superposition of MALT1Casp-Ig3 bound to hex-LRSR-peptide (light blue)
and in complex with thioridazine (gold). The biologically relevant
MALT1 dimer is shown in gray. b) The conformational rearrangement of
helix aC and aD and subsequently b-sheets 3A and 3B is inhibited due
to the steric hindrance by thioridazine. c) Close-up view of the hydro-
phobic binding pocket; the residues that interact with thioridazine are
shown as sticks. The violet mesh represents the refined 2FoÀFc electron
density map (contoured at 1s) for the ligand, shown together with the
anomalous difference density map (contoured at 2.2s and shown in
yellow).
the caspase domain and the Ig3 domain connecting helix a1Ig3
of MALT1 (Figure 1a). This allosteric binding site, far from
the catalytic center well explains the fact that phenothiazine
derivatives act as noncompetitive, reversible inhibitors.[13]
Superposition of the enzymatic active MALT1Casp-Ig3 construct
bound to the hex-LRSR-AOMK peptide (unpublished data)
with the thioridazine-bound structure indicate, that binding of
the compound between helices a1Ig3 and aC prevents the
conformational change into an active enzyme, the so-called
second activation step of MALT1[7b] (Figure 1). Besides
ligand-induced rearrangements of the active site loops three
major shifts of helices aC, aD, and of b-sheets 3A and 3B are
essential to achieve the enzymatic proficient protease con-
formation (Figure 1b). The movement of helices aC and aD
is hampered by the sandwiched thioridazine, and subse-
quently b-sheets 3A and 3B cannot perform their pivotal shift
(Figure 1b). A detailed analysis of the inhibitor binding site
shows that the tricyclic ring system of thioridazine is bound in
a hydrophobic pocket composed of residues A394, F398, and
L401 in helix aC and L346, V344, and V381 in b-sheets 1 and
2, respectively (Figure 1c). The orientation of the 2-methyl-
thiophenothiazine ring was proven by collecting a dataset at
a wavelength of 1.9 ꢀ to detect the anomalous signal of sulfur
(Figure 1c).
Figure 2. MALT1 mutant E397A is less responsive to the inhibitory
potential of mepazine and thioridazine. a) Close up view of W580
flipping and the interaction of thioridazine with residue E397. The
thioridazine-bound and ligand-free MALT1 structures are shown in
gold and gray, respectively (PDB code: 3V55).[7b] b–d) Tryptophan
fluorescence quenching titration study of monomeric MALT1Casp-Ig3 and
the E397A mutant with thioridazine, mepazine, and promethazine
(negative control). The binding constants of the different phenothia-
zine inhibitors to the wt and mutant protein are listed below the
respective graphs. e) Activity of MALT1 wt and E397A is inhibited by
the peptide-based MALT1 inhibitor Z-VRPR-FMK. Increasing amounts
of the peptide led to an equivalent loss of MALT1 activity. The data
confirmed that the mutant E397A has no influence on the activity of
MALT1. f–h) MALT1 cleavage assay performed with wt and E397A GST-
MALT1325-760 after incubation with mepazine, thioridazine, and pro-
methazine. Mepazine and thioridazine have a lower inhibitory potential
on the E397A mutant than on the wt, while the impact of the weak
MALT1 inhibitor promethazine on both MALT1 variants is equivalent.
a substantial displacement of helix a1Ig3 (Figure 2a). Probably
triggered by rotation of this domain-connecting helix, the
entire Ig3 domain becomes more flexible (Figure 1a). To
verify the mechanism of inhibition by phenothiazine-based
drugs with MALT1 in solution, we developed a tryptophan-
fluorescence-quenching assay. For this assay we took advant-
age of the fact that W580 is the only tryptophan residue in the
MALT1Casp-Ig3 construct and in close proximity to the bound
tricyclic compound. The tryptophan fluorescence of mono-
Upon inhibitor binding, the side chain of residue trypto-
phan W580 on helix a1Ig3 is flipped out of the hydrophobic
groove into a solvent-exposed environment which leads to
2
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
meric MALT1Casp-Ig3 was recorded with increasing amounts of
thioridazine (Figure 2b). The titration was continued until
saturation of quenching was observed.
both compounds reduced the viability of cells transduced with
MALT1 wildtype, while the HBL-1 cells transduced with the
E397A mutant were largely resistant to mepazine and
thioridazine (Figure 3). FACS histograms and western blot
show that MALT wildtype and E397A mutant proteins are
only expressed after addition of doxycycline (Figure 2a,b in
the Supporting Information).
To confirm that the titration of thioridazine has no effect
on the general folding of MALT1Casp-Ig3 under the applied
conditions, circular dichroism spectra were collected
(Figure 1 in the Supporting Information). The association
constants were calculated by plotting the percentage of W580
quenching versus the concentration of the compound and
fitting the curves to a one-site binding model. The equivalent
experiment was conducted using mepazine; similar results led
to the proposal that different phenothiazine derivatives
inhibit MALT1 by the same mechanism (Figure 2c). As
a negative control, the weak inhibitory phenothiazine-based
compound promethazine was used in the tryptophan-quench-
ing assay (Figure 2d); promethazine showed a significant
lower binding constant, consistent with the weak inhibitory
effect in vivo.[13]
The structure of inhibitor-bound MALT1 well explains
the effect of different phenothiazine modifications on their
inhibitory effectiveness as Nagel et al. had described. On the
one hand, the phenothiazine ring system has an optimal size
to fit in the hydrophobic pocket and only minor changes that
increase the size or enhance the solubility are tolerated. On
the other hand, the hydrogen bond between the N-methyl-
piperidine nitrogen of thioridazine and glutamic acid E397
seems to play a major role in MALT1 recognition (Figure 2a),
as the phenothiazine backbone alone is a very weak MALT1
inhibitor. All derivatives tested with a piperidyl ring system
and/or nitrogen in a similar position to the piperidyl nitrogen
showed an inhibitory effect in comparable range.[13] Modifi-
cations in the connecting alkyl chain, for example keto- or
hydroxy groups, increased the IC50 values up to tenfold.[13] We
concluded that the tricyclic ring system and the methylpiper-
idyl group play equally important roles in drug recognition.
To verify the influence of the E397–thioridazine interaction
and to further support the proposed binding mechanism for
phenothiazine derivatives, an E397A mutant of MALT1 was
tested with regard to its inhibition by different phenothiazine
derivatives in vitro and in vivo. To ensure correct folding of
the mutant protein, an in vitro enzymatic activity assay was
performed (Figure 2e). To define the inhibitory potential on
wild-type (wt) and mutant MALT1, IC50 values were deter-
mined for thioridazine, mepazine, and promethazine on GST-
MALT1325-760 wt and E397A (Figure 2 f–h). The two com-
pounds with a piperidyl ring system, thioridazine and
mepazine, display a lower inhibitory potential on the E397A
mutant than on the wt, whereas the mutation had no influence
on the weak inhibitory effect of promethazine. These results
are consistent with the tryptophan-quenching data and
emphasize the importance of the E397–inhibitor interaction
on the inhibitory potential of these phenothiazine derivatives
in vitro.
Figure 3. HBL-1 cells were lentivirally transduced with either MALT1
wildtype or the point mutant E397A. Subsequently, they were treated
with mepazine and thioridazine, and cell viability was monitored after
four days. The data demonstrate that mepazine and thioridazine
diminish cell viability of cells expressing MALT1 wt, while the E397A
mutant confers increased resistance to the compounds.
A detailed inspection of the electron density map of the
inhibitor suggests that solely the S enantiomer of thioridazine
is bound in the crystal structure (Figure 1c). However, we
cannot exclude that the R enantiomer is able to bind with
comparable affinity at this position. To analyze the influence
of chirality on the binding affinity and inhibitory potential of
the individual enantiomers, (R)- and (S)-mepazine and
thioridazine were prepared in enantiomerically pur form
pure as described in the Supporting Information and analyzed
accordingly. Whereas (R)- and (S)-thioridazine show equiv-
alent binding affinities and IC50 values (Figure 4a,c) (S)-
mepazine exhibits a significantly higher binding affinity and
an up to eight times higher inhibitory potential over (R)-
mepazine (Figure 4b,d). So far thioridazine has been applied
in the racemic form as the hydrochloride of 10-[2-(1-
methylpiperid-2-yl)ethyl]-2-methylthiophenothiazine for the
symptomatic therapy of psychotic disorders.[14] Nevertheless,
the antipsychotic effect is believed to be associated with (R)-
thioridazine.[15] Since (S)-thioridazine binds MALT1 with
a comparable affinity, it may be possible to reduce the
sedative effects by using the pure S enantiomer for treatment
of MALT1-driven cancer or autoimmune diseases. In any
case, the increased affinity of (S)-mepazine represents the
first step towards an MALT1-optimized phenothiazine-based
compound.
Since the 1950s thioridazine, mepazine, and other pheno-
thiazines have been used clinically as antipsychotic drugs,
where they exert their sedative effects by acting as dopamine
receptor antagonists in the brain.[14] Recently, thioridazine
was shown to induce toxicity in cancer stem cells (CSCs) while
having no effect on normal human pluripotent stem cells
(hPSCs). However, quite in contrast to the situation found in
MALT1-dependent ABC-DLBCL, CSCs express dopamine
receptors, and thioridazine affects CSC survival by acting as
To test the proposed inhibitor-binding mechanism in cells,
the human cell line HBL-1 derived from DLBCL patients was
transduced with wildtype or E397A mutant MALT1. After
doxycycline-induced MALT1 expression the cells were
exposed to either mepazine or thioridazine. Analysis of cell
viability by cell count four days after treatment revealed that
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
b) H. Zhou, M. Q. Du, V. M. Dixit, Cancer Cell 2005, 7, 425 –
431.
[3] a) L. Sun, L. Deng, C. K. Ea, Z. P. Xia, Z. J. Chen, Mol. Cell
2004, 14, 289 – 301; b) O. Gaide, B. Favier, D. F. Legler, D.
Bonnet, B. Brissoni, S. Valitutti, C. Bron, J. Tschopp, M. Thome,
Nat. Immunol. 2002, 3, 836 – 843.
[4] A. Oeckinghaus, E. Wegener, V. Welteke, U. Ferch, S. C. Arslan,
J. Ruland, C. Scheidereit, D. Krappmann, EMBO J. 2007, 26,
4634 – 4645.
[5] a) F. Rebeaud, S. Hailfinger, A. Posevitz-Fejfar, M. Tapernoux,
R. Moser, D. Rueda, O. Gaide, M. Guzzardi, E. M. Iancu, N.
Rufer, N. Fasel, M. Thome, Nat. Immunol. 2008, 9, 272 – 281;
b) B. Coornaert, M. Baens, K. Heyninck, T. Bekaert, M.
Haegman, J. Staal, L. Sun, Z. J. Chen, P. Marynen, R. Beyaert,
Nat. Immunol. 2008, 9, 263 – 271; c) M. Duwel, V. Welteke, A.
Oeckinghaus, M. Baens, B. Kloo, U. Ferch, B. G. Darnay, J.
Ruland, P. Marynen, D. Krappmann, J. Immunol. 2009, 182,
7718 – 7728.
[6] a) S. Hailfinger, H. Nogai, C. Pelzer, M. Jaworski, K. Cabalzar,
J. E. Charton, M. Guzzardi, C. Decaillet, M. Grau, B. Dorken, P.
Lenz, G. Lenz, M. Thome, Proc. Natl. Acad. Sci. USA 2011, 108,
14596 – 14601; b) S. Rosebeck, P. C. Lucas, L. M. McAllister-
Lucas, Future Oncol. 2011, 7, 613 – 617; c) J. Staal, Y. Driege, T.
Bekaert, A. Demeyer, D. Muyllaert, P. Van Damme, K. Gevaert,
R. Beyaert, EMBO J. 2011, 30, 1742 – 1752.
[7] a) J. Hachmann, S. J. Snipas, B. J. van Raam, E. M. Cancino, E. J.
Houlihan, M. Poreba, P. Kasperkiewicz, M. Drag, G. S. Salvesen,
Biochem. J. 2012, 443, 287 – 295; b) C. Wiesmann, L. Leder, J.
Blank, A. Bernardi, S. Melkko, A. Decock, A. D’Arcy, F. Villard,
P. Erbel, N. Hughes, F. Freuler, R. Nikolay, J. Alves, F.
Bornancin, M. Renatus, J. Mol. Biol. 2012, 419, 4 – 21; c) J. W.
Yu, P. D. Jeffrey, J. Y. Ha, X. Yang, Y. Shi, Proc. Natl. Acad. Sci.
USA 2011, 108, 21004 – 21009.
[8] C. Pelzer, K. Cabalzar, A. Wolf, M. Gonzalez, G. Lenz, M.
Thome, Nat. Immunol. 2013, 14, 337 – 345.
[9] L. M. Staudt, Cold Spring Harbor Perspect. Biol. 2010, 2,
a000109.
[10] V. N. Ngo, R. E. Davis, L. Lamy, X. Yu, H. Zhao, G. Lenz, L. T.
Lam, S. Dave, L. Yang, J. Powell, L. M. Staudt, Nature 2006, 441,
106 – 110.
[11] a) A. G. Baxter, Nat. Rev. Immunol. 2007, 7, 904 – 912; b) C.
Mc Guire, P. Wieghofer, L. Elton, D. Muylaert, M. Prinz, R.
Beyaert, G. van Loo, J. Immunol. 2013, 190, 2896 – 2903; c) A.
Brꢂstle, D. Brenner, C. B. Knobbe, P. A. Lang, C. Virtanen,
B. M. Hershenfield, C. Reardon, S. M. Lacher, J. Ruland, P. S.
Ohashi, T. W. Mak, J. Clin. Invest. 2012, 122, 4698 – 4709.
[12] B. D. Trapp, K. A. Nave, Annu. Rev. Neurosci. 2008, 31, 247 –
269.
[13] D. Nagel, S. Spranger, M. Vincendeau, M. Grau, S. Raffegerst, B.
Kloo, D. Hlahla, M. Neuenschwander, J. P. von Kries, K. Hadian,
B. Dorken, P. Lenz, G. Lenz, D. J. Schendel, D. Krappmann,
Cancer Cell 2012, 22, 825 – 837.
[14] P. Seeman, T. Lee, Science 1975, 188, 1217 – 1219.
[15] S. A. Jortani, A. Poklis, J. Anal. Toxicol. 1993, 17, 374 – 377.
[16] E. Sachlos, R. M. Risueno, S. Laronde, Z. Shapovalova, J. H.
Lee, J. Russell, M. Malig, J. D. McNicol, A. Fiebig-Comyn, M.
Graham, M. Levadoux-Martin, J. B. Lee, A. O. Giacomelli, J. A.
Hassell, D. Fischer-Russell, M. R. Trus, R. Foley, B. Leber, A.
Xenocostas, E. D. Brown, T. J. Collins, M. Bhatia, Cell 2012, 149,
1284 – 1297.
Figure 4. Mepazine enantiomers have different MALT1 inhibitory
potentials. a,b) Tryptophan-quenching assay with the separated R and
S enantiomers of thioridazine and mepazine. c,d) MALT1 cleavage
assay with wt GST-MALT1325-760 after incubation with either the R or S
enantiomer of mepazine and thioridazine. c) Thioridazine enantiomers
have an equivalent MALT1 inhibitory potential. d) (S)-mepazine has
a higher MALT1 inhibitory potential than the R enantiomer.
a dopamine receptor antagonist.[16] Whereas side effects
associated with dopamine receptors may be tolerated to
some extent for treatment of high-grade MALT1-dependent
lymphoma, more selective MALT1 inhibitors are certainly
required for a therapeutic application in diseases associated
with autoimmunity or allergic inflammation. By identifying
the thioridazine and mepazine binding pocket on the new
target MALT1, we pave the way for medicinal chemistry to
develop effective MALT1 inhibitors. From the perspective of
structure-based drug design, the detailed analysis of the key
binding interactions in the co-crystal structure provides
important insights for further optimization of a phenothia-
zine-based MALT1 inhibitor. The identification of the
improved affinity of (S)-mepazine represents a significant
result towards this goal. Thus, our structural insights into the
noncompetitive, allosteric mode of MALT1 inhibition will be
of significance for the development of more effective and
selective drugs to treat MALT1-driven cancer or autoimmune
diseases.
Received: May 18, 2013
Published online: && &&, &&&&
Keywords: cancer · inhibitors · medicinal chemistry ·
.
multiple sclerosis · thioridazine
[17] The structure of MALT1Casp-Ig3 in complex with thioridazine is
deposited in the Protein Data Base under accession code 4I1R.
[1] M. Thome, Nat. Rev. Immunol. 2008, 8, 495 – 500.
[2] a) A. G. Uren, K. OꢁRourke, L. A. Aravind, M. T. Pisabarro, S.
Seshagiri, E. V. Koonin, V. M. Dixit, Mol. Cell 2000, 6, 961 – 967;
4
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
Communications
Medicinal Chemistry
Second site: In the crystal structure of
human MALT1casp-Ig3 (mucosa-associated
lymphoid tissue lymphoma translocation
protein 1) in complex with the tricyclic
phenothiazine derivative thioridazine
(violet in the picture), the inhibitor is
bound in a hydrophobic pocket far from
the active site. This explains the action of
phenothiazine derivatives as noncompe-
titive, reversible inhibitors.
F. Schlauderer, K. Lammens,* D. Nagel,
M. Vincendeau, A. C. Eitelhuber,
S. H. L. Verhelst, D. Kling, A. Chrusciel,
J. Ruland, D. Krappmann,
K.-P. Hopfner
&&&& — &&&&
Structural Analysis of Phenothiazine
Derivatives as Allosteric Inhibitors of the
MALT1 Paracaspase
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!