Structural basis for the blockage of IL-2 signaling by therapeutic antibody basiliximab.
ABSTRACT IL-2 signaling plays a central role in the initiation and activation of immune responses. Correspondingly, blockage of this pathway leads to inhibition of the immune system and would provide some therapeutic benefits. Basiliximab (Simulect), a therapeutic mAb drug with specificity against IL-2R alpha of T cells, was approved by U.S. Food and Drug Administration in 1998. It has been proven to be effective in the suppression of the IL-2 pathway and hence has been widely used to prevent allograft rejection in organ transplantation, especially in kidney transplants. In this study, we report the crystal structure of the basiliximab Fab in complex with the ectodomain of IL-2R alpha at 2.9 A resolution. In the complex structure, the Fab interacts with IL-2R alpha with extensive hydrophobic and hydrophilic interactions, accounting for a high binding affinity of 0.14 nM. The Ag binding site of basiliximab consists of all six CDR loops that form a large binding interface with a central shallow hydrophobic groove surrounded by four hydrophilic patches. The discontinuous epitope is composed of several segments from the D1 domain and a minor segment from the D2 domain that overlap with most of the regions responsible for the interactions with IL-2. Thus, basiliximab binding can completely block the interactions of IL-2 with IL-2R alpha and hence inhibit the activation of the IL-2 signal pathway. The structural results also provide important implications for the development of improved and new IL-2R alpha-targeted mAb drugs.
- SourceAvailable from: Chen Zhong[Show abstract] [Hide abstract]
ABSTRACT: Interleukin-2 (IL)-2 signaling plays a pivotal role in the activation of immune responses, and drugs that block this pathway have been shown to be effective for the immunosuppression in patients with organ transplantation to alleviate/eliminate allograft rejection. The first humanized monoclonal antibody (mAb) daclizumab falls into this category and shows high specificity and affinity against a key component of the IL-2 receptor complex, namely IL-2Rα. To reveal the molecular mechanism of the inhibition of the IL-2 signaling pathway by daclizumab, we determined the crystal structures of the daclizumab Fab in free form and in complex with the IL-2Rα ectodomain at 2.6 and 2.8 Å resolution, respectively. The daclizumab Fab adopts a similar conformation in the presence or absence of the IL-2Rα ectodomain. The antigen-binding site of daclizumab is mainly composed of five complementarity determining regions (CDRs) that form a large positively charged surface depression and two flanking patches that are generally hydrophobic. The conformational epitope consists of several discontinuous segments of the IL-2Rα ectodomain, a large portion of which overlaps with the regions that interact with IL-2, suggesting that the binding of daclizumab to IL-2Rα would prevent the IL-2 binding to IL-2Rα and the subsequent formation of the IL-2/IL-2Rαβγ(c) complex, and therefore block the IL-2 signaling pathway. These results also have implications for the design and development of improved mAb drugs targeting IL-2Rα.Cell Research 12/2010; 20(12):1361-71. · 10.53 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: A number of different methods are commonly used to map the fine details of the interaction between an antigen and antibody. Undoubtedly the method that is now most commonly used to give details at the level of individual amino acids and atoms is X-ray crystallography. The feasibility of undertaking crystallographic studies has increased over recent years due to the introduction of automation, miniaturisation and high throughput processes. However this still requires a high level of sophistication and expense and cannot be used when the antigen is not amenable to crystallisation. NMR spectroscopy offers a similar level of detail to crystallography but the technical hurdles are even higher such that it is rarely used in this context. Mutagenesis of either antigen or antibody offers the potential to give information at the amino acid level but suffers from the uncertainty of not knowing whether an effect is direct or indirect due to an effect on the folding of a protein. Other methods such as hydrogen deuterium exchange coupled to mass spectrometry (HDX-MS) and the use of short peptides coupled with ELISA based approaches tend to give mapping information over a peptide region rather than at the level of individual amino acids. It is quite common to use more than one method because of the limitations and even with a crystal structure it can be useful to use mutagenesis to tease apart the contribution of individual amino acids to binding affinity. This article is protected by copyright. All rights reserved.Immunology 03/2014; · 3.71 Impact Factor
- Transplantation 04/2011; 91(8):e59-61. · 3.78 Impact Factor
The Journal of Immunology
Structural Basis for the Blockage of IL-2 Signaling by
Therapeutic Antibody Basiliximab
Jiamu Du,* Hui Yang,*,†Dapeng Zhang,‡Jianchuan Wang,*,†Huaizu Guo,‡Baozhen Peng,*
Yajun Guo,‡and Jianping Ding*
IL-2 signaling plays a central role in the initiation and activation of immune responses. Correspondingly, blockage of this pathway
leads to inhibition of the immune system and would provide some therapeutic benefits. Basiliximab (Simulect), a therapeutic mAb
drug with specificity against IL-2Ra of T cells, was approved by U.S. Food and Drug Administration in 1998. It has been proven to
be effective in the suppression of the IL-2 pathway and hence has been widely used to prevent allograft rejection in organ
transplantation, especially in kidney transplants. In this study, we report the crystal structure of the basiliximab Fab in complex
with the ectodomain of IL-2Ra at 2.9 A˚resolution. In the complex structure, the Fab interacts with IL-2Ra with extensive
hydrophobic and hydrophilic interactions, accounting for a high binding affinity of 0.14 nM. The Ag binding site of basiliximab
consists of all six CDR loops that form a large binding interface with a central shallow hydrophobic groove surrounded by four
hydrophilic patches. The discontinuous epitope is composed of several segments from the D1 domain and a minor segment from
the D2 domain that overlap with most of the regions responsible for the interactions with IL-2. Thus, basiliximab binding can
completely block the interactions of IL-2 with IL-2Ra and hence inhibit the activation of the IL-2 signal pathway. The structural
results also provide important implications for the development of improved and new IL-2Ra–targeted mAb drugs.
of Immunology, 2010, 184: 1361–1368.
ing the defense against pathogens, recognition and binding of the
foreign Ags by the TCRs stimulate both the secretion of IL-2 and
the expression of IL-2Rs on the T cell surface. Subsequently, the
IL-2/IL-2R interaction activates the intracellular Ras/Raf/MAPK,
JAK/STAT, and PI3K/AKT signal pathways and ultimately stim-
ulates the growth, differentiation, and survival of the Ag-selected
cytotoxic T cells (9–14). IL-2 also regulates the functions of B
cells, NK cells, and regulatory T cells (6, 7).
The biological functions of IL-2 depend on its interactions with
three IL-2Rs to form a quaternary complexIL-2/IL-2Rab common
g chain (gc) to trigger the IL-2 signaling process. Among the re-
ceptors, IL-2Ra (p55, Tac Ag, or CD25) is an IL-2–specific re-
nterleukin-2 is the first cytokine to be identified, character-
ized, purified, and cloned (1–5). It plays a pivotal role in
immune responses against pathogenic infection (6–8). Dur-
ceptor, IL-2Rb (p75 or CD122) is shared with IL-15 (15–18), and
the gc(p65 or CD132) is a common receptor shared by many cy-
and IL-2Rb contribute to the rapid association and slow dissocia-
tion of IL-2, respectively (16), whereas receptors b and g mediate
the transmembrane signal transduction (21, 22). Structural studies
is composed of two b-strand–swapped “sushi-like” domains, and
both IL-2Rb and gcare composed of two fibronectin type III do-
mains (24–26). In the structures of the IL-2/IL-2Rabgcquaternary
complexes, IL-2Ra binds to a surface groove of IL-2 and makes
residues of the D2 domain, and IL-2Rb and gcbind to IL-2 on the
but no contact with IL-2Ra (24–26). The gcreceptor alone has no
detectable affinity to IL-2, and binding of gcto IL-2 needs the
presence of IL-2Rb (27). It is inferred that binding of IL-2Ra to
together form a composite binding site for gc. The structural, bio-
chemical, and computational data together suggest a sequential
binding scenario of IL-2 by its receptors: first IL-2Ra, which is
abundantly expressed on the T cell surface, captures and enriches
the secreted IL-2 and changes the conformation of IL-2 to a favor-
able IL-2Rb binding state, then the formed IL-2/IL-2Ra complex
approaches IL-2Rb through two-dimensional cell surface diffusion
to form the IL-2/IL-2Rab complex, and finally gcis recruited to
form the biologically active IL-2/IL-2Rabgccomplex to transduce
the signaling cascade (25–28).
immune system (7). The activation of T cells through the IL-2 sig-
naling is initiated by the binding of IL-2Ra to IL-2. IL-2Ra is
a specific receptor for IL-2, whereas IL-2Rb and gcare shared by
B cells but abundantly expressed on activated T cells, especially by
*State Key Laboratory of Molecular Biology and Research Center for Structural
Biology, Institute of Biochemistry and Cell Biology, Shanghai Institutes for Biolog-
ical Sciences, Chinese Academy of Sciences;†Graduate School of Chinese Academy
of Sciences; and‡International Joint Cancer Institute, Second Military Medical Uni-
versity, Shanghai, China
Received for publication September 28, 2009. Accepted for publication November
This work was supported by grants from the Ministry of Science and Technology of
China (2004CB720102, 2006AA02A313, and 2007CB914302), the National Natural
Science Foundation of China (30730028 and 90713046), the Chinese Academy of
Sciences (KSCX2-YW-R-107 and SIBS2008002), and the Science and Technology
Commission of Shanghai Municipality (07XD14032).
The coordinates and structure factors presented in this article have been deposited in
the Research Collaboratory for Structural Bioinformatics Protein Data Bank under
accession number 3IU3.
Address correspondence and reprint requests to Dr. Jianping Ding, Institute of Bio-
chemistry and Cell Biology, Shanghai Institutes for Biological Sciences, Chinese
Academy of Sciences, 320 Yue-Yang Road, Shanghai 200031, China. E-mail address:
Abbreviations used in this paper: gc, common g chain; PDB, Protein Data Bank.
the T cells participating in some pathological conditions such as
organ allograft rejection (29, 30), some autoimmune diseases (31,
32), and T cell leukemia (31, 33). The critical role of IL-2Ra in the
IL-2 signal pathway and its specific expression pattern make it
interfere with the interactions between IL-2 and IL-2R and hence
inhibit the IL-2 signal pathway, resulting in suppression of the im-
mune system, which provides clinical benefits to organ trans-
plantation patients. Two mAb drugs against IL-2Ra, basiliximab
(Simulect; Novartis Pharmaceuticals, East Hanover, NJ) and dacli-
zumab (Zenapax; Roche, Basel, Switzerland), have been approved
by the U.S. Food and Drug Administration for the prevention of
allograft rejection in organ transplantation, especially in kidney
transplants. Basiliximab is a mouse‑human chimeric mAb with the
domains of human IgG1(k) (34) that has had great success in the
prevention of renal allograft rejection (35). This mAb binds spe-
cifically to the ectodomain of IL-2Ra. With the phage display
method, the epitope recognized by basiliximab was mapped to
residues 116–122 of the D2 domain of IL-2Ra, which is part of the
region interacting with IL-2 and thus explains in part why the
binding of basiliximab with IL-2Ra can block IL-2 signaling (36).
However, the detailed molecular mechanism of the inhibition of
IL-2 signaling by basiliximab remains unclear.
complex with the IL-2Ra ectodomain. Structural analysis of this com-
plex and its comparison with the crystal structures of IL-2 in complex
with IL-2Rabgcreveal the molecular basis for the high specificity and
for the blockage of the IL-2 signaling by basiliximab. The structural
of improved and new mAb drugs against IL-2Ra.
Materials and Methods
Protein preparation and purification
The mAb basiliximab was purchased from Novartis Pharmaceuticals. The
and 100 mM sodium acetate (pH 5.5), followed by digestion with 10 mg/ml
papain (Sigma-Aldrich, St. Louis, MO) at 37˚C for 5 h. The reaction was
quenched with 20 mM iodoacetamide. The Fab fragment was separated by
ion exchange chromatography using a Q Sepharose Fast Flow column (GE
Healthcare). The Fab sample was concentrated to ∼10 mg/ml and then ex-
changed to a buffer of 10 mM HEPES (pH 7.0) and 50 mM NaCl.
The cDNA encoding the ectodomain of human IL-2Ra (residues 1–217)
was cloned into a modified pFastBac vector (Invitrogen, Carlsbad, CA) that
of the target protein, respectively. The recombinant protein was expressed
and secreted into the medium using a Bac-to-Bac baculovirus expression
the remaining cells and concentrated to a suitable volume followed by di-
alysis against a buffer of 10 mM Tris (pH 8.0) and 500 mM NaCl. The re-
Superflow column (Qiagen, Valencia, CA). Because the wild-type IL-2Ra
would form a disulfide-linked dimer with the free cysteine at position 192
(37), the purified disulfide-linked protein was reduced with 10 mM cysteine
and further alkylated with 20 mM iodoacetamide as previously described
(27). The resultant protein was purified further by gel filtration chromatog-
to the monomeric IL-2Ra was collected for further structural and bio-
chemical studies. The recombinant IL-2Ra is heavily glycosylated, and its
whereas the calculated molecular weight is ∼25 kDa.
HEPES (pH 7.0) and 50 mM NaCl and then concentrated to 3 mg/ml for
crystallization. The purity and homogeneity of the complex were confirmed
by SDS-PAGE and dynamic light scattering analysis.
Crystallization and diffraction data collection
Crystallization was performed using the hanging drop vapor diffusion
method at 20˚C. In a drop containing 0.5 ml of the protein complex sample
and 0.5 ml of the reservoir solution (0.2 M KCl, 0.05 M HEPES (pH 7.5),
and 45% pentaerythritol propoxylate (5/4 PO/OH) (38) equilibrated against
of 0.1 3 0.1 3 0.05 mm3after 15 d. The crystal was directly mounted on
a nylon loop and flash-cooled into the liquid N2stream (2170˚C). Dif-
fraction data were collected at Shanghai Synchrotron Radiation Facility
beamline BL-17U1 and processed with the program HKL2000 (39). The
statistics of the diffraction data are summarized in Table I.
Structure determination and refinement
The structure of the basiliximab Fab in complex with the IL-2Ra ecto-
domain was determined by the molecular replacement method im-
plemented in the program Phaser (40) with the structure of the basiliximab
Fab as the search model, followed by manual fitting of the IL-2Ra ecto-
domain. The rotation function search and the subsequent translational
function search with the structure of the free-form basiliximab Fab (Protein
Data Bank [PDB] code 1MIM) (41) used as the search model yielded an
outstanding solution in the asymmetric unit. To further locate the position
of the IL-2Ra ectodomain, we used all of the three available structures of
IL-2Ra (PDB codes 2B5I, 1Z92, and 2ERJ) (24–26) as templates with
Phaser and other commonly used programs implemented with the mo-
lecular replacement method. However, these attempts were unsuccessful to
find a solution for IL-2Ra. After several cycles of structure refinement
using the program CNS (42), the electron density for the D1 domain of IL-
2Ra was developed gradually. The structure of the D1 domain of IL-2Ra
(PDB code 2B5I) (25) was manually placed into the electron density. After
several rounds of structure refinement using the program Phenix (43) and
model building using the program O (44), the complete D1 and D2 do-
mains of IL-2Ra were modeled and fit well into the electron density.
However, as in all of the other IL-2Ra structures (24–26), several flexible
regions of IL-2Ra have no defined electron density and thus could not be
modeled, including the linker region between domains D1 and D2 (resi-
dues 62–100) and the C-terminal region (residues 159–217). The final
model contains 119 of 217 residues of the IL-2Ra ectodomain. There was
a long stretch of electron density near residue Asn49of IL-2Ra that could
be modeled as an N-linked core trisaccharide (MANb-1, 4GlcNAcb-1,
4GlcNAcb-1-Asn49) without ambiguity. All of the diffraction data were
used in the structure refinement except 5% of randomly chosen diffraction
data were set aside for free R factor cross-validation. The stereochemical
geometry of the final structure model was analyzed with the program
Procheck (45). The statistics of the refinement and structure model also are
listed in Table I. Structural analysis was performed using the programs in
the CCP4 suite (46) and the PISA server (47). Figures were prepared using
the program Pymol (www.pymol.org).
Surface plasmon resonance analysis
The kinetic studies of the interaction between basiliximab and the IL-2Ra
ectodomain were performed by the surface plasmon resonance method
using a Biacore 3000 instrument (GE Healthcare) at 25˚C. The mAb ba-
siliximab was immobilized on a CM5 sensor chip (GE Healthcare) using
an amine coupling kit (GE Healthcare). The purified IL-2Ra ectodomain
was dialyzed against the HBS (0.01 M HEPES, pH 7.4, 0.15 M NaCl,
3 mM EDTA, 0.005% Surfactant P20) buffer (GE Healthcare) and used as
the analyte in the binding assay. The association was monitored for a 240 s
period, and the disassociation was monitored by flowing the HBS buffer
for 600 s subsequently. An irrelevant chimeric mAb, the anti-CD20 mAb
Rituximab (Roche), was used as a reference. The experimental data were
analyzed with a 1:1 Langmuir model using the program BIAevalutation
The coordinates and structure factors of the basiliximab Fab in complex
with the ectodomain of IL-2Ra have been deposited in the Research
Collaboratory for Structural Bioinformatics PDB (www.rcsb.org/pdb/)
with the accession code 3IU3.
Overall structure of the basiliximab Fab in complex with the
The crystal structure of the basiliximab Fab in complex with the IL-
2Ra ectodomain was determined by the molecular replacement
method implemented with the structure of the free-form basiliximab
1362 STRUCTURE OF BASILIXIMAB Fab IN COMPLEX WITH IL-2Ra
Fab as the search model, followed by manual fitting of the IL-2Ra
ectodomain. This structure was refined to a resolution of 2.9 A˚,
yielding an R factor of 21.5% and a free R factor of 26.2% (Table I).
There are three Fab/IL-2Ra complexes in an asymmetric unit. The
structure model has good stereochemical geometry with only Thr50
of each L chain located in the disallowed regions of the Ram-
in the generously allowed region of the Ramachandran plot in the
free-form basiliximab Fab and has slightly differed f and c angles
residues,the three complexesareverysimilar(superposition of allof
the Ca atoms yields a root-mean-square deviation of 1.3–1.4 A˚for
IL-2Ra, 0.2–0.3 A˚for the Fab, and 0.5–0.6 A˚for the complex, re-
spectively), and the one with the most detectable residues and best
electron density has been selected for structural analysis.
The basiliximab Fab in the complex consists of the H chain
residues 1–215 that fold into the VHand CH1domains and the L
chain residues 1–208 that fold into the VLand CLdomains (Fig.
1A). The overall structures of the variable domains and constant
domains of the Fab in the complex are similar to those in the free-
form Fab reported previously (41) (superposition of all of the Ca
atoms yields a root-mean-square deviation of 0.7 A˚for the vari-
able domains and 0.9 A˚for the constant domains). The con-
formations of the CDRs of the Fab in the complex also resemble
those in the free-form Fab, indicating that binding of IL-2Ra does
not induce a significant conformational change of the Fab.
IL-2Ra complex is composed of two typical b-strand‑swapped
that assemble like a bent arm with an elbow angle of ∼90˚. The
electron density for both D1 and D2 domains was well defined,
especially in the regions participating in interactions with the Fab
(Fig. 1B). However, again similar to that seen in the other IL-2Ra
structures (24–26), the linker region between the D1 and D2 do-
mains (residues 62–100) and the C-terminal region (residues 159–
217) connecting the D2 domain to the transmembrane domain of
a total of 217 residues of the IL-2Ra ectodomain were modeled,
reflecting the flexible nature of IL-2Ra. A detailed structure com-
parison indicates that the overall structure of the IL-2Ra ectodo-
main in this complex is similar to that in its complexes with IL-2
(24–26). Superposition of the different structures yields a root-
mean-square deviation of 0.9–1.8 A˚for the D1 domain, 1.6–2.0 A˚
for the D2 domain, and 1.4–1.7 A˚for the whole IL-2Ra molecule.
The major structural differences occur in two solvent-exposed re-
in contact with other protein molecules. It is noteworthy that Asn49
of IL-2Ra in this complex is glycosylated and an N-linked core
IL-2Ra ectodomain. A, A stereoview of the overall structure of the com-
plex. The Fab is colored with the L chain in yellow and the H chain in
green, and IL-2Ra is colored with the D1 domain in cyan and the D2
domain in purple. The sugar chain of the glycosylated Asn49of IL-2Ra is
shown with a ball-and-stick model. B, A stereoview of a representative
SIGMAA-weighted 2Fo2 Fcelectron density map (1s contour level) in
regions of the D1 domain of IL-2Ra (residues 21–30 and 53–61) that are
involved in the interactions with the Fab. The atomic coordinates of the
residues are shown with ball-and-stick models. The disulfide bonds be-
tween Cys28and Cys59and between Cys30and Cys61are clearly defined.
Overall structure of the basiliximab Fab in complex with the
Table I.Summary of diffraction data and structure refinement statistics
Summary of Diffraction Data
a = b (A˚)
Resolution range (A˚)a
Unique reflections (I/s(I) . 0)
Wilson B factor (A˚2)
Statistics of Refinement and Model
Number of reflections [Fo. 0s(Fo)]
Free R set
R factor/free R factor (%)c
Number of protein atoms
Number of sugar atoms
Average B factor of all atoms (A˚2)
Root-mean-square bond lengths (A˚)
Root-mean-square bond angles (˚)
Ramachandran plot (%)
Most favored regions
Generously allowed regions
aNumbers in parentheses refer to the highest resolution shell.
cR factor = ||Fo| 2 |Fc||/|Fo|.
The Journal of Immunology1363
trisaccharide (MANb-1, 4GlcNAcb-1, 4GlcNAcb-1-Asn49) could
be modeled without ambiguity. This is the first observation of gly-
cosylation of this residue, because in the previously reported
structures Asn49is either disorderedor mutated to serine toprohibit
the glycosylation of the protein (24–26). In our structure, although
Asn49and the sugar chain do not have direct contact with the ba-
the interactions with the Fab, which might contribute to the stabi-
lization of the glycosylated Asn49. However, the biological signifi-
cance of the glycosylation is unclear.
Interactions between the basiliximab Fab and the IL-2Ra
In the complex structure, the basiliximab Fab forms extensive hy-
drophilic and hydrophobic interactions with IL-2Ra via a large
interface, including 10 hydrogen bonds, 4 salt bridges, and 138 van
der Waals contacts (Fig. 2, Tables II and III). Although the complex
than the average value (0.64–0.68) for Ag/Ab complexes (49),
formation of the complex buries a very large solvent-accessible
surface area of 2255.4 A˚2(1108.6 A˚2on the Fab and 1146.8 A˚2on
IL-2Ra), which is much higher than the common value seen in the
A˚2of the buried surface area, and the L chain 339.7 A˚2, consistent
with the notion that the mAb H chain usually makes more con-
tributions than the L chain in Ag binding (52, 53).
The Ag binding site of basiliximab is comprised of all six CDR
loops that form a large, flat surface to accommodate a conforma-
tional epitope of IL-2Ra (Fig. 2A, 2B). The basiliximab epitope
consists of several discontinuous segments of IL-2Ra, including
residues 1–6, 21–29, 38–48, and 56–57 of the D1 domain and
residues 118–120 of the D2 domain. In total, 23 residues of IL-
2Ra (21 residues of the D1 domain and 2 residues of the D2
domain) are involved in the interactions with the Fab, which is
more than those in most Ag/Ab complexes (51). The sugar chain
of the glycosylated Asn49of IL-2Ra extends away from the in-
teraction interface and has no direct contact with the Fab. These
results are in agreement with the biochemical data that basilix-
imab has a very high binding affinity to the IL-2Ra ectodomain
with an apparent Kd(KD) of 0.14 nM determined by the surface
plasmon resonance technique (data not shown).
A detailed analysis of the interactions between the basiliximab
Fab and the IL-2Ra ectodomain shows that the Ag binding site of
basiliximab is composed of a shallow hydrophobic groove sur-
rounded by four hydrophilic patches (hereafter residues of the Fab
H chain and L chain and IL-2Ra are designated with the chain
identifiers H, L, and I, respectively) (Fig. 2A, 2B). At the center of
the Ag binding site, there is a shallow groove of a hydrophobic
nature consisting of six aromatic residues TyrL93, TyrH30, TrpH31,
TyrH50, TyrH98, and TyrH100at the interface of the VLand VH
domains (Fig. 2A, 2B). Residues in this groove form extensive
hydrophobic interactions with a hydrophobic surface patch of IL-
down at the Fab. The basiliximab Fab is shown with an electrostatic potential surface with the locations of some of the residues involved in the interactions
with IL-2Ra labeled in green. IL-2Ra is shown with a stick model with the D1 domain in cyan and the D2 domain in purple, and the residues involved in
the interactions with the Fab are shown in ball-and-stick models and labeled in black. B, The interaction interface between the Fab and IL-2Ra viewing
down at IL-2Ra in the same orientation as in Fig. 2A. IL-2Ra is shown with an electrostatic potential surface with the locations of some of the residues
involved in the interactions with the Fab labeled in cyan. The Fab is shown with a stick model with the CDR loops H1, H2, and H3 in green and the CDR
loops L1, L2, and L3 in yellow, and the residues involved in the interactions with IL-2Ra are indicated in ball-and-stick models and labeled in black. C, A
stereoview showing the hydrogen bonding and salt bridge interactions between IL-2Ra (in cyan) and the L chain of the Fab (in yellow). The hydrogen
bonds are indicated with gray dashes, and the salt bridges are indicated with orange dashes. D, A stereoview showing the hydrogen bonding and salt bridge
interactions between IL-2Ra (D1 domain in cyan and D2 domain in purple) and the CDR loop H1 of the Fab (in green). E, A stereoview showing the
hydrogen bonding and salt bridge interactions between IL-2Ra (in cyan) and the CDR loops H2 and H3 of the Fab (in green). The color coding of the
structural elements is the same as that in Fig. 1A.
Interactions between the basiliximab Fab and the IL-2Ra ectodomain. A, The interaction interface between the Fab and IL-2Ra viewing
1364STRUCTURE OF BASILIXIMAB Fab IN COMPLEX WITH IL-2Ra
2Ra consisting of the residues LeuI2, MetI25, LeuI42, TyrI43, LeuI45,
IleI118, and HisI120. This part of the interface contributes 59 van
der Waals contacts and 2 hydrogen bonds in total and plays
a pivotal role in the formation of the complex (Tables II and III).
In particular, the residues MetI25, LeuI42, and TyrI43of IL-2Ra
intrude their side chains into the groove to form numerous hy-
drophobic contacts with the surrounding residues, and TyrI43in-
serts its side chain into a small cavity in the groove and forms
a hydrogen bond with AspH97. In addition, AsnI27forms a hydro-
gen bond with TrpH31. These results are in agreement with the
notion that tyrosine is the most frequently observed residue in the
CDRs and plays the most important role in the Ag/Ab interaction
because it can form both hydrophobic and hydrogen bonding in-
Surrounding the hydrophobic groove there are three positively
charged surface patches and one negatively charged surface patch
several residues of CDRs L1 and L3, including ArgL29, ArgL90,
and SerL91that form both hydrophilic and hydrophobic inter-
actions with residues AspI56and AsnI57of IL-2Ra (Fig. 2A, 2B).
In particular, AspI56stretches its side chain into a shallow cavity
on the surface of the Fab and forms two salt bridges with ArgL29
and ArgL90, and AsnI57forms two hydrogen bonds with ArgL90
and SerL91(Fig. 2C, Table II). The second positively charged
patch is composed of the residues AspL49and LysL52of CDR L2
that form three hydrogen bonds and several hydrophobic inter-
actions with the residues ThrI47and GlyI48of IL-2Ra (Fig. 2A–C,
Table II). The third positively charged patch is composed of
mainly CDR H1 residues ArgH29and TyrH30that interact with
a negatively charged surface patch of IL-2Ra formed by two
short segments (residues 1–6 and residues 116–120) (Fig. 2A,
2B). In particular, ArgH29forms a salt bridge with AspI6and
a hydrogen bond with HisI120, and TyrH30forms a hydrogen bond
with AspI4(Fig. 2D, Table II). The fourth surface patch is neg-
atively charged and composed of mainly CDR H2 residues
AsnH53, AspH55, and GluH63that interact with a positively
charged surface patch of IL-2Ra formed by the residues ArgI35,
ArgI36, and LysI38(Fig. 2A, 2B). Specifically, AsnH53forms
a hydrogen bond with GluI29, and AspH55forms a salt bridge with
ArgI36(Fig. 2E, Table II). In addition, GluH63has a weak hy-
drophilic interaction with LysI38(4.1 A˚). Among the many resi-
dues of IL-2Ra involved in the interactions with the Fab, the key
residues of the epitope appear to be ThrI47, AspI56, and AsnI57,
each of which makes two hydrogen bonds or salt bridges with the
Fab, and LeuI42and TyrI43, each of which makes many hydro-
phobic interactions with the Fab.
The epitope of basiliximab
Previously, the basiliximab epitope was mapped to residues 116–
122 of the D2 domain of IL-2Ra using the phage display method
(36). In the basiliximab Fab/IL-2Ra complex structure, only two
Specifically, IleI118makes twovan der Waals contacts with the Fab,
and HisI120forms one hydrogen bond with ArgH29and five van der
Waals contacts with the Fab (Tables II and III). These interactions
constitute only a small portion of the interactions at the third hy-
recognition and binding of basiliximab. These results indicate that
the potential epitope identified by the phage display method is in-
complete and the functional role of the identified region in the
recognition is undefined. A similar situation also was seen in the
structural studies of the Rituximab Fab in complex with a peptide
corresponding to its epitope on CD20. With the phage display
method, the Rituximab binding epitope was mapped to two seg-
loop. The structural studies of the Ag/Ab complex indicate that the
182YCYSI186motif is not directly involved in the interaction with
the Ab; instead it plays a critical role in the formation of a disulfide
bond to define the proper geometry of the170ANPS173motif so that
the latter can be recognized by Rituximab (57, 58).
170ANPS173and182YCYSI186on the CD20 extracellular
Molecular mechanism of the inhibition of the IL-2 signal
pathway by basiliximab
Structural analysis of the basiliximab Fab/IL-2Ra complex and its
comparison with the crystal structures of the IL-2/IL-2Ra and IL-
2/IL-2Rabgc complexes provide insights into the molecular
Hydrogen bonds and salt bridges between the basiliximab Fab
Fab Atom CDR LoopIL-2Ra Atom Distance (A˚)
Fab Residue CDR LoopIL-2Ra Residue
IL-2Ra (#4.0 A˚)
van der Waals contacts between the basiliximab Fab and
IL-2Ra Residue Fab Residue
SerH26(5), ArgH29(8), TyrH30(4)
TrpH31(1), AspH55(5), SerH57(1)
TyrL93(4), TrpH31(5), SerH57(1)
TyrL93(7), TrpH31(1), AspH97(6), GlyH99(2)
TyrL31(4), AspL49(5), LysL52(1), TyrH100(1)
SerL30(2), TyrL31(6), ArgL90(5)
ArgL29(1), ArgL90(3), SerL91(2)
There are a total of 138 van der Waals contacts. Numbers in parentheses refer to
the number of van der Waals contacts.
The Journal of Immunology1365
mechanism of the inhibition of the IL-2 signal pathway by basi-
liximab. In the crystal structures of the IL-2/IL-2Ra and IL-2/IL-
2Rabgccomplexes, IL-2Ra makes interactions with IL-2 mainly
via the D1 domain (24–26). Specifically, MetI25, LeuI42, and TyrI43
of IL-2Ra form a hydrophobic surface patch to interact with
residues Phe42and Leu72of IL-2 (24). Surrounding the hydro-
phobic patch, several hydrophilic residues of IL-2Ra interact with
IL-2 by forming numerous hydrogen bonds, salt bridges, and van
der Waals contacts. In total, there are 21 residues of IL-2Ra
participating in the interactions with IL-2, forming 8 hydrogen
bonds, 2 salt bridges, and 100 van der Waals contacts (25). The
interaction interface buries 971.2 A˚2of the solvent-accessible
surface area on IL2-Ra. Structural comparison of these complexes
with the basiliximab Fab/ IL-2Ra complex indicates that the resi-
dues of IL-2Ra responsible for the interactions with IL-2 overlap
largely with the epitope of basiliximab. Fifteen out of the 21 resi-
dues (71.4%) are involved in the interactions with the basiliximab
Fab, and ∼641.6 A˚2(66.1%) of the buried solvent-accessible sur-
face area on IL-2Ra is covered by the Ab (Fig. 3A). Especially,
the hydrophobic patch formed by residues MetI25, LeuI42, and
TyrI43of IL-2Ra plays a keyrole in the binding of both basiliximab
and IL-2. Furthermore, the basiliximab binding epitope comprises
several other residues besides those involved in the interactions
with IL-2, and the basiliximab Fab/IL-2Ra interface comprises
more hydrophobic and hydrophilic interactions than the IL-2/IL-
2Ra interface. These results may explain in part the biochemical
data that the binding affinity of basiliximab to IL-2Ra (0.14 nM) is
∼71-fold higher than that of IL-2 to IL-2Ra (10 nM) (27).
Therefore, the binding of basiliximab to IL-2Ra would compete
for IL-2 binding to the receptor. In the presence of a sufficient
amount of basiliximab, the IL-2 binding sites of IL-2Ra would be
blocked, and thus IL-2 signaling cannot be initiated and executed
due to the lack of binding of IL-2 to IL-2Ra and the formation of
the functional IL-2/IL-2Rabgccomplex (Fig. 3A, 3B). This pro-
vides the molecular mechanism of the inhibition of IL-2 signaling
Implications for drug development
IL-2 signaling plays an important role in the activation of immune
responses against foreign intrusion. Blockage of this signal path-
way could provide therapeutic benefits to reduce or eliminate al-
lograft rejection in organ transplantation. The biological and
structural data have shown that activation of IL-2 is initiated by the
binding of IL-2Ra and further facilitated by the binding of IL-2Rb
and gc(24–27). Among the three receptors, only IL-2Ra is IL-2–
specific, whereas IL-2Rb and gcare shared with other cytokines
and are less specific. Therefore, IL-2Ra is a more suitable drug
target for blocking the IL-2 signal pathway. Basiliximab and da-
clizumab are two IL-2Ra–specific mAb drugs that have been used
in clinical applications for the prevention of allograft rejection in
Although basiliximab binds to IL-2Ra with a high affinity (Kd=
0.14 nM), the Fab/IL-2Ra complex has a relatively lower shape
complementarity value of 0.57, suggesting that the paratope of ba-
siliximab could be modified to have higher shape and chemical
complementarities with IL-2Ra and thus achieve a tighter binding
and better specificity. In our previous structural studies of the Rit-
uximab Fab in complex with a CD20 peptide, we have proposed
some mutations on the CDR loops of the Ab that might be able to
increase its binding affinity (57). Recently, these suggestions have
been validated, and the results have shown that some of the muta-
tions could substantially improve the affinity (59). Structural anal-
ysis of the basiliximab Fab/IL-2Ra complex also provides some
hints for improving the mAb drug. For instance, MetI25intrudes its
but forms only one van der Waals contact with GlyH99. Modeling
studies show that a substitute for GlyH99with a slightly larger hy-
drophobic residue, such as alanine, valine, leucine, or isoleucine,
could form more hydrophobic interactions with MetI25and the sur-
rounding residues LeuI2, LeuI42, and TyrI43without steric conflict,
complementarities. Moreover, because LysL52forms a hydrogen
bond with the main-chain carbonyl of GlyI48and a weak salt bridge
with GluI22(4.3 A˚), mutation of LysL52to arginine would make it
form a more favorable salt bridge with GluI22. In addition, SerL55is
located near the interaction interface but has no direct contact with
IL-2Ra; however, substitutionof this residuewitha large positively
bridge with GluI1(within 3.5 A˚). In other words, mutations of the
aforementioned residues at the paratope of basiliximab could in-
troduce additional favorable interactions between the mAb and IL-
2a andthusimprove the binding affinityand specificityof themAb.
2Ra and further transmitted by the binding of IL-2Rb and gcto form the biologically active complex IL-2/IL-2Rabgc. The binding of basiliximab to IL-2Ra
blocks its binding with IL-2 and prevents the formation of the IL-2/IL-2Rabgccomplex, thus inhibits the IL-2 signal pathway.
Molecular mechanism of the inhibition of IL-2 signaling by basiliximab. A, Overlay of the basiliximab Fab/IL-2Ra complex onto the IL-2/IL-
1366STRUCTURE OF BASILIXIMAB Fab IN COMPLEX WITH IL-2Ra
Structural analysis of the basiliximab Fab/IL-2Ra complex also
has some implications for the development of new anti–IL-2Ra
for IL-2 is a relatively flat surface. Thus, it is difficult to design
the binding of IL-2Ra with IL-2. Basiliximab uses all six CDRs to
form a large, flat paratope to accommodate a large conformational
the region involved in the interactions with IL-2 with extensive hy-
drophobic and hydrophilic interactions. This is different from most
other Ag/Ab complexes in which both VHand VLCDRs are in-
volvedinthe formationof adeeppocket orgroovetoaccommodate
an epitope intruding from the Ag with the VHCDRs playing
crystal structures of the Ag/Ab complexes in the PDB, Collis et al.
paratope to bind the epitope. Normally, the CDR loops L1, L3, H1,
and H2 of the mAbs have ∼10–17, 7–13, 5–7, and 16–19 residues,
respectively, the CDR loop L2 has ∼7 residues with less variation,
and the CDR loop H3 has a great variation of 3–19 residues with an
average value of 9 in murine mAbs (60, 61). For basiliximab, the
CDR loop L3 contains 7 residues, which is the shortest compared
with the other mAbs, and the CDR loop H3 contains 8 residues,
which is also slightly shorter than the average value of the murine
mAbs. In addition, the CDR loops L1, H1, and H2 of basiliximab
has 7 residues as usual. Thus, these short CDR loops make basi-
liximab form a large, flat depression instead of a deep pocket or
groove as the paratope that is more advantageous to bind a large
conformational epitope with a flat surface. It is interesting to note
that the CDR loops L1, L2, L3, H1, H2, and H3 of daclizumab, the
other anti–IL-2Ra mAb drug, have 10, 7, 9, 5, 17, and 7 residues,
like basiliximab, daclizumab also has a large, flat paratope com-
2Ra. These results together indicate that because IL-2Ra uses
binding affinity and a high specificity, basiliximab uses all six CDR
large conformational epitope of IL-2Ra. This notion should be
taken into account in the design and development of new anti–IL-
2Ra mAb drugs. In addition, this result might also be useful in the
design and development of mAb drugs against proteins that have
a flat molecular surface with no large surface intrusion for Ab rec-
ognition and no proper pocket for small molecule binding.
We are grateful to the staff members at Shanghai Synchrotron Radiation
Facility for their assistance in diffraction data collection.
The authors have no financial conflicts of interest.
1. Morgan, D. A., F. W. Ruscetti, and R. Gallo. 1976. Selective in vitro growth of T
lymphocytes from normal human bone marrows. Science 193: 1007–1008.
2. Robb, R. J., and K. A. Smith. 1981. Heterogeneity of human T-cell growth factor
(s) due to variable glycosylation. Mol. Immunol. 18: 1087–1094.
3. Smith, K. A., M. F. Favata, and S. Oroszlan. 1983. Production and character-
ization of monoclonal antibodies to human interleukin 2: strategy and tactics.
J. Immunol. 131: 1808–1815.
4. Smith, K. A., T. N. Fredrickson, L. E. Mobraaten, and E. DeMaeyer. 1977. The
interaction of erythropoietin with fetal liver cells. II. Inhibition of the erythro-
poietin effect by interferon. Exp. Hematol. 5: 333–340.
5. Taniguchi, T., H. Matsui, T. Fujita, C. Takaoka, N. Kashima, R. Yoshimoto, and
J. Hamuro. 1983. Structure and expression of a cloned cDNA for human in-
terleukin-2. Nature 302: 305–310.
6. Klebb, G., I. B. Autenrieth, H. Haber, E. Gillert, B. Sadlack, K. A. Smith, and
I. Horak. 1996. Interleukin-2 is indispensable for development of immunological
self-tolerance. Clin. Immunol. Immunopathol. 81: 282–286.
7. Smith, K. A. 1988. Interleukin-2: inception, impact, and implications. Science
8. Nelson, B. H., and D. M. Willerford. 1998. Biology of the interleukin-2 receptor.
Adv. Immunol. 70: 1–81.
9. Friedmann, M. C., T. S. Migone, S. M. Russell, and W. J. Leonard. 1996. Dif-
ferent interleukin 2 receptor b-chain tyrosines couple to at least two signaling
pathways and synergistically mediate interleukin 2-induced proliferation. Proc.
Natl. Acad. Sci. USA 93: 2077–2082.
10. Ravichandran, K. S., V. Igras, S. E. Shoelson, S. W. Fesik, and S. J. Burakoff.
1996. Evidence for a role for the phosphotyrosine-binding domain of Shc in
interleukin 2 signaling. Proc. Natl. Acad. Sci. USA 93: 5275–5280.
11. Gadina, M., C. Sudarshan, R. Visconti, Y. J. Zhou, H. Gu, B. G. Neel, and
J. J. O’Shea. 2000. The docking molecule Gab2 is induced by lymphocyte ac-
tivation and is involved in signaling by interleukin-2 and interleukin-15 but not
other common g chain-using cytokines. J. Biol. Chem. 275: 26959–26966.
12. Gaffen, S. L., S. Y. Lai, M. Ha, X. Liu, L. Hennighausen, W. C. Greene, and
M. A. Goldsmith. 1996. Distinct tyrosine residues within the interleukin-2 re-
ceptor b chain drive signal transduction specificity, redundancy, and diversity.
J. Biol. Chem. 271: 21381–21390.
13. Minami, Y., T. Kono, T. Miyazaki, and T. Taniguchi. 1993. The IL-2 receptor
14. Smith, K. A. 1980. T-cell growth factor. Immunol. Rev. 51: 337–357.
15. Robb, R. J., A. Munck, and K. A. Smith. 1981. T cell growth factor receptors.
Quantitation, specificity, and biological relevance. J. Exp. Med. 154: 1455–1474.
16. Wang, H. M., and K. A. Smith. 1987. The interleukin 2 receptor. Functional
consequences of its bimolecular structure. J. Exp. Med. 166: 1055–1069.
17. Tsudo, M., R. W. Kozak, C. K. Goldman, and T. A. Waldmann. 1986. Dem-
onstration of a non-Tac peptide that binds interleukin 2: a potential participant in
a multichain interleukin 2 receptor complex. Proc. Natl. Acad. Sci. USA 83:
18. Sharon, M., R. D. Klausner, B. R. Cullen, R. Chizzonite, and W. J. Leonard.
1986. Novel interleukin-2 receptor subunit detected by cross-linking under high-
affinity conditions. Science 234: 859–863.
19. He, Y. W., and T. R. Malek. 1998. The structure and function of gc-dependent
cytokines and receptors: regulation of T lymphocyte development and homeo-
stasis. Crit. Rev. Immunol. 18: 503–524.
20. Takeshita, T., H. Asao, K. Ohtani, N. Ishii, S. Kumaki, N. Tanaka, H. Munakata,
M. Nakamura, and K. Sugamura. 1992. Cloning of the g chain of the human IL-2
receptor. Science 257: 379–382.
21. Nakamura, Y., S. M. Russell, S. A. Mess, M. Friedmann, M. Erdos, C. Francois,
Y. Jacques, S. Adelstein, and W. J. Leonard. 1994. Heterodimerization of the IL-
2 receptor b- and g-chain cytoplasmic domains is required for signalling. Nature
22. Nelson, B. H., J. D. Lord, and P. D. Greenberg. 1994. Cytoplasmic domains of
the interleukin-2 receptor b and g chains mediate the signal for T-cell pro-
liferation. Nature 369: 333–336.
23. Brandhuber, B. J., T. Boone, W. C. Kenney, and D. B. McKay. 1987. Three-
dimensional structure of interleukin-2. Science 238: 1707–1709.
24. Rickert, M., X. Wang, M. J. Boulanger, N. Goriatcheva, and K. C. Garcia. 2005.
The structure of interleukin-2 complexed with its a receptor. Science 308: 1477–
25. Wang, X., M. Rickert, and K. C. Garcia. 2005. Structure of the quaternary
complex of interleukin-2 with its a, b, and gcreceptors. Science 310: 1159–
26. Stauber, D. J., E. W. Debler, P. A. Horton, K. A. Smith, and I. A. Wilson. 2006.
Crystal structure of the IL-2 signaling complex: paradigm for a heterotrimeric
cytokine receptor. Proc. Natl. Acad. Sci. USA 103: 2788–2793.
27. Rickert, M., M. J. Boulanger, N. Goriatcheva, and K. C. Garcia. 2004. Compen-
satory energetic mechanisms mediating the assembly of signaling complexes be-
tween interleukin-2 and its a, b, and gcreceptors. J. Mol. Biol. 339: 1115–1128.
28. Forsten, K. E., and D. A. Lauffenburger. 1994. The role of low-affinity in-
terleukin-2 receptors in autocrine ligand binding: alternative mechanisms for
enhanced binding effect. Mol. Immunol. 31: 739–751.
29. Rubin, L. A., and D. L. Nelson. 1990. The soluble interleukin-2 receptor: bi-
ology, function, and clinical application. Ann. Intern. Med. 113: 619–627.
30. Waldmann, T. A. 2007. Daclizumab (anti-Tac, Zenapax) in the treatment of
leukemia/lymphoma. Oncogene 26: 3699–3703.
31. Waldmann, T. A. 1989. The multi-subunit interleukin-2 receptor. Annu. Rev.
Biochem. 58: 875–911.
32. Waldmann, T. A. 2003. The meandering 45-year odyssey of a clinical immu-
nologist. Annu. Rev. Immunol. 21: 1–27.
33. Waldmann, T. A. 1986. The structure, function, and expression of interleukin-2
receptors on normal and malignant lymphocytes. Science 232: 727–732.
34. Kahan, B. D., P. R. Rajagopalan, and M. Hall; United States Simulect Renal
Study Group. 1999. Reduction of the occurrence of acute cellular rejection
among renal allograft recipients treated with basiliximab, a chimeric anti-
interleukin-2-receptor monoclonal antibody. Transplantation 67: 276–284.
The Journal of Immunology1367
35. Tang, I. Y., H. U. Meier-Kriesche, and B. Kaplan. 2007. Immunosuppressive
strategies to improve outcomes of kidney transplantation. Semin. Nephrol. 27:
36. Binder, M., F. N. Vo ¨gtle, S. Michelfelder, F. Mu ¨ller, G. Illerhaus, S. Sundararajan,
R. Mertelsmann, and M. Trepel. 2007. Identification of their epitope reveals the
structural basis for the mechanism of action of the immunosuppressive antibodies
basiliximab and daclizumab. Cancer Res. 67: 3518–3523.
37. Kato, K., and K. A. Smith. 1987. Tac antigen forms disulfide-linked homo-
dimers. Biochemistry 26: 5359–5364.
38. Gulick, A. M., A. R. Horswill, J. B. Thoden, J. C. Escalante-Semerena, and
I. Rayment. 2002. Pentaerythritol propoxylate: a new crystallization agent and
cryoprotectant induces crystal growth of 2-methylcitrate dehydratase. Acta
Crystallogr. D Biol. Crystallogr. 58: 306–309.
39. Otwinowski, Z., and W. Minor. 1997. Processing of X-ray diffraction data col-
lected in oscillation mode. Methods Enzymol. 276: 307–326.
40. McCoy, A. J., R. W. Grosse-Kunstleve, L. C. Storoni, and R. J. Read. 2005.
Likelihood-enhanced fast translation functions. Acta Crystallogr. D Biol. Crys-
tallogr. 61: 458–464.
41. Mikol, V. 1996. Structure of the Fab fragment of SDZ CHI621: a chimeric an-
tibody against CD25. Acta Crystallogr. D Biol. Crystallogr. 52: 534–542.
42. Bru ¨nger, A. T., P. D. Adams, G. M. Clore, W. L. DeLano, P. Gros, R. W. Grosse-
Kunstleve, J. S. Jiang, J. Kuszewski, M. Nilges, N. S. Pannu, et al. 1998.
Crystallography & NMR system: a new software suite for macromolecular
structure determination. Acta Crystallogr. D Biol. Crystallogr. 54: 905–921.
43. Adams, P. D., R. W. Grosse-Kunstleve, L. W. Hung, T. R. Ioerger, A. J. McCoy,
N. W. Moriarty, R. J. Read, J. C. Sacchettini, N. K. Sauter, and T. C. Terwilliger.
2002. PHENIX: building new software for automated crystallographic structure
determination. Acta Crystallogr. D Biol. Crystallogr. 58: 1948–1954.
44. Jones, T. A., J. Y. Zou, S. W. Cowan, and M. Kjeldgaard. 1991. Improved
methods for building protein models in electron density maps and the location of
errors in these models. Acta Crystallogr. A 47: 110–119.
45. Laskowski, R. A., M. W. Macarthur, D. S. Moss, and J. M. Thornton. 1993.
PROCHECK: a program to check the stereochemical quality of protein struc-
tures. J. Appl. Crystallogr. 26: 283–291.
46. Collaborative Computational Project, Number 4. 1994. The CCP4 suite: pro-
grams for protein crystallography. Acta Crystallogr. D Biol. Crystallogr. 50:
47. Krissinel, E., and K. Henrick. 2007. Inference of macromolecular assemblies
from crystalline state. J. Mol. Biol. 372: 774–797.
48. Al-Lazikani, B., A. M. Lesk, and C. Chothia. 1997. Standard conformations for
the canonical structures of immunoglobulins. J. Mol. Biol. 273: 927–948.
49. Lawrence, M. C., and P. M. Colman. 1993. Shape complementarity at protein/
protein interfaces. J. Mol. Biol. 234: 946–950.
50. Wilson, I. A., and R. L. Stanfield. 1993. Antibody-antigen interactions. Curr.
Opin. Struct. Biol. 3: 113–118.
51. Davies, D. R., and G. H. Cohen. 1996. Interactions of protein antigens with
antibodies. Proc. Natl. Acad. Sci. USA 93: 7–12.
52. Wilson, I. A., and R. L. Stanfield. 1994. Antibody-antigen interactions: new
structures and new conformational changes. Curr. Opin. Struct. Biol. 4: 857–867.
53. Davies, D. R., E. A. Padlan, and S. Sheriff. 1990. Antibody-antigen complexes.
Annu. Rev. Biochem. 59: 439–473.
54. Koide, S., and S. S. Sidhu. 2009. The importance of being tyrosine: lessons in
molecular recognition from minimalist synthetic binding proteins. ACS Chem.
Biol. 4: 325–334.
55. Mian, I. S., A. R. Bradwell, and A. J. Olson. 1991. Structure, function and
properties of antibody binding sites. J. Mol. Biol. 217: 133–151.
56. Birtalan, S., Y. Zhang, F. A. Fellouse, L. Shao, G. Schaefer, and S. S. Sidhu.
2008. The intrinsic contributions of tyrosine, serine, glycine and arginine to the
affinity and specificity of antibodies. J. Mol. Biol. 377: 1518–1528.
57. Du, J., H. Wang, C. Zhong, B. Peng, M. Zhang, B. Li, S. Huo, Y. Guo, and
J. Ding. 2007. Structural basis for recognition of CD20 by therapeutic antibody
rituximab. J. Biol. Chem. 282: 15073–15080.
58. Binder, M., F. Otto, R. Mertelsmann, H. Veelken, and M. Trepel. 2006. The
epitope recognized by rituximab. Blood 108: 1975–1978.
59. Li, B., L. Zhao, H. Guo, C. Wang, X. Zhang, L. Wu, L. Chen, Q. Tong, W. Qian,
H. Wang, and Y. Guo. 2009. Characterization of a rituximab variant with potent
antitumor activity against rituximab-resistant B-cell lymphoma. Blood 114:
60. Collis, A. V., A. P. Brouwer, and A. C. Martin. 2003. Analysis of the antigen
combining site: correlations between length and sequence composition of the
hypervariable loops and the nature of the antigen. J. Mol. Biol. 325: 337–354.
61. Wu, T. T., G. Johnson, and E. A. Kabat. 1993. Length distribution of CDRH3 in
antibodies. Proteins 16: 1–7.
1368 STRUCTURE OF BASILIXIMAB Fab IN COMPLEX WITH IL-2Ra