The Structure of Chagasin in Complex
with a Cysteine Protease Clarifies the Binding
Mode and Evolution of an Inhibitor Family
Stephanie X. Wang,1Kailash C. Pandey,2Julio Scharfstein,6James Whisstock,7Rick K. Huang,8
Jordan Jacobelli,1Robert J. Fletterick,3Philip J. Rosenthal,2Magnus Abrahamson,9Linda S. Brinen,4
Andrea Rossi,5Andrej Sali,5and James H. McKerrow1,*
1Department of Pathology
2Department of Medicine
3Department of Biochemistry and Biophysics
4Department of Cell and Molecular Pharmacology
5Department of Biopharmaceutical Sciences
University of California, San Francisco, San Francisco, CA 94143, USA
6Instituto de Biofı ´sica Carlos Chagas Filho, Universidade Federal do Rio de Janeiro, CEP 21990-400 Rio de Janeiro, Brazil
7Department of Biochemistry and Molecular Biology, Victorian Bioinformatics Consortium, School of Biomedical Sciences,
Monash University Clayton Campus, Wellington Road, Clayton, Victoria 3800, Australia
8Molecular and Cell Biology Undergraduate Program, University of California, Berkeley, Berkeley, CA 94720, USA
9Department of Clinical Chemistry, Institute of Laboratory Medicine, University of Lund, University Hospital,
S-221 85 Lund, Sweden
Protein inhibitors of proteolytic enzymes regu-
late proteolysisand preventthepathological ef-
fects of excess endogenous or exogenous pro-
teases. Cysteine proteases are a large family of
kingdoms. Disturbance of the equilibrium be-
tween cysteine proteases and natural inhibitors
is a key event in the pathogenesis of cancer,
rheumatoid arthritis, osteoporosis, and emphy-
sema. A family (I42) of cysteine protease in-
hibitors (http://merops.sanger.ac.uk) was dis-
covered in protozoan parasites and recently
found widely distributed in prokaryotes and eu-
karyotes. We report the 2.2 A˚crystal structure
of the signature member of the I42 family, cha-
gasin, in complex with a cysteine protease.
Chagasin has a unique variant of the immuno-
globulin fold with homology to human CD8a. In-
teractions of chagasin with a target protease
are reminiscent of the cystatin family inhibitors.
Protein inhibitors of cysteine proteases may
have evolved more than once on nonhomolo-
Cysteine proteases are a large and diverse family of en-
zymes found throughout the plant and animal kingdoms,
and represent the dominant protease family in inverte-
brates. Disturbance of the equilibrium between cysteine
proteases and their natural inhibitors is a key event in
the pathogenesis of cancer, rheumatoid arthritis, osteo-
porosis, and emphysema (Turk et al., 2003; Riese and
Chapman, 2000). Chagasin is a protease inhibitor that
was first identified in Trypanosoma cruzi as the physiolog-
ical regulator of cruzain (also known as cruzipain), the ma-
jor protease of this protozoan parasite (Monteiro et al.,
2001; Rigden et al., 2002; Sanderson et al., 2003). Cruzain
is a papain-like (Clan CA) cysteine protease that is ex-
pressed in all stages of the parasite life cycle. It is a key
virulence factor of T. cruzi, the infectious agent responsi-
ble for the leading cause of heart disease in Latin America,
Chagas disease (Scharfstein et al., 1986; Engel et al.,
1998). Chagasin is associated with cruzain during its traf-
ficking to specific compartments of the parasite cell, and
accumulated evidence suggests that the primary role of
chagasin is in posttranslational regulation of protease
activity (Monteiro et al., 2001). Following the discovery of
chagasin, homologous proteins were identified in numer-
ous other eukaryotic and prokaryotic organisms (Rigden
et al., 2002; Sanderson et al., 2003). In many cases, these
related protease inhibitors likely regulate cysteine prote-
asesproduced bytheircognate organism. However,in or-
ganisms such as the pathogenic bacterium Pseudomonas
encoding a cysteine protease target has been identified in
the genome (Sanderson et al., 2003). An alternative func-
tionfor thisfamily ofprotease inhibitorswas therefore pro-
posed: inhibiting the activity of host cysteine proteases
elaborated as part of the host defense against pathogens
(Sanderson et al., 2003). The amino acid sequence of cha-
gasin provides few clues to its function, as chagasin and
other I42 family inhibitors share no sequence homology
with any known protease inhibitors (Rigden et al., 2002).
Recently, NMR solution structures were solved for cha-
gasin (Salmon et al., 2006) and the homolog of chagasin in
Structure 15, 535–543, May 2007 ª2007 Elsevier Ltd All rights reserved 535
Leishmania mexicana (Smith et al., 2006). Predictions
were made about the possible mode of interaction be-
tween members of this new family of inhibitors and their
protease targets. But without a protease-inhibitor com-
plex, these remained speculative predictions. Solving of
an X-ray structure of chagasin in complex with a target
cysteine protease now allows clarification of the binding
mode of this novel inhibitor.
Confirmation that Recombinant Chagasin Inhibits
Target Proteases and Purification of the Chagasin-
Falcipain 2 Complex
Before embarking on crystallization and structure analysis
of the complex between chagasin and falcipain 2 (FP2),
we confirmed that chagasin was indeed a tight-binding
inhibitor of this parasite protease as well as homologous
a more potent inhibitor than the cystatins against five
cathepsin L-like proteases. As was the case with cystatin,
the Kiversus cathepsin B was substantially higher due to
the impediment of the occluding loop of cathepsin B to
itor leupeptin. Having confirmed tight binding of chagasin
to FP2, we then confirmed purification of that complex
for crystallography and subsequent structural analysis
General Characteristics of the Chagasin-Falcipain
2 Structure: Chagasin Is a New Variant
of the Immunoglobulin Fold
The coordinates and structure of chagasin have been de-
posited in the Protein Data Bank (PDB) under ID code
2OUL. The X-ray structure of chagasin is found to differ
from the structures of all known classes of protease inhib-
1988; Dubin et al., 2003; Guncar et al., 1999). Chagasin
adopts an immunoglobulin (Ig)-like b sandwich structure
(Figure 2B). Most Ig-like b sandwich domains are catego-
rized into four subgroups: constant-type, variable-type,
switched-type, and hybrid-type, as defined by the topo-
logical arrangement of the strands in the front and back
b sheets (Bork et al., 1994). The hypervariable loops con-
regions (CDRs) that contribute to the versatility of Ig-like
folds in numerous protein-protein interactions (Garcia
et al., 1998, 1999). In chagasin, the CDR equivalent loops
Table 1. Inhibition Constants for Chagasin versus Clan
CA (Papain) Cysteine Proteases
Protease CystatinChagasin Leupeptin
Falcipain 26.5 ± 1.41.7 ± 0.53 0.20 ± 0.11
Falcipain 3100 ± 8.6 0.62 ± 0.270.30 ± 0.09
Cathepsin B 101 ± 7.7100 ± 9.5 0.37 ± 0.14
Cathepsin L11.5 ± 3.6 0.35 ± 0.100.52 ± 0.22
Cathepsin K 25.4 ± 3.02.0 ± 0.28 0.64 ± 0.14
Cathepsin H 0.63 ± 0.2415 ± 4.83.2 ± 1.4
Figure 1. Stoichiometry Analysis of Cha-
gasin-FP2 by Size-Exclusion Chroma-
The standard curve based on molecular weight
standards and the elution volume for the
chagasin-FP2 complex and chagasin alone
are highlighted. The peak fraction containing
the putative complex was analyzed by SDS-
PAGE followed by silver staining. The apparent
molecular weights for the chagasin-FP2 com-
plex and chagasin alone were calculated as
46 kDa and 14 kDa, respectively. The molecu-
lar weight for FP2 is 28 kDa.
Structure of Chagasin
536 Structure 15, 535–543, May 2007 ª2007 Elsevier Ltd All rights reserved
are the BC, DE, and FG loops (Figures 3B and 3C). These
loops approximately correspond to the three most con-
servedmotifs defined inthe I42family ofinhibitors (Rigden
detailed in the current structure is likely to apply to the
entire family of inhibitors.
The immunoglobulin fold (i.e., the Ig-like fold) derives its
name from the core structure of the immunoglobulins
themselves, but it is also found in other immune system
proteins including interleukin 1, CD4, CD1, and CD8. The
elucidation of the structure of chagasin confirms that this
backbone has utility for protein-protein interactions that
is much broader. A query on DBAli (Marti-Renom et al.,
2001) (http://salilab.org/DBAli/), with no restriction on se-
quence identity, shows that there are 284 chains in the
ture of chagasin. Among the proteins identified with high
similarity to chagasin, 11 of the 31 are immune system
proteins. A structural core of four conserved b strands,
B, C, E, and F, constitute one of the signature features
of an Ig-like fold. This conserved core is found in all sub-
types of Ig domains, but on first analysis it appeared to
be absent in chagasin because the N-terminal strand
A in chagasin is positioned differently from known Ig-like
domains. If strand A were switched from its traditional
front-sheet position to the back sheet, the topology
arrangement of chagasin would be identical to that of
a typical c-type Ig domain (Figure 3A). We therefore clas-
sify the novel topology arrangement in chagasin as an N-
sc-type Ig domain (amiNo-terminal switched constant-
Comparison of Chagasin X-Ray
and Solution Structures
The Ig-like b sandwich structure of chagasin largely vali-
dates previous modeling predictions (Rigden et al.,
2001) and two solution structures (Salmon et al., 2006;
Smith et al., 2006). However, differences exist at the N ter-
minus and among the CDR-like loops. Sequence align-
also adopt similar Ig-like folds, and this is confirmed at
least for the NMR solution structure for L. mexicana ICP
(Salmon et al., 2006). Structural variations are expected
due to sequence inserts between strands D and E (Rigden
Figure 2. The Structure of Chagasin Bound to a Cysteine Protease Target, FP2
(A) Overall structure of chagasin-FP2 is shown with chagasin in red and FP2 in blue.
(B) Chagasin is shown in stereo view in the same orientation as in (A).
(C) Sequence and secondary structure alignment of chagasin and chagasin-like protease inhibitors (CCPI) from T. brucei and L. major versus human
CD8a. The secondary structure distributions for chagasin and CD8a are depicted with the sequence alignment.
Structure 15, 535–543, May 2007 ª2007 Elsevier Ltd All rights reserved 537
Structure of Chagasin
et al., 2002). For example, the chagasin ortholog from
Plasmodium falciparum contains an insert of 50 amino
acids between its motifs I and II that may introduce signif-
icant structural deviation, or even include an additional
Homology of Chagasin and CD8a
The chagasin X-ray structure confirms homology to the
X-ray structure of the T cell (thymus-derived lymphocyte)
surface protein CD8a (Leahy et al., 1992), a key element
in antigen activation of T cells by antigen presenting cells.
ment very similar to that of chagasin (Figures 3A and 3B).
Like CD8a, the b strands in chagasin are held together by
strong hydrophobic interactions. Residues such as Val,
Phe, Trp, Leu, and Ile constitute a solid hydrophobic
core that likely ensures the stability of chagasin in the ab-
sence of any disulfide bond (Figure 3C). More importantly,
the BC loop in chagasin (LPSNPTTGFAW) closely resem-
(LLSNPTSGCSW) (Figure 2C). What is most remarkable is
that, to date, the sequence motif found in the binding loop
of CD8a and chagasin has not been found in any of the
other immunoglobulin superfamily members, including
the 284 chains in the PDB with at least 50% of their resi-
dues structurally equivalent to those in chagasin.
An analysis of the binding of the chagasin BC loop to
a target protease, compared to the binding of CD8a to
a major histocompatibility complex (MHC), shows that
four of the seven conserved residues play very similar
roles (protein-protein interactions, flexibility or stabiliza-
tionof theloop, andstabilization of thecore).Itisnotewor-
thythatthisshortyet importantsequencehomology could
not be detected using sensitive search engines such as
PSI-BLAST but only became apparent after the structural
homology was revealed. The BC loop is one of the three
signature motifs of the I42 family of inhibitors, and it at
least partially accounts for the activity of the inhibitor
against target proteases (Figure 3C). A synthetic version
of the BC loop peptide in the chagasin-like protein from
Entamoeba histolytica specifically blocked the activity of
cysteine proteases that prefer Phe at the P1 site (Rieken-
berg et al., 2005). It is also noteworthy that Thr31 in the
BC loop of chagasin binds to the catalytic Cys at the
FP2 active site through water-mediated hydrogen bonds
(Figure 4A). This highly conserved Thr likely serves as
Figure 3. Comparison of Chagasin and CD8a
(A and B) Depiction of the topology arrangement of chagasin and CD8a, respectively. Both molecules are in a rainbow color scheme to reveal their
b strands, with N termini in dark blue and C termini in bright red. Both molecules are oriented to reveal the conserved F, C, B, and E core strands. All
structure figures were generated with PyMOL (http://www.pymol.org/; Delano Scientific LLC, South San Francisco, CA).
(C) The hydrophobic core of the chagasin structure is depicted. Selected residues involved in hydrophobic packing are shown with side chains which
are highlighted by electron density at 1 (s) from the 3fo2fc omit map.
538 Structure 15, 535–543, May 2007 ª2007 Elsevier Ltd All rights reserved
Structure of Chagasin
a key functional residue among all chagasin-like inhibitors
(identified to date) (Figure 2C).
How Chagasin Binds to a Target Protease
Models of the binding loops and predictions of interaction
of these loops with target proteases were made based on
two solution structures, one of chagasin itself (Salmon
(Smith et al., 2006). As expected from the observed flexi-
bility of these loops in solution, the modeled structures
differ from those determined here by X-ray analysis. Nev-
ertheless, the BC loop, the highly mobile DE loop, and the
RPW/F motif in the FG loop are all confirmed as key ele-
ments for binding. The previous report of the lack of inhib-
itor activity by a mutant form of the L. mexicana ICP high-
lights the importance of the GXG motif in the DE loop
(Smith et al., 2006). However, as shown in the X-ray anal-
ysis, a wedge-like interaction with the target by all three
loops is key to binding and, by inference, full inhibition
activity of chagasin.
Despite its clear homology to the LLSNPTSGCSW motif
in human CD8a, the BC loop in chagasin appears to func-
tion via a different mechanism from what has been estab-
lished for the CDR loops. Chagasin demonstrates a strict
1:1 binding to a target protease, FP2, in structure (Fig-
ure 4A) and in gel-filtration analysis (Figure 1). Both human
and mouse CD8a form dimers that bind MHCs (Garcia
et al., 1996; Gao et al., 1997; Liu et al., 2003) (Figure 4B).
In CD8a, these interactions rely upon the CDR loops and
include the BC loop motif from each CD8a monomer
(Gao et al., 1997; Liu et al., 2003). By definition, the CDR
loops adopt multiple conformations to complement anti-
gen binding; thus, it is no surprise that the BC loop in cha-
gasin and the equivalent loops in human and mouse CD8a
have different structural features (Figures 4A and 4B).
chagasin does not compete with binding of CD8a to an
MHC by biacore analysis (H. Cheroutre, personal commu-
nication) or in a T cell activation assay.
The structure of chagasin was determined bound to the
cathepsin L-like cysteine protease FP2 (Figure 2A). The
original FP2 structure was determined in a cystatin-FP2
Figure 4. Key Binding Interactions between Chagasin and a Target Protease Compared to CD8a and HLA
Binding interactions between (A) theBCloopand FP2arecomparedwiththosebetween (B)human CD8aand HLA(Protein DataBank IDcode1AKJ).
The protease and the MHC are rendered by surface presentation, while the BC loop in chagasin and the equivalent in CD8a are highlighted to reveal
their conformational differences. The BC loop in the side view (A) and the BC-like loop 1 in (B) are oriented similarly.
Structure 15, 535–543, May 2007 ª2007 Elsevier Ltd All rights reserved 539
Structure of Chagasin
complex (Wang et al., 2006). Overall, the structural differ-
ence between chagasin-bound FP2 and cystatin-bound
FP2 is minimal, with a root-mean-square deviation (rmsd)
around 0.5 A˚when compared by all 241 Ca atoms, sug-
gesting that both inhibitors interact with a consistently
The protease binding loops (BC, DE, and FG) in chaga-
sin form a well-aligned wedge that fills the active site
groove of target cysteine proteases to obstruct substrate
binding. Thetripartite binding in chagasin isreminiscent of
that found in the cystatins (Bode et al., 1988; Wang et al.,
2006). Similar interactions are found between human ca-
thepsin L and the MHC class II-associated p41 Ii fragment
(Guncar et al., 1999). Kinetic analysis confirms the advan-
tage of tripartite binding. Chagasin inhibits both papain
and cruzain with picomolar affinity (Monteiro et al., 2001).
It also inhibits the malarial cysteine proteases FP2 and
falcipain 3 (FP3), as well as the human cathepsins B, H,
K, and L (Table 1).
Chagasin, cystatins, and p41 inhibitory fragment come
from protist, vertebrate, and human sources, respectively.
They share no overall sequence homology and contain
no conserved binding motifs. Nevertheless, they have
evolved to bind to the same class of enzymes via remark-
ably similar binding interactions (Figures 5A and 5B). The
tripartite mode of inhibition effectively blocks enzymatic
activities in papain-like cysteine proteases and provides
unique advantages that may have driven such evolution-
ary convergence. First, all tripartite binding results in
very large protein-protease interaction surfaces that often
correspond to strong binding energy. All known tripartite
inhibitors bind to target cysteine proteases over buried
surface areas larger than 2000 A˚2(Figure 5A), whereas
most other enzyme-inhibitor complexes only interact
over surfaces averaging approximately 950 A˚2in size
(Jones and Thornton, 1996). Second, the three protease
contact sites are discontinuous in amino acid sequence
and relatively independent from each other (Figure 5B).
This organization, which is reminiscent of the versatility
of antibodies in antigen binding, enables chagasin to
adapt to the active sites of different cysteine proteases
and likely accounts for its broad inhibitory activities. Third,
it is intriguing that such multiloop binding interactions
at the protease active site are only found between pa-
pain-like enzymes and their inhibitors (Stubbs et al.,
1990; Jenko et al., 2003; Guncar et al., 1999; Wang et al.,
2006). By comparison to other protease classes, papain-
like cysteine proteases harbor many characteristics that
facilitate protein-protein interactions, including an acces-
sible active site, a relatively flat substrate binding cleft,
and the predominantly hydrophobic nature of the binding
surface (Jones and Thornton, 1996; DeLano et al., 2000).
The convergent evolution of chagasin, p41 inhibitory
fragment, and cystatin reflects the importance of post-
as ancient as protists and as complex as primates. In both
monality of function and location in that they interact with
their target proteases in protein-trafficking pathways be-
tween the Golgi (chagasin) and the lysosome (cystatins).
While the chagasin family proteins commonly function to
inhibit endogenous papain-like cysteine proteases (San-
tos et al., 2005), there are clearly examples of organisms
without such enzymes in which a chagasin family inhibitor
is produced (Pseudomonas aeruginosa and Thermobifida
fusca, for example). In situations where a pathogen is in-
volved, Sanderson et al. (2003) suggested that the chaga-
sin homolog may function to protect the organism from
Figure 5. Comparison of Chagasin Target Interactions to that of a Cystatin Inhibitor and the p41 Inhibitory Fragment
(A) Thecomplexesof p41-catL(ProteinDataBankID code1ICF;p41ingreen andcatL inlight yellow),cystatin-FP2 (Protein DataBank IDcode1YVB;
wheat) are superimposed by the 110 most conserved residues in the protease domains.
(B) Front and side views of the superimposed inhibitors.
540 Structure 15, 535–543, May 2007 ª2007 Elsevier Ltd All rights reserved
Structure of Chagasin
host proteases elaborated as part of an innate immune re-
sponse or other defense mechanism. Nevertheless, there
are also examples of chagasin family inhibitors being ex-
pressed by nonpathogenic organisms such as the archae
Functional convergence may account for the binding of
It is more challenging to explain why distant proteins such
as chagasin and CD8a share structural and sequence ho-
mology. It is possible that they share a common ancestor
but have diverged in function. The highly homologous
loops evolved to meet different functional requirements
in protease binding and antigen presentation. An alter-
native hypothesis has also been proposed, invoking a
horizontal gene transfer mechanism from human to the
porting nucleotide sequence data (Rigden et al., 2001,
2002). Alternatively, the discovery of homology between
chagasin and CD8a suggests that a very effective and
commonly used scaffold for protein binding, the Ig-like
fold, arose in ancient eukaryotes. Because of the physical
principles that hold b sheets within the Ig-like fold, the
overall scaffold can remain stable yet diversify in the loop
structures to provide novel biologic functions withspecific
binding partners. In the case of CD8a and chagasin, this
led to a very similar motif presented by two different loops
but interacting with the target ligand using four of seven
similar amino acid side-chain modes. Protein-protein in-
teractions using this scaffold likely diverged into distinct
ing cofactors) prior to or concurrent with the evolution of
the adaptive immune system in vertebrates.
of this new inhibitor family is now clarified by X-ray struc-
ture. Predictions made from solution structures and muta-
aspects of the molecular evolution of cysteine protease
inhibitors are also mirrored in the structure. First, the ho-
mology between chagasin and CD8a suggests that a very
effective and commonly used scaffold for protein binding
arose in agent eukaryotes, but diverged into distinct enti-
ties (inhibitors versus binding cofactors) with the evolution
gent evolution of similar target binding by chagasin, p41
inhibitory fragment, and cystatin.
Cloning, Expression, Purification, Gel Filtration, and Kinetic
Analysis of Chagasin
Two primers (50-CTTAAAATCGGATCCCACAAGGTGACGAAAGCCC
GTGAA-30) were used to amplify the chagasin gene from T. cruzi chro-
mosomal DNA. It was subsequently inserted between the BamHI and
HindIII restriction sites in the pQE30 vector (QIAGEN). Upon transfor-
mation into m15 (pREP4) cells, the 12 kDa chagasin was expressed
as an insoluble protein at 37?C, and was partially solubilized by slower
cell growth in LB at 15?C. Soluble chagasin was then purified by an
N-terminal His tag on an Ni-NTA column (QIAGEN) preequilibrated
with 25 mM imidazole, 50 mM phosphate (pH 7.0), and 200 mM
in the presence of equilibrium buffer.
The apparent Kiof chagasin against FP2, FP3, and cathepsin B, H,
K, and L was determined by standard protocols (Wang et al., 2006).
Purification and Crystallization of the Chagasin-FP2 Complex
Purified and activated FP2 was incubated with recombinant chagasin
at 4?C and further purified as previously described (Wang et al., 2006).
The purified chagasin-FP2 complex was dialyzed into 50 mM phos-
phate buffer (pH 7.0) and 200 mM NaCl and concentrated to 6 mg/
ml. The complex was subsequently crystallized by the sitting drop va-
por diffusion method by mixing 0.8 ml of chagasin-FP2 with 0.8 ml of
well solution at 4?C in the presence of 20% (v/v) PEG300, 0.1 M Tris
(pH 8.5), 5% (w/v) PEG8000, and 10%–15% (v/v) glycerol. Molecular
weight standards (Bio-Rad) and a preincubated mixture of chagasin-
FP2 were analyzed by size-exclusion chromatography on a Superose
12column (Pharmacia)in 50 mMBis-Tris (pH 5.6) with 200mM NaCl at
a flow rate of 0.5 ml/min.
Data Collection and Structure Determination
of the Chagasin-FP2 Complex
oprotection against synchrotron radiation. Complete data sets of cha-
gasin-FP2 were collected at 100K on beamline 8.3.1 at the Advanced
Light Source (ALS), using a CCD camera. The data sets were indexed
and integrated using DENZO/SCALEPACK (Otwinowski and Minor,
1997) to yield a data set 96.7% complete at 2.2 A˚(Rmerge= 10.8%).
The chagasin-FP2 crystals belong to the space group P43212, with
cell dimensions a = b = 94.236 A˚, c = 119.764 A˚, and a = b = g =
90?. Each asymmetric unit contains only one chagasin-FP2 complex.
Data up to 3.5 A˚were included in the molecular replacement search
with an FP2 model (PDB ID code 1YVB), using the rotational and trans-
lational functions from the Crystallography & NMR System software
suites (CNS 1.1) (Brunger et al., 1998). Following a rigid body refine-
ment, the FP2 model gave an initial Rworkingof 42.9%. Chagasin was
built into the extra electron densities by de novo building using
QUANTA 2000 (Molecular Simulations). The chagasin-FP2 structure
was refined by alternate cycles of energy minimization, simulated an-
nealing, and group B factor refinements in the CNS suites. Model
building and fitting were done using QUANTA 2000. The final chaga-
sin-FP2 structure has been refined to 2.2 A˚. Complete data and refine-
ment statistics are listed in Table 2.
Sequence and Structure Homology Analyses
Various segments of the chagasin amino acid sequence were used to
search against the general nonredundant database as well as the PDB
database using the NCBI BLAST site for possible protein homologs.
Results from the first search include similar protease inhibitors from
other parasitic organisms, for example T. brucei and P. falciparum.
Although a search against the PDB database identified scores of pro-
teins from the immune system, regardless of which segment was used
to search the PDB database, CD8a was always identified as one of the
top hits. Alignment of the chagasin and CD8a sequences revealed that
CD8a harbors a short sequence motif that is highly similar to motif I in
The structure of chagasin was searched against the DABli database
(http://www.salilab.org/DBAli/) to identify all known structures with
significant structural similarity to chagasin (i.e., with p value > 8.0).
DBAli contains 1,086,905,585 pairwise structural alignments and fam-
ily-based multiple structure alignments for 26,950 nonredundant
chains in the PDB. Three of the six matches are classified as immuno-
globulins in SCOP (http://scop.mrc-lmb.cam.ac.uk/scop/; including
Structure 15, 535–543, May 2007 ª2007 Elsevier Ltd All rights reserved 541
Structure of Chagasin
the highest significance hit): one belongs to the PaPD superfamily of
the Ig-like fold, and the remaining two are classified as non-Ig-like
but b sheet-containing folds. Therefore, the available evidence sug-
gests that chagasin belongs to the immunoglobulin superfamily with
a very ancient divergence.
T Cell Cytotoxicity Assay
Cytotoxic activity was measured in a standard51Cr release assay. Ac-
tivated CD8+T cells from OT-I transgenic mice were used as effectors,
generatedasin Krummel et al.(1999),and LB27.4Bcells wereused as
targets for cytotoxicity. Target B cells were labeled with 50 milliCi of
51Cr per 1 3 106cells for 1 hr. One milligram/milliliter of SIINFEKL cog-
nate ovalbumin-derived peptide (pOVA) was added to the target
centrations of chagasin protein were diluted in the wells for the cyto-
toxicity assay. In addition, one group of target cells was preincubated
with 50 mg/ml of chagasin protein for 1 hr. Various dilutions of OT-I
effector cells were incubated with 1 3 104 51Cr-labeled target B cells
in round-bottom 96-well plates for 4 hr at 37?C. Subsequently,
100 ml of supernatant was removed and counted on a scintillation
counter to determine experimental51Cr release (ER). Spontaneous
release (SR) was determined using the target cells alone, while maxi-
mum release (MR) was determined by lysing target cells with 1%
SDS and measuring the amount of51Cr release. Specific lysis (in %)
was calculated as follows: (ER ? SR)/(MR ? SR)*100. All samples
were set up in triplicate.
This work was supported by NIH grants AI35707 and AI35800, the
Sandler Family Supporting Foundation, and the Medicines for Malaria
Venture. S.X.W. was a recipient of a postdoctoral fellowship from the
American Heart Association.
Received: November 21, 2006
Revised: March 1, 2007
Accepted: March 20, 2007
Published: May 15, 2007
Bode, W., Engh, R., Musil, D., Thiele, U., Huber, R., Karshikov, A.,
Brzin, J., Kos, J., and Turk, V. (1988). The 2.0 A˚X-ray crystal structure
of chicken egg white cystatin and its possible mode of interaction with
cysteine proteinases. EMBO J. 7, 2593–2599.
Bork, P., Holm, L., and Sander, C. (1994). The immunoglobulin fold.
Structural classification, sequence patterns and common core. J.
Mol. Biol. 242, 309–320.
Brunger, A.T., Adams, P.D., Clore, G.M., DeLano, W.L., Gros, P.,
Grosse-Kunstleve, R.W., Jiang, J.S., Kuszewski, J., Nilges, M., Pannu,
N.S., et al. (1998). Crystallography & NMR System: a new software
suite for macromolecular structure determination. Acta Crystallogr. D
Biol. Crystallogr. 54, 905–921.
DeLano, W.L., Ultsch, M.H., de Vos, A.M., and Wells, J.A. (2000). Con-
vergent solutions to binding at a protein-protein interface. Science
Dubin, G., Krajewski, M., Popowicz, G., Stec-Niemczyk, J., Bochtler,
teine protease inhibitors: solution structure of staphostatin A from
Staphylococcus aureus. Biochemistry 42, 13449–13456.
Engel, J.C., Doyle, P.S., Hsieh, I., and McKerrow, J.H. (1998). Cysteine
protease inhibitors cure an experimental Trypanosoma cruzi infection.
J. Exp. Med. 188, 725–734.
Gao, G.F., Tormo, J., Gerth, U.C., Wyer, J.R., McMichael, A.J., Stuart,
D.I., Bell, J.I., Jones, E.Y., and Jakobsen, B.K. (1997). Crystal structure
of the complex between human CD8a (a) and HLA-A2. Nature 387,
Garcia, K.C., Scott, C.A., Brunmark, A., Carbone, F.R., Peterson, P.A.,
Wilson, I.A., and Teyton, L. (1996). CD8 enhances formation of stable
T-cell receptor/MHC class I molecule complexes. Nature 384, 577–
Garcia, K.C., Degano, M., Pease, L.R., Huang, M., Peterson, P.A.,
Teyton, L., and Wilson, I.A. (1998). Structural basis of plasticity in
T cell receptor recognition of a self peptide-MHC antigen. Science
Garcia, K.C., Teyton, L., and Wilson, I.A. (1999). Structural basis of
T cell recognition. Annu. Rev. Immunol. 17, 369–397.
Guncar, G., Pungercic, G., Klemencic, I., Turk, V., and Turk, D. (1999).
Crystal structure of MHC class II-associated p41 Ii fragment bound
to cathepsin L reveals the structural basis for differentiation between
cathepsins L and S. EMBO J. 18, 793–803.
Jenko, S., Dolenc, I., Guncar, G., Dobersek, A., Podobnik, M., and
Turk, D. (2003). Crystal structure of Stefin A in complex with cathepsin
H: N-terminal residues of inhibitors can adapt to the active sites of
endo- and exopeptidases. J. Mol. Biol. 326, 875–885.
Jones, S., and Thornton, J.M. (1996). Principles of protein-protein
interactions. Proc. Natl. Acad. Sci. USA 93, 13–20.
Kleywegt, G.J., and Brunger, A.T. (1996). Checking your imagination:
applications of the free R value. Structure 4, 897–904.
Krummel, M.F., Heath, W.R., and Allison, J. (1999). Differential cou-
pling of second signals for cytotoxicity and proliferation in CD8+
Table 2. Diffraction Data Processing and Refinement
Cell parameters (A˚)
a = b94.236
a = b = g (?)
Data resolution (A˚)
Unique reflections 27,127
Completeness (%)96.7 (80.2)a
Average B factor (A˚2)
Rmsd bond (A˚)0.009
Rmsd angle (?) 1.6
aStatistics for the highest-resolution shell.
bRmerge=Pj(I – <I>)j/P(I).
dRfree: crossvalidation R calculated by omitting 5% of the re-
flections (Kleywegt and Brunger, 1996).
h,k,l(jFobs(h,k,l)j – kjFcalc(h,k,l)j)/P
542 Structure 15, 535–543, May 2007 ª2007 Elsevier Ltd All rights reserved
Structure of Chagasin
T cell effectors: amplification of the lytic potential by B7. J. Immunol.
Leahy, D.J., Axel, R., and Hendrickson, W.A. (1992). Crystal structure
of a soluble form of the human T cell coreceptor CD8 at 2.6 A˚resolu-
tion. Cell 68, 1145–1162.
Liu, Y., Xiong, Y., Naidenko, O.V., Liu, J.H., Zhang, R., Joachimiak, A.,
Kronenberg, M., Cheroutre, H., Reinherz, E.L., and Wang, J.H. (2003).
The crystal structure of a TL/CD8aa complex at 2.1 A˚resolution: impli-
cations for modulation of T cell activation and memory. Immunity 18,
Marti-Renom, M.A., Ilyin, V.A., and Sali, A. (2001). DBAli: a database of
protein structure alignments. Bioinformatics 17, 746–747.
Monteiro, A.C., Abrahamson, M., Lima, A.P., Vannier-Santos, M.A.,
and Scharfstein, J. (2001). Identification, characterization and localiza-
tion of chagasin, a tight-binding cysteine protease inhibitor inTrypano-
soma cruzi. J. Cell Sci. 114, 3933–3942.
Otwinowski, Z., and Minor, W. (1997). Processing of X-Ray Diffraction
Data Collected in Oscillation Mode, Volume 276 (New York: Academic
Riekenberg, S., Witjes, B., Saric, M., Bruchhaus, I., and Scholze, H.
(2005). Identification of EhICP1, a chagasin-like cysteine protease
inhibitor of Entamoeba histolytica. FEBS Lett. 579, 1573–1578.
Riese, R.J., and Chapman, H.A. (2000). Cathepsins and compartmen-
talization in antigen presentation. Curr. Opin. Immunol. 12, 107–113.
Rigden, D.J., Monteiro, A.C., and Grossi de Sa, M.F. (2001). The pro-
tease inhibitorchagasin of Trypanosoma cruzi adopts animmunoglob-
ulin-type fold and may have arisen by horizontal gene transfer. FEBS
Lett. 504, 41–44.
Rigden,D.J.,Mosolov,V.V.,and Galperin, M.Y.(2002). Sequence con-
servation in the chagasin family suggests a common trend in cysteine
proteinase binding by unrelated protein inhibitors. Protein Sci. 11,
Salmon, D., do Aido-Machado, R., Diehl, A., Leidert, M., Schmetzer,
O., de A. Lima, A.P.C., Scharfstein, J., Oschkinat, H., and Pires, J.R.
(2006). Solution structure and backbone dynamics of the Trypano-
soma cruzi cysteine protease inhibitor chagasin. J. Mol. Biol. 357,
Sanderson, S.J., Westrop, G.D., Scharfstein, J., Mottram, J.C., and
Coombs, G.H. (2003). Functional conservation of a natural cysteine
peptidase inhibitor in protozoan and bacterial pathogens. FEBS Lett.
Santos, C.C., Sant’anna, C., Terres, A., Cunha-e-Silva, N.L., Scharf-
stein, J., and de A. Lima, A.P.C. (2005). Chagasin, the endogenous
cysteine-protease inhibitor of Trypanosoma cruzi, modulates parasite
differentiation and invasion of mammalian cells. J. Cell Sci. 118, 901–
Scharfstein, J., Schechter, M., Senna, M., Peralta, J.M., Mendonca-
Previato, L., and Miles, M.A. (1986). Trypanosoma cruzi: characteriza-
tion and isolation of a 57/51,000 m.w. surface glycoprotein (GP57/51)
expressed by epimastigotes and bloodstream trypomastigotes.
J. Immunol. 137, 1336–1341.
Smith, B.O., Picken, N.C., Westrop, G.D., Bromek, K., Mottram, J.C.,
and Coombs, G.H. (2006). The structure of Leishmania mexicana ICP
provides evidence for convergent evolution of cysteine peptidase
inhibitors. J. Biol. Chem. 281, 5821–5828.
Stubbs, M.T., Laber, B., Bode, W., Huber, R., Jerala, R., Lenarcic, B.,
and Turk, V. (1990). The refined 2.4 A˚X-ray crystal structure of re-
combinant human stefin B in complex with the cysteine proteinase
papain: a novel type of proteinase inhibitor interaction. EMBO J. 9,
Turk, D., Turk, B., and Turk, V. (2003). Papain-like lysosomal cysteine
proteases and their inhibitors: drug discovery targets? Biochem. Soc.
Symp. (70), 15–30.
Wang, S.X., Pandey, K.C., Somoza, J.R., Sijwali, P.S., Kortemme, T.,
Brinen, L.S., Fletterick, R.J., Rosenthal, P.J., and McKerrow, J.H.
(2006). Structural basis for unique mechanisms of folding and hemo-
globin binding by a malarial protease. Proc. Natl. Acad. Sci. USA
The coordinates and structure of chagasin have been deposited in the
Protein Data Bank under ID code 2OUL.
Structure 15, 535–543, May 2007 ª2007 Elsevier Ltd All rights reserved 543
Structure of Chagasin