Structural Basis of Integrin Regulation and Signaling

ANRV306-IY25-21 ARI 11 February 2007 13:4
Structural Basis of Integrin

Regulation and Signaling
Bing-Hao Luo, Christopher V. Carman,
Annu. Rev. Immunol. 2007.25:619-647. Downloaded from arjournals.annualreviews.org
and Timothy A. Springer

The CBR Institute for Biomedical Research and Department of Pathology, Harvard
by HARVARD UNIVERSITY on 04/13/07. For personal use only.
Medical School, Boston, Massachusetts 02115; email: luo@cbr.med.harvard.edu,

ccarman@bidmc.harvard.edu, SpringerOfce@cbr.med.harvard.edu
Annu. Rev. Immunol. 2007. 25:61947 Key Words

First published online as a Review in Advance on cell adhesion, conformational change, ICAM-1, I domain,
January 2, 2007
migration
The Annual Review of Immunology is online at
immunol.annualreviews.org Abstract
This articles doi: Integrins are cell adhesion molecules that mediate cell-cell, cell
10.1146/annurev.immunol.25.022106.141618
extracellular matrix, and cell-pathogen interactions. They play
Copyright c 2007 by Annual Reviews. critical roles for the immune system in leukocyte trafcking
All rights reserved
and migration, immunological synapse formation, costimulation,
0732-0582/07/0423-0619$20.00 and phagocytosis. Integrin adhesiveness can be dynamically regu-
lated through a process termed inside-out signaling. In addition,
ligand binding transduces signals from the extracellular domain to
the cytoplasm in the classical outside-in direction. Recent struc-
tural, biochemical, and biophysical studies have greatly advanced
our understanding of the mechanisms of integrin bidirectional sig-
naling across the plasma membrane. Large-scale reorientations of
the ectodomain of up to 200 A couple to conformational change in
ligand-binding sites and are linked to changes in and subunit
transmembrane domain association. In this review, we focus on in-
tegrin structure as it relates to afnity modulation, ligand binding,
outside-in signaling, and cell surface distribution dynamics.
619
INTRODUCTION As adhesion molecules, integrins are

unique in that their adhesiveness can be dy-
The immune system relies heavily on in-
namically regulated through a process termed
tegrins for (a) adhesion during leukocyte
inside-out signaling or priming. Thus, stim-
trafcking from the bloodstream, migration
uli received by cell surface receptors for
within tissues, immune synapse formation,
chemokines, cytokines, and foreign antigens
and phagocytosis; and (b) signaling during
initiate intracellular signals that impinge on
costimulation and cell polarization. Inte-
integrin cytoplasmic domains and alter ad-
grins are so named because they integrate
hesiveness for extracellular ligands. In addi-
the extracellular and intracellular environ-
tion, ligand binding transduces signals from
ments by binding to ligands outside the cell
the extracellular domain to the cytoplasm in
and cytoskeletal components and signaling
the classical outside-in direction (outside-in
molecules inside the cell. Integrins are nonco-
signaling). These dynamic properties of in-
valently associated heterodimeric cell surface

tegrins are central to their proper function
adhesion molecules. In vertebrates, 18 sub-
in the immune system. Indeed, mutations or
units and 8 subunits form 24 known pairs
small molecules that stabilize either the in-
(Figure 1). This diversity in subunit composi-

active state or the active adhesive stateand
tion contributes to diversity in ligand recogni-
thereby block the adhesive dynamics of leuko-
tion, binding to cytoskeletal components and
cyte integrinsinhibit leukocyte migration
coupling to downstream signaling pathways.
and normal immune responses.
Immune cells express at least 10 members of
the integrin family belonging to the 2, 7,
and 1 subfamilies (Table 1). The 2 and 7
integrins are exclusively expressed on leuko- INTEGRIN I DOMAINS
cytes, whereas the 1 integrins are expressed Half of integrin subunits contain a domain
on a wide variety of cells throughout the body. of about 200 amino acids known as an inserted
Distribution and ligand-binding properties of (I) domain, or a von Willebrand factor A do-
the integrins on leukocytes are summarized in main (Figure 1). In integrins in which it is
Table 1. For reviews, see References 1 and present, the I domain is the major or ex-
2. Mutations that block expression of the 2 clusive ligand-binding site. In this review, we
integrin subfamily lead to leukocyte adhesion begin with the I domain because it serves
deciency, a disease associated with severe im- as a paradigm for understanding conforma-
munodeciency (3). tional regulation and ligand binding for all
1*
10* IIb
11*
L* 2*
E* 7 3
Figure 1 M* 3*
2 5
The 24 integrin X* 4* 1* V*
heterodimers. The 6
subunits with I D* 5*
8
domains are 4 6*
asterisked. Integrin
heterodimers on 7*
immune cells are 8*
shown with red
lines. 9*
620 Luo Carman Springer

Table 1 Integrins on leukocytesa

Integrin Distributionb Ligandc
L2, LFA-1, CD11a/CD18 Lymphocytes, NK cells, monocytes, ICAM-1, -2, -3, -5
macrophages, dendritic cells, neutrophils
M2, Mac-1, CR3, CD11b/CD18 Monocytes, macrophages, neutrophils, NK iC3b, brinogen, heparin, many others
cells
X2, p150,95, CR4, Monocytes, macrophages, NK cells, dendritic iC3b, brinogen, heparin, many others
CD11c/CD18 cells
D2 Monocytes, macrophages, eosinophils, ICAM-3, VCAM-1
neutrophils
41, VLA-4, CD49d/CD29 Lymphocytes, monocytes, eosinophils VCAM-1, bronectin
47, LPAM-1 Lymphocytes, monocytes, NK cells MAdCAM-1, bronectin
E7, HML-1 Intra-epithelial T lymphocytes E-cadherin
11, VLA-1, CD49a/CD29 Long-term activated T lymphocytes, B Collagen

lymphocytes, monocytes
21, VLA-2, GPIa, CD49b/CD29 long-term activated T lymphocytes, B Collagen

lymphocytes, monocytes
51, VLA-5, CD49e/CD29 T lymphocytes, monocytes Fibronectin
61, VLA-6, GPIc, CD49f/CD29 T lymphocytes, monocytes Laminin
a
From References 126128.
b
Only leukocytes are listed. The 1 integrins are all expressed on nonhematopoietic cells, and 21 and 61 are expressed on platelets.
c
Only major ligands are listed.
integrins. Subsequently, we describe the com- from the top face. A divalent cation-binding
plex ectodomain architecture shared by all in- site, which physiologically binds Mg2+ , de-
tegrins, including 12 different domains, one nes the top face of the domain. The bound
MIDAS: metal
of which in the subunit is homologous to Mg2+ is ligated by ve side chains located in iondependent
the I domain. three different loops (Figure 3). The rst of adhesion site
these loops, between -strand 1 and -helix 1,
i.e., the 1-1 loop, contains three coordinat-
I Domain Structure ing residues in a sequence that is a signature of
I domains, Asp-Xaa-Ser-Xaa-Ser or DXSXS.
The I domain can be expressed indepen-
The second loop donates a coordinating Thr
dently of other integrin domains and was
residue, and the third loop donates an Asp.
the rst domain to be crystallized (4). Sev-
Divalent cations are universally required for
eral structures of I domains bound to lig-
ligand binding by integrins, and in I do-
ands are now available, including the 2 I
mains the metal-coordinating residues, and
domain bound to a triple-helical collagen pep-
the residues surrounding the metal-binding
tide (5) and L I domains with mutation-
site, are important for ligand binding. There-
ally introduced disulde bonds bound to in-
fore, this site has been designated the metal
tercellular adhesion molecule (ICAM)-1 and
iondependent adhesion site (MIDAS).
ICAM-3 (6, 7) (Figure 2). The I domain
adopts the dinucleotide-binding or Rossmann
fold, with -helices surrounding a central Conformational Regulation
-sheet (Figure 2). -strands and -helices of I Domains
tend to alternate in the secondary structure, Structural studies of I domains in the pres-
with the -helices wrapping around the do- ence and absence of ligand, and with mu-
main in counterclockwise order when viewed tations that stabilize distinct afnity states,
www.annualreviews.org Integrin Regulation and Signaling 621

surrounding the metal (Figure 3). In the

open conformation of the MIDAS, two ser-
ines and one threonine are in the primary co-
ordination sphere, whereas two aspartic acid
residues are in the secondary coordination
sphere (Figure 3b). Notably, the glutamic
acid residue, which is contributed by the lig-
and or ligand-mimetic lattice contact, donates
the only negatively charged oxygen to the pri-
mary coordination sphere in the open confor-
mation (E314 in Figure 3b). The lack of any
charged group in the primary coordination
sphere donated by the I domain is hypoth-
esized to enhance the strength of the metal-

ligand bond. In the closed conformation of
the I domain (Figure 3a), the threonine
moves from the primary to the secondary co-

ordination sphere, and one of the aspartic acid
residues moves from the secondary to the pri-
mary coordination sphere. The backbone and
side chain rearrangements in the I domain
are accompanied by a 2.3 A sideways move-
ment of the metal ion away from the threo-
nine and toward the aspartic acid on the oppo-
site side of the coordination shell. The closed
and open structures are consistent with the
Figure 2
idea that an energetically favorable MIDAS
requires at least one primary coordination to
A mutant, high-afnity L I domain (gold -sheet and coil and green
-helices) in complex with domain 1 of ICAM-3 (cyan). The Mg2+ is a negatively charged oxygen. In the absence
shown as a gray sphere. The side chain of the key integrin-binding of a ligand, pseudoligand, and the remainder
residue, Glu37 of ICAM-3, is shown. The mutationally introduced of the integrin ectodomain, wild-type I do-
K287C/K294C disulde bond that stabilizes the open conformation is mains crystallize in the closed conformation.
shown in pink. ICAM-3 domain 2 is omitted for clarity. [From Protein
The closed conformation therefore appears to
Data Bank (PDB) ID code 1T0P (7).]
be the low-energy conformation, as veried
computationally (9). However, with an engi-
have provided a mechanistic understanding of neered disulde bond that was designed to sta-
conformational regulation during both prim- bilize the open conformation, an L I domain
ing and ligand binding. I domains have been crystallized in the open conformation in the
crystallized in three distinct conformations, absence of a ligand-mimetic lattice contact (6).
termed closed, intermediate, and open (46, This demonstrates that, in principle, interac-
8). These demonstrate distinct coordination tions with other integrin domains might be
of the metal in the MIDAS, arrangement of capable of stabilizing an unliganded I domain
the 6-7 loop, and axial disposition of the in the open conformation and activating or
C-terminal 7-helix along the side of the I priming it for ligand binding.
domain (5, 6, 8) (Figure 4a). At the I do- Change in coordination at the MIDAS of
main MIDAS, ve residues and several wa- I domains is coupled to backbone move-
ter molecules contribute oxygen atoms to the ments of loops that bear the coordinating
primary and secondary coordination spheres residues. Several of these loops, including the

ANRV306-IY25-21
Annu. Rev. Immunol. 2007.25:619-647. Downloaded from arjournals.annualreviews.org ARI 11 February 2007 13:4
Figure 3
Structural rearrangement of the M I domain MIDAS. (a) Structure of the closed M I domain MIDAS.
(b) Structure of the open I domain MIDAS. Glu-314 from a neighboring M I domain coordinates
with the MIDAS magnesium. Purple and green spheres are Mn2+ and Mg2+ ions, respectively, and red
spheres are coordinating water-molecule oxygens. [From PDB ID codes 1JLM and 1IDO (4, 8).]
1-1 and 4-5 loops, also bear residues example, Figure 2 shows a complex between
that directly contact ligand, and thus their ICAM-3 and a high-afnity L I domain mu-
movement increases complementarity to lig- tant with a disulde bond introduced into the
ADMIDAS:
and. To accommodate these rearrangements, 6-7 loop to stabilize the open conforma- adjacent to metal
the 6-7 loop undergoes the largest shift tion. Conversely, binding of wild-type I do- iondependent
of all, although it is not a MIDAS loop nor mains to ligand at the extremely high, mM, adhesion site
does it contact ligand. Coupled to the 6- concentrations used in crystallization can in-
7 loop rearrangement, the C-terminal 7- duce MIDAS rearrangements and downward
helix moves 7 A down the side of the do- displacement of the 7-helix (5). Thus, the
main (Figure 4a). The axial displacement of transmission of inside-out and outside-in sig-
the 7-helix represents the critical linkage naling within the I domain occurs along
for transmission of conformational signals be- the same pathway but ows in opposite
tween the MIDAS of the I domain and other directions.
integrin domains, as discussed below. Engi- The ability to modulate afnity by 10,000-
neered disulde bonds that stabilize the 7- fold demonstrates the exquisite efciency of
helix in intermediate and open conformations the I domain in coupling change in con-
(shifted axially downward by approximately formation to change in afnity. Remarkably,
3 A or 7 A relative to the closed conforma- through directed evolution an engineered
tion, respectively) induced rearrangements in L I domain mutant (F265S/F292G) was re-
the MIDAS and surrounding loops that were cently obtained with an increase of 200,000-
coupled to 500- and 10,000-fold increases in fold in afnity for ICAM-1 (10). However,
afnity for ICAM-1, respectively (6). As men- whether I domains achieve such high afni-
tioned above, the downward movement of the ties for ligands under physiological condi-
7-helix is sufcient for priming the I do- tions is unknown. A high-afnity state of
main into higher-afnity states. Crystal struc- integrins on intact cells can be induced by
tures have been obtained of intermediate- and addition of Mn2+ , which, as reviewed below,
high-afnity mutant L I domains both in the increases integrin afnity by replacing Ca2+
absence and presence of ligands (6, 7). For at the ADMIDAS site of the I domain.

ANRV306-IY25-21
Figure 4
Conformational change and transmission of allostery by and I domains. (a) The I domain.
Nonmoving segments of the backbone are shown as a gray worm. The moving segments of the backbone
and the MIDAS metal ions are closed (gold ) and open (cyan). The direction of movement is indicated
with arrows. [From PDB ID codes 1JLM and 1IDO (4, 8).] (b) The I domain and its linkage to the
hybrid and plexin/semaphorin/integrin (PSI) domain. Nonmoving segments of the I backbone are
shown as a gray worm. Moving segments and metal ions are color coded as shown. Directions of 1- and
7-helix movements are shown with arrows. [PDB ID codes are 1U8C, 1L5G, and 1TXV (32, 36, 40).]
Interestingly, activation of integrin adhesive- pathway from the closed to the open confor-
ness on intact cells by physiologic stimuli ap- mation of the L and M I domains (12).
pears to result in lower afnity than that in- A monoclonal antibody (mAb), AL-57, has
duced by Mn2+ . Therefore, investigators have been developed by phage display that selec-
hypothesized that an intermediate conforma- tively targets the high-afnity open confor-
tional state with intermediate afnity for lig- mation of the L I domain (13, 14). AL-57
and is important for physiologic, ne-tuned does not bind the low-afnity state of leuko-
regulation of L2 adhesiveness (11). Molec- cyte functionassociated antigen (LFA)-1
ular dynamic studies showed that the interme- (L2) but does bind the intermediate- and
diate conformation of the I domain is on the high-afnity states of the L I domain with

KD of 4.7 M and 23 nM, respectively. AL-57 (Figure 5). They share much more sequence
is ligand-mimetic because it binds only upon identity with one another (30% to 50%) than
activation and requires Mg2+ for binding. In- with other IgSF molecules and thus comprise
terestingly, monovalent Fab AL-57 demon- a subfamily of the Ig superfamily. Except for
strates afnity increases on a subset (10%) ICAM-4, they all bind to L2 through a key
of lymphocyte cell surface LFA-1 molecules glutamic acidic residue in domain 1 (21). The
upon stimulation with chemokine CXCL-12 order of binding afnities for L2 is ICAM-
and PMA (phorbol 12-myristate 13-acetate). 1 > ICAM-2 > ICAM-3 (11). In the struc-
These results are consistent with previous ob- ture of the L I domain bound to ICAM-1
servations on Mac-1 on neutrophils (15) and (6) or ICAM-3 (7), Glu-34 (ICAM-1) or Glu-
suggest that after physiologic activation a sub- 37 (ICAM-3) at the end of the -strand C of
set of cell surface Mac-1 molecules on neu- domain 1 forms a direct coordination to the
trophils and LFA-1 molecules on lymphocytes Mg2+ in the I domain MIDAS (Figure 2).
are converted to a higher-afnity state. This This metal-coordination bond is surrounded

active subset of molecules mediates adhesion by a ring of hydrophobic residues in both
because the antibodies specic for this sub- the I domain and ICAM-1. The surround-
set of 10% to 30% of surface molecules com- ing nonpolar environment strengthens the
pletely inhibit cell adhesion. Coulombic interaction between the Glu and
Mg2+ . Surrounding the hydrophobic ring are
polar interactions involving hydrogen bonds
Allosteric I Domain Inhibitors and salt bridges. The nonpolar region in
Small molecule allosteric inhibitors provide ICAM-1 is more polar in ICAM-3 and ap-
further support for the role of the 7-helix pears to account for the lower afnity of L2
in I domain regulation. One class of small for ICAM-3 than for ICAM-1 (7). In sup-
molecule inhibitors, termed I allosteric an- port of this nding, increasing the hydropho-
tagonists, binds underneath the C-terminal bicity of the nonpolar ring even further in
-helix of the L I domain (1618). Such ICAM-1 by structure-guided mutagenesis in-
antagonists stabilize the closed conformation creases the afnity of L2 for ICAM-1 (22).
of the I domain by preventing downward ax- ICAM-1 and ICAM-3 dock in the same orien-
ial shift of the 7-helix and thereby prevent- tation to the L I domain, and the structure of
ing MIDAS rearrangements necessary for ef- ICAM-2 (23) suggests an essentially identical
cient ligand binding. The mode of action of docking mechanism.
these antagonists is conrmed by the nd- The structure of the L I domain bound
ing that a mutant L I domain that stabi- to a portion of ICAM-1 (6) combined with a
lizes the high-afnity, open conformation of complementary structure containing the re-
the C-terminal 7-helix with an engineered maining portion of ICAM-1 (24) have pro-
disulde bond is resistant to inhibition by I vided a topological view of the L2-ICAM-
allosteric antagonists (11, 19, 20). 1 interaction as it might take place during
cell-cell interactions (Figure 5b). L2- and
M2-binding sites on ICAM-1 are located
Ligand Recognition by I Domains on D1 and D3, respectively. ICAM-1 exists
Ligand recognition by I domains has been on the cell surface predominantly as a ho-
elucidated by crystal structures of the 2 I modimer (25, 26). Relatively strong but re-
domain in complex with a triple-helical col- versible dimerization takes place in D4 by
lagenous peptide (5) and the L I domain merging of the -sheets of two D4 do-
in complex with ICAM-1 and ICAM-3 (6, mains into -supersheets, as revealed by an
7). ICAM-1, -2, -3, -4, and -5 are cell sur- ICAM-1 D3-D5 crystal structure (24). An
face molecules with 2 to 9 IgSF domains apparently weaker hydrophobic dimerization

Figure 5
ICAM structure
and integrin
binding.
(a) Schematic of
ICAM-1, -2, -3,
and -5. The
domains are color
coded, and
integrin-binding
sites are shown. (b)
Structural model of
ICAM-1 oligomers
bound to L I
domain. The
model was
constructed from
the structure of
ICAM-1 D1-D2 in
complex with L I
domain (PDB ID
code 1MQ8) and
from the structure
of ICAM-1 D3-D5
(PDB ID code
1P53) (6, 24).
ICAM-1 is cyan,
with the rst
carbohydrate
residue at each site
in yellow; the L I
domain is purple.
interface in D1 has been revealed in differ- yielding a one-dimensional array of ICAM-

ent crystal structures, including the ICAM- 1 molecules on the cell surface (Figure 5b),
1 D1-D2 complex with the L I domain which has an architecture appropriate for dis-
(6). Together, the L I domain-ICAM-1 D1- playing the L2-binding site in D1 and the
D2 and the ICAM-1 D3-D5 structures show M2-binding site in D3 for recognition by
that ICAM-1 dimers are Y-shaped and that integrins on an opposing cell (24).
the dimeric interface at D4 and D5 provides In contrast to L2, the M2 and X2
a rigid stem to orient D1-D3 optimally for integrins bind to a range of diverse ligands,
binding integrins on opposing cell surfaces including ICAMs, brinogen, iC3b, heparin,
(24) (Figure 5a). Furthermore, dimerization and denatured and proteolyzed proteins (27
at D1 could link neighboring Y-shaped dimers 29). Proteolysis and denaturation enhance

binding of M and especially of X I domains associated with a large-scale global confor-

to brinogen, and investigators have proposed mational rearrangement in which the inte-
that X2 functions as a danger receptor for grin extends with a switchblade-like motion
proteolyzed and denatured proteins (29). In (3336) (Figure 6c,d ). The most recent stud-
marked contrast to L, the X and M I do- ies have elucidated the detailed mechanisms
mains show a KD for small molecules with car- for linking these global rearrangements to in-
boxyl groups of 100 M. This relaxed ligand tradomain conformational changes that are
specicity is consistent with the ability of the associated with afnity modulation and ligand
wild-type M I domain to engage in ligand- binding.
mimetic lattice contacts in crystals. In these
contacts, the MIDAS binds to a glutamic acid
residue in a neighboring M I domain in the The Ligand-Binding Head
crystal lattice (Figure 3b), and the I domain The N-terminal region of the integrin
crystallizes in the open conformation (4). subunit contains seven segments of about
60 amino acids, each with weak sequence sim-
ilarity to one another. These were initially
INTEGRIN GLOBAL TOPOLOGY predicted (37) and later conrmed by crystal

A complete understanding of integrin regu- structures (31, 36) to fold into a seven-bladed
lation requires knowledge of how conforma- -propeller domain. When present, the I
tional information is transmitted through the domain is inserted between -sheets 2 and 3
many domains that link the ligand-binding of the -propeller (Figure 6a,b). An inserted
domains to the transmembrane and cytoplas- domain in the subunit, the I domain, is
mic domains. Both the integrin and homologous to the I domain, except that it
subunits are type I transmembrane glycopro- contains two additional segments; one of these
teins with large extracellular domains, single- helps form the interface with the -propeller
spanning transmembrane domains, and, with and the other is known as the specicity-
the exception of 4, short cytoplasmic do- determining loop (SDL) because of its role in
mains (Figure 6a,b). From electron mi- ligand binding (Figure 6a,b and Figure 7c).
croscopy (EM), investigators have known for One side of the I domain binds to the upper
years that the overall topology of integrins in- hub of the -propeller domain directly over
cluded an extracellular, globular, N-terminal the pseudosymmetry axis of the -propeller
ligand-binding head domain, representing a (Figure 7c). The extensive interface, which
critical and subunit interface, standing buries 1700 A2 of solvent-accessible surface
on two long and extended C-terminal legs or area on each side, is much greater than any
stalks, which connect to the transmembrane other domain-domain interface in integrins,
and cytoplasmic domains of each subunit (30). including interfaces between domains that are
However, X-ray crystal structures of the ex- contiguous in sequence in the and sub-
tracellular domain of the integrin V3 pro- units. Mutations in the 2 I domain that dis-
vided the surprising nding that the legs were rupt this interface cause leukocyte adhesion
severely bent, generating a V-shaped topology deciency (38, 39). The opposite, lower face
in which the head domain was closely juxta- of the subunit -propeller domain is stabi-
posed to the membrane-proximal portions of lized by Ca2+ ions that bind to Ca2+ -binding
the legs (31, 32) (Figure 6c and Figure 7a). -hairpin motifs (Figure 7b,c). Like the I
Since the elucidation of these initial struc- domain, the I domain contains a MIDAS for
tures, an increasing number of studies have binding negatively charged residues, which
together established that the bent conforma- physiologically binds Mg2+ (36). Addition-
tion represents the physiological low-afnity ally, there are two adjacent metal ionbinding
state, whereas priming and ligand binding are sites, which physiologically bind Ca2+ (36),

a I domain
b
Headpiece
Head
Thigh Calf-1 Calf-2
Genu TM Cytoplasmic
domain N
PSI domain Upper PSI Hybrid
leg Thigh
Hybrid Cytoplasmic I-EGF1
I-EGF domain N I-EGF2
Tailpiece
1 2 3 4 Calf-1 I-EGF3
Lower
leg I-EGF4
Calf-2
TM TM
Cyto Cyto
C C
c 2 3
1 Hybrid
Thigh
PSI
Genu
Calf-1 I-EGF1-4
Calf-2
TM TM
Extrinsic ligand
Closed headpiece, Closed headpiece, Open headpiece,
bent extended extended
3
2
d I domain
1 Hybrid
Thigh
PSI
Genu
Calf-1 I-EGF1-4
Calf-2
TM TM
Closed headpiece, Closed headpiece, Open headpiece,

bent extended extended

share some coordinating residues in com- mains represents the subunit genu, the key
mon with the MIDAS, and are known as the pivot point for switchblade extension in the
LIMBS (ligand-induced metal ionbinding subunit (Figure 6 and Figure 7a).
I-EGF domain:
site) and ADMIDAS (adjacent to metal ion The topology of the subunit is more integrin epidermal
dependent adhesion site) (31, 32) (Figure 4b complex. The I domain is inserted in growth factorlike
and Figure 7b). the hybrid domain, which forms the upper domain
Structures of V3 (32) and IIb3 portion of the upper leg (Figure 6a,b).
(36) show that peptides containing ligand- In turn, the hybrid domain is inserted in
mimetic Arg-Gly-Asp (RGD) sequences bind the plexin/semaphorin/integrin (PSI) domain
across the - subunit interface in the head (Figure 6a,b). The second segment of the PSI
(Figure 7c,d ). The Asp carboxyl group di- domain is very short but can be assigned as
rectly coordinates the I domain MIDAS part of the PSI domain because it contains
Mg2+ ion, whereas the Arg side chain hy- 3-Cys435, which is involved in a long-range
drogen binds to Asp residues in the V or disulde bond to 3-Cys11 in the rst seg-
IIb -propeller domains (Figure 7d ). The ment of the PSI domain, and this disulde is
binding site for macromolecular ligands is structurally conserved in other PSI domains
larger. Residues shown by mutagenesis to be (36, 40, 41).

important for binding to brinogen (smaller The remainder of the leg is built from
spheres, Figure 7c) decorate the cap subdo- four integrin epidermal growth factorlike (I-
main of the -propeller (in Figure 7c, green), EGF) domains and a tail domain. I-EGF do-
the remainder of the -propeller (magenta), mains 1 and 2 were not resolved in the V3
and the I domain (cyan). The cap subdo- crystal structure. However, an NMR struc-
main is formed by several insertions that are ture of 2 I-EGF3 and studies on I-EGF2
unusually long in IIb in -propeller domain and I-EGF3 established an extended orienta-
-sheets (blades) 2 and 3. The ligand-binding tion between these domains. Furthermore, a
site in the -propeller is formed largely by structure of the 2 PSI, hybrid, and I-EGF1
-sheets 2 and 3, which lie opposite the domains has been solved (41). Superposition
ligand-binding MIDAS and SDL of the I of these structures on the bent V3 structure
domain. establishes that the bend in the leg, or knee,
is located between I-EGF domains 1 and 2, as
suggested earlier (42). The PSI and I-EGF1
The and Subunit Legs domains are located side by side (Figure 6b).
In the subunit, the region C-terminal to The bends in the leg at the genu and in the
the -propeller comprises the leg of the leg between I-EGF1 and I-EGF2 are located
subunit and contains three -sandwich do- close to one another and in a geometry appro-
mains. The upper leg contains the thigh do- priate so that extension can occur by pivoting
main and the lower leg consists of the calf-1 of the headpiece about an axis through the
and calf-2 domains. A small Ca2+ -binding and subunit knees (Figure 6c,d ), as shown
loop located between the thigh and calf-1 do- by EM studies (33, 43). Consistent with these

Figure 6
Integrin architecture. (a) Organization of domains within the primary structures. Some subunits
contain an I domain inserted in the position denoted by the dashed lines. Cysteines and disuldes are
shown as lines below the stick gures. (b) Schematic of the course of the and subunit polypeptide
chains through domains from the N to C termini. (cd ) Rearrangement of domains during activation of
integrins that lack (c) or contain (d ) an I domain. The subunit lower legs are exible and are
therefore shown in what may be the predominant (solid representation) and less predominant (dashed lines)
orientations.

ANRV306-IY25-21
Figure 7
Crystal structures of integrins V3 and IIb3. (a) The structure of V3 in the bent conformation.
The V and 3 subunits are colored in green and red, respectively. (b) Superposition of liganded-open
IIb3 and unliganded-closed V3 headpieces. The and subunits are colored in green and yellow
in V3 and in purple and light blue in IIb3. Calcium and magnesium ions in IIb3 only are gold
and gray spheres, respectively. (c) The drug tiroban is shown bound to the IIb3 head, and mapping is
shown of brinogen bindingsensitive mutations. The clinically approved antagonist tiroban is shown
with yellow carbons, blue nitrogens, and red oxygens. The cap subdomain of the -propeller is in green.
Ca2+ and Mg2+ ions are large gold and gray spheres, respectively. C atoms of brinogen
bindingsensitive residues are shown as small spheres in the same color as the domains in which they are
present. Disulde bonds are yellow cylinders. (d ) The binding of eptibatide to IIb3 interface is
depicted. The fragment of eptibatide that mimics RGD is shown as a stick model with carbon, nitrogen,
and oxygen colored yellow, blue, and red, respectively. The binding pocket is shown with IIb and 3 in
purple and light blue, respectively. Hydrogen bonds are shown as gray dashed lines. Ca2+ and Mg2+ are
gold and gray spheres, respectively. The coordinations to the metal ions are shown as green dashed lines.
[Structure PDB ID codes are, for V3, 1U8C (40); for IIb3 bound to eptibatide, 1TY6; and for
IIb3 bound to tiroban, 1TY5 (36).]

ndings, many antibodies that either activate pled, so that reshaping to the high-afnity,
or report activation in cell surface integrins ligand-binding site is allosterically linked to
map to the PSI and I-EGF domains (44 downward movement of the 7-helix. This
47). Furthermore, L antibodies that report linkage is critical for propagation of con-
extension map to the inner face of the thigh formational signals from the ligand-binding
domain and require genu Ca2+ -coordinating pocket to the other integrin domains and vice
residues for binding and thereby provide evi- versa (Figure 4b). When the RGD-mimetic
dence that integrin extension occurs by a re- is soaked into preformed crystals (liganded-
arrangement at the thigh/genu interface (48). closed, Figure 4b), the 1-1 loop moves al-
most as much but does not have as optimal
an interaction with ligand as in the liganded-
CONFORMATIONAL open structure, and the remaining movements
REGULATION OF INTEGRIN are frustrated by crystal lattice contacts.
EXTRACELLULAR DOMAINS
Conformational Activation Effects of Mn2+ and Ca2+
of I Domains
Compared with results in the physiologic di-
In the bent conformation, the ligand-binding valent cations Mg2+ and Ca2+ , which are
site is not in an optimal orientation for bind- present at 1 mM in body uids, addition of
ing macromolecular ligands in the extracel- Mn2+ or removal of Ca2+ increases ligand-
lular matrix or on the surface of other cells binding afnity and adhesiveness of almost all
(Figure 6c,d ). However, integrins in the bent integrins. Recent studies show that binding of
conformation can bind ligands, as clearly metal ions to the LIMBS and ADMIDAS sites
shown by soaking an RGD ligand-mimetic can explain these effects (49, 50). Mutational
peptide into preformed crystals (32). Ligand studies show that the LIMBS functions as a
binding induced movements in the 1-1 and positive regulatory site, and the ADMIDAS
6-7 loops (liganded-closed conformation, functions as a negative regulatory site (4951).
Figure 4b). However, downward displace- Additionally, in 51 and L2 integrins, the
ment of the 7-helix was not seen (32). ADMIDAS functions in transmission of al-
When an IIb3 headpiece was rst mixed lostery between the I domain and other do-
with ligand-mimetic drugs, and then crys- mains (50, 51). For most integrins, Ca2+ has
tals were allowed to form, a different confor- both positive and negative regulatory effects.
mation, termed liganded-open, was obtained High concentrations of Ca2+ inhibit adhe-
(36). In the high-afnity, liganded-open sion, whereas low concentrations of Ca2+ syn-
I domain, compared with the low-afnity, ergize with suboptimal Mg2+ concentrations
unliganded-closed I domain, marked move- to support adhesion. The LIMBS mediates
ments occur of the 1-1 and 6-7 loops the synergistic effects of low Ca2+ concentra-
and of the 1- and 7-helices (Figure 4b). tions (49, 50), whereas the ADMIDAS medi-
Coordination of the Met335 backbone car- ates the negative regulatory effects of higher
bonyl in the 6-7 loop to the ADMIDAS Ca2+ concentrations, which are competed by
Ca2+ ion in the low-afnity, unliganded con- Mn2+ (49).
formation is broken in the high-afnity,
liganded conformation. This enables the
ADMIDAS metal and residues in the 1-1 Communication Between the I
loop that coordinate to both the ADMIDAS and I Domains
and MIDAS metals to shift markedly, remodel Conformational regulation of integrins con-
the ligand-binding site, and increase afnity taining an I domain requires the additional
for ligand. These movements are tightly cou- step of transmission of allostery from the I

I domain I domain
Inactive MIDAS
I-like I-like
domain domain
Active MIDAS
Hybrid Hybrid Invariant Glu residue

domain domain
subunit subunit subunit subunit
b
I domain I domain I domain

C
C
C C
I-like I-like I-like
domain domain domain
Hybrid Hybrid Hybrid

domain domain domain
subunit subunit subunit subunit subunit subunit
L-E310C 2-A210C L-E310C / 2-A210C

Inactive Inactive Constitutively active
Figure 8
Communication between I and I domains. (a) It has been proposed that L-Glu-310 acts as an
intrinsic ligand that binds to the 2 I domain MIDAS and, thus, axially displaces the L I domain
7-helix in the C-terminal direction, reshapes the 6-7 loop, and activates the L I domain MIDAS.
(b) Individual mutation of L-Glu-310 or 2-Ala-210 to cysteine abolishes I domain activation, whereas
the double mutation of L-E310C with 2-A210C forms a disulde bond that constitutively activates
ligand binding (104).
domain to the I domain (Figure 6d ). An domain (53, 54) (Figure 8a). Yang et al. (54)
invariant Glu residue, E-310 in L, in the provided direct evidence for an activating in-
linker between the C-terminal 7-helix of the teraction between L residue 310 and the 2
I domain and -sheet 3 of the -propeller MIDAS. Individual mutation of the L linker
domain is required for I domain activation residue Glu-310 or 2 MIDAS residue Ala-
(52, 53). It has been proposed that this in- 210 to cysteine abolishes I domain activation,
variant Glu residue acts as an intrinsic lig- whereas the double mutation of L-E310C
and and binds to the I MIDAS when it and 2-A210C results in formation of a disul-
is activated and that it exerts a downward de bond that constitutively activates ligand
pull on the 7-helix and activates the I binding (54) (Figure 8b).

/ I Domain Allosteric Antagonists where in the LFA-1 molecule. This mutant

is also decient in supporting rolling interac-
Small molecule integrin antagonists have
tions on ICAM-1 substrates. However, treat-
yielded further insight into the mechanisms
ment of L Glu-310 LFA-1 mutants with /
for I I communication. A class of L2
I allosteric antagonists induces epitope ex-
and M2 small molecule antagonists per-
posure and renders them competent to sup-
turbs the interface between the I domain
port rolling, consistent with the hypothesis
and the I domain (5557). These antago-
that these antagonists bind to the same site
nists do not inhibit binding of isolated wild-
to which L Glu-310 binds (59).
type or mutant intermediate- or high-afnity
I domains to ICAM-1 (56). Furthermore,
these inhibitors bind to L2 even when the
I domain is deleted, but do not bind when Hybrid Domain Swing-Out
and Integrin Extension
the I domain MIDAS is mutated (56). Some

but not all compounds of the series exhibit The orientation between the I and hybrid
subunit selectivity, suggesting that a portion domains appears to be the critical translator
of the subunit nearby the I domain, likely converting global conformational change into
the -propeller domain or its linkers to the local intradomain conformational changes
I domain, is involved in binding. There- that regulate integrin afnity for ligand (see
fore, these inhibitors have been designated as Figure 4b and Figure 6c,d ). As a conse-
/ I domain allosteric antagonists. quence of the topology of insertion of the
These antagonists apparently bind to the I domain in the hybrid domain, the piston-
MIDAS of the 2 I domain, competitively like displacement of the 7-helix in the high-
inhibit binding of the intrinsic ligand in the afnity, liganded structure results in complete
I domain linker, and thus leave the I do- remodeling of the interface between these do-
main in its default low-energy, inactive, closed mains, leading to the swing-out of the hy-
conformation. At the same time, the / I brid domain (36) (Figure 4b). Actually, the
allosteric antagonists mimic intrinsic ligand 7-helix functions more like a connecting rod
binding and thereby stabilize the I domain than a piston because as it moves downward,
in its active conguration, as shown by induc- its angle changes (Figure 4b), like a connect-
tion of activation epitopes in the 2 I domain, ing rod between a piston and a crankshaft.
as well as the L and 2 legs (56). The antag- This forces rotation about a crankshaft bear-
onists induce integrin extension as shown in ing (circled in Figure 4b) between the last
EM studies (43). In in vitro shear ow assays -strand of the hybrid domain and the rst
and in vivo, the antagonists enhance rolling -strand of the I domain. Note the struc-
of leukocytes and inhibit rm adhesion (57). tural design of this machine: Hydrogen bonds
These results on ICAM-1 substrates suggest in -helices are all internal, allowing them to
that the postulated L Glu-3102 MIDAS move independently of other structural units,
interaction is not required for rolling adhe- whereas the three other connecting units be-
sion, in agreement with the ability of isolated, tween the I and hybrid domains are -
wild-type L I domains to support rolling strands, which are xed in position within -
adhesion (58, 59), but is required for rm sheets by hydrogen bonds. Therefore, there
adhesion. LFA-1 containing an L Glu-310 is little compliance in the central I domain
Ala mutation shows lowered expression of -sheet or the two hybrid domain -sheets.
activation epitopes in the 2 I domain and This enables the rearrangement of the loops
leg, demonstrating cooperativity between the around the I domain MIDAS to be trans-
postulated L Glu-3102 MIDAS interac- mitted as a 60 swing of the hybrid domain
tion and conformational rearrangements else- away from the subunit and a 70 A movement

of the rigidly connected PSI domain, i.e., a main is highly buried in the interfaces that
70 A separation of the integrin knees (36, 41) stabilize the bent structure, and therefore
(Figure 4b). its swing-out destabilizes the bent conforma-
Three major integrin conformations have tion (33). By contrast, the extended integrin
been resolved by crystal and EM studies conformation is compatible with both closed
(Figure 6c,d ). The bent conformation con- and open headpiece conformations (33, 43)
tains a closed headpiece (31). The hybrid do- (Figure 6c,d and Figure 9a,b). The inu-
ence on equilibration between these states has
been studied of a exible, C-terminal clasp
fused to the C termini of the and sub-
unit ectodomains, which mimics association
between the and subunit transmembrane
domains (33, 43). Whereas clasped V3 or
X2 particles are predominantly in the bent

conformation (see panel 1 of both Figure 9a
and b), unclasped particles are predominantly
extended. For V3 and X2, about half of

unclasped, extended particles have the closed
headpiece (see panel 2 of both Figure 9a
and b) and half have the open headpiece (see
panel 3 of both Figure 9a and b) (33, 43).
Therefore, once these integrins extend, the
energies of the closed and open headpiece
conformations must be comparable. How-
ever, the energetics for conformational tran-
sitions appear to vary among integrins, as ex-
emplied with L2. Thus, clasped L2
shows about equal proportions of bent and
extended particles, and unclasped L2 par-
ticles adopt predominantly the closed head-
piece, with a smaller proportion of particles
having the open headpiece (43).
A large number of studies are in agree-
ment with the three integrin conformational
states described above and support the im-
portance of hybrid domain swing-out in in-
Figure 9 ducing high afnity for ligand. EM stud-
EM negative-stain class averages of integrins V3 and X2 in bent ies of the 51 headpiece show that hybrid
and extended conformations (33, 43). The EM images of the extended domain swing-out occurs upon binding -
conformations only are colored according to the scheme shown in d. (a)
bronectin fragments (34). Electron tomogra-
V3 in bent ( panel 1), extended with closed headpiece ( panel 2), and
extended with open headpiece ( panel 3) conformations. (b) X2 in bent phy of negatively stained, active, detergent-
( panel 1), extended with closed headpiece ( panel 2), and extended with soluble IIb3 puried on an RGD peptide
open headpiece ( panel 3) conformations. (c) X2 in complex with CBR afnity column reveals an extended confor-
LFA-1/2 Fab illustrates exibility of the leg: panel 1, closed headpiece mation with >90% of particles showing an
with parallel legs; panel 2, closed headpiece with crossed legs; panels 3
open headpiece structure that matches per-
and 4, open headpiece. Panels 13 are with clasped X2, and panel 4 is
with unclasped X2. In ac, a schematic in the same orientation as the fectly (60) the open, liganded IIb3 head-
right-most panel is shown to the right; the dashed lower legs symbolize piece crystal structure (36). In addition to
exibility and averaging-out. structural investigations (33, 34, 36, 43, 60,

61), integrin hybrid domain swing-out is sup- bodies have been mapped to the I-EGF2 and
ported by a range of other studies. Stabilizing I-EGF3 domains, respectively (45, 77). Al-
the open headpiece by mutationally introduc- though clasped X2 was >95% in the bent
ing an N-glycosylation site into the hybrid conformation, binding of CBR LFA-1/2 Fab
I domain interface increases ligand-binding induced complete conversion to the extended
afnity (62, 63). As shown by epitope map- conformation (43) (Figure 9c, panels 13).
ping and EM, an allosteric 1 antibody that Furthermore, KIM127 Fab bound only when
inhibits ligand binding restricts the swing-out extension was induced by another agent such
of the hybrid domain (63). The functional as CBR LFA-1/2 Fab or an / allosteric an-
properties of an inhibitory 2 mAb suggest it tagonist. In combination with the following
also inhibits by blocking signal transmission at cited functional studies, the EM study (43) es-
the Ihybrid domain interface (64). More- tablished that (a) extension is sufcient to ac-
over, activation-dependent mAbs map to the tivate ligand-binding competence by 2 inte-
inner face of the hybrid domain, consistent grins (35, 45, 54, 7276, 78), (b) ligand-bound
with exposure of this face after swing-out (65, 2 integrins on cell surfaces are extended (70,
66), and specic mutations of the I domain 71), (c) binding to soluble ligand induces ex-
7-helix facilitate hybrid domain swing-out tension (57), and (d ) extracellular activation of
(65). integrins by Mn2+ and inside-out activation
In contrast to the consensus in the above of integrins stimulated by protein kinase C or
studies, an alternative deadbolt model (67) cytoplasmic domain mutations induce the ex-
has received little experimental support, as re- tended conformation in the absence of ligand
viewed in more detail elsewhere (68). The pre- binding (35, 45, 75).
sumed deadbolt interface between 3 Val332 When viewed in combination, the crystal
and Ser674 is extremely small, at 60 A2 and EM studies demonstrate two structurally
(67), and a three-residue deletion of 3 linked mechanisms for activating integrin ad-
residues 672674 that removes this interface hesiveness. First, extension moves the ligand-
has no effect on ligand binding to cell sur- binding head 100 A to 200 A further away
face integrins V3 and IIb3 ( J. Zhu, B. from the cell surface and orients it optimally
Luo & T.A. Springer, unpublished observa- for adhesion to another cell or to the extracel-
tion). One negative-stain EM study found lular matrix. Second, extension enables hybrid
that ligand binding to V3 did not in- domain swing-out, which induces increased
duce extension (69); however, much greater afnity for ligand.
particle aggregation was present than in
other studies and must have complicated the
analysis. The Compliant Integrin Legs
The use of functionally well-characterized The design of the connecting rod and
antibodies in EM experiments has provided crankshaft bearing between the I and hy-
denitive evidence that integrin extension brid domains and the rigidity of the hybrid
occurs on intact cells in response to phys- domain/PSI domain unit amplify reshaping
iologic stimuli and is sufcient to activate of the ligand-binding site into a 70 A separa-
integrin adhesiveness (43). Extensive, physio- tion at the integrin knees. Such a large move-
logically relevant studies of 2 integrins on in- ment appears to be important for transmission
tact cells have shown that CBR LFA-1/2 mAb of conformational change to the transmem-
induce the high-afnity state and that, de- brane and cytoplasmic domains because the
pending on the experimental system, KIM127 leg in particular is highly compliant, i.e., ex-
mAb can either stabilize or report the high- ible. Below, we discuss the role of integrin
afnity state (35, 45, 54, 57, 7076). The bind- and subunit transmembrane domain sepa-
ing sites for KIM127 and CBR LFA-1/2 anti- ration in activation. Transmembrane domain

separation, extension, and hybrid domain that do not heterodimerize causes constitutive
swing-out are linked; however, this linkage activation of L2 (84). Fluorescent proteins
is not tight because of the exibility of the were fused to the L and 2 cytoplasmic do-
lower leg. When extended V3 or X2 mains for uorescent resonance energy trans-
particles are imaged and class averaged, the fer (FRET) studies. These studies on live cells
domains in the lower leg tend to disappear show that in the resting state the integrin
because they appear in different orientations and subunit cytoplasmic domains are close
and are averaged out (33, 43) (panels 2 and to one another (35). However, they undergo
3 in both Figure 9a and b). Fab binding re- signicant spatial separation upon inside-out
sults in better resolution of the lower leg, activation induced by stimulation of protein
both in clasped (Figure 9c, panels 13) and kinase C, stimulation by a chemoattractant of
unclasped preparations (Figure 9c, panel 4). a G proteincoupled receptor, or transfection
Both parallel and crossed orientations of the with the talin head domain, which binds the
and legs are seen (Figure 9c, panels 1 integrin cytoplasmic domain. Furthermore,
and 2, respectively), and the leg is clearly extracellular addition of Mn2+ and soluble
exible above the Fab-binding sites in I-EGF ICAM-1, which induces integrin extension as
domains 2 and 3, i.e., at the knee between shown by exposure of the KIM127 epitope,
I-EGF1 and I-EGF2, and appears to be ex- also induces and subunit cytoplasmic do-
ible at other locations as well. This exibil- main separation (35).
ity is symbolized with the dashed legs in NMR studies of the integrin cytoplas-
Figure 6c,d. In V3 the leg snaps into mic tails suggest that their association is
a preferred orientation when it is extended weak, with signicant differences between
(33) (Figure 9a, panels 2 and 3). In X2 published structures of associated cytoplasmic
the extended leg is exible at the genu (43) domains (85, 86) or with undetectable asso-
(Figure 9b,c). ciation between and subunit cytoplasmic
domains (87, 88). These studies imply that the
cytoplasmic interaction is modest and/or tran-
CONFORMATIONAL CHANGE sient and that other domains are required for
OF THE INTEGRIN stable and association. Binding of intracel-
CYTOPLASMIC AND lular proteins such as RAPL (89) and the talin
TRANSMEMBRANE DOMAINS head domain (9092) to the integrin cytoplas-
In the bent V3 crystal structure, the and mic tails activates integrins for ligand binding,
subunit ectodomain C termini are a few presumably by disrupting and association.
angstroms apart (31), consistent with associ- Other proteins also bind to the cytoplasmic
ation of the and subunit transmembrane tails, including lamin, which competes with
domains. Ectodomain constructs with the C talin for binding to the tail and modulates
termini clasped have lower afnity for ligand cell migration (93), and ICAP-1, which binds
than unclasped constructs (79). Many stud- to the same motif as talin and has a related fold
ies show that deletions or mutations in the (94). The structural basis for talin and lamin
and subunit transmembrane and cytoplas- binding to the integrin cytoplasmic domain
mic domains, which are expected to destabi- has been demonstrated by NMR and crystal
lize / association, activate integrins (80 studies (91, 95, 96).
83). Furthermore, replacement of the L and Mutational studies have dened interfaces
2 cytoplasmic domains with ACID/BASE on the integrin and subunit transmem-
peptides that form a heterodimeric -helical brane domains that, when substituted, result
coiled-coil stabilizes L2 in an inactive state, in activation (97100). Furthermore, disul-
whereas replacement with similar peptides de scanning of the exofacial portions of

the transmembrane domains revealed a spe- to enable a substantial proportion of integrin

cic -helical interface between the and molecules to equilibrate to the high-afnity,
transmembrane domains in the resting state open headpiece conformation. The set point
(97). Disulde scanning also revealed that af- for the equilibria between bent and extended
ter activation from inside the cell, the and conformations and between extended open
subunits moved apart in the membrane headpiece and extended closed headpiece con-
instead of rearranging into a distinct / formations is integrin dependent (43) and
interface. may help explain differences between in-
tegrins in their susceptibility to activation
(101).
A MODEL FOR BIDIRECTIONAL The mechanochemical design of inte-
SIGNAL TRANSMISSION BY grins favors extension and hybrid domain
INTEGRINS ACROSS THE swing-out when integrins function in ad-
PLASMA MEMBRANE hesion. The distance in the subunit be-

Based on the preponderance of the results de- tween the ligand-binding site and the I-EGF1
scribed above, the following model is evident. domain is 20 A further (x) in the open
Integrins are in equilibrium between differ- than in the closed headpiece conformation
ent conformational states (Figure 6c,d ). The (see Figure 4b). Therefore, in cell migra-
bent conformation is stabilized by interfaces tion or as a consequence of cytoskeleton con-
between the headpiece and the lower legs, traction, when tensile force (F) is exerted
between the lower and legs (33), and on a ligand-bound integrin and resisted by
between the and transmembrane and cy- cytoskeletal proteins bound to the sub-
toplasmic domains. However, none of these unit cytoplasmic domain, the open headpiece
interfaces is tight, and small perturbations can conformation will be stabilized relative to
readily shift the equilibrium toward extension the closed headpiece by approximately Fx.
and separation. Perturbation of the cytoplas- Notably, the extended conformation would
mic domains by mutation or by binding of similarly be favored over the bent confor-
the talin head domain or other effector pro- mation and has a substantially greater x.
teins induces separation of the cytoplasmic Thus, a mechanochemical switch favors the
and transmembrane domains. This in turn re- high-afnity state when tensile force is ap-
sults in separation of the and lower legs. plied to integrins, and this is expected to
Lower leg separation destabilizes the interface be of great importance for force resistance
between the lower legs and the headpiece and and mechanotransduction by integrins dur-
results in integrin extension. Transmembrane ing cell adhesion and migration (102). This
domain separation would favor the open over mechanochemical design stabilizes the high-
the closed headpiece because the upper and afnity state when tensile force is applied to
legs are 70 A further apart in the open than selectins and integrin I domains as well, and
the closed headpiece. However, because the the importance of how force is linked to al-
lower leg is highly exible, transmembrane lostery has been experimentally demonstrated
domain and lower leg separation and exten- for I domains (103).
sion would not be sufcient to enforce hy- Conformational change can also be trans-
brid domain swing-out [compare panel 2 of mitted from the integrin ligand-binding site
Figure 6c or d (with dashed leg) with panel to the cytoplasm, as demonstrated with FRET
3 of Figure 6c or d (with solid leg)]. EM (35). Which integrin conformation rst binds
studies and results with activation-dependent ligand is unknown and may depend on (a) the
antibodies demonstrate that extension is suf- rate of equilibration between different con-
cient to induce integrin adhesiveness and formational states, (b) the population of the

different states, and (c) the binding kinetics soluble ligand-binding assays and studies with
and afnities of the different states. However, Fabs specic for the high-afnity conforma-
the preponderance of EM and crystal struc- tion have clearly demonstrated rapid and tran-
ture studies demonstrates that once ligand is sient integrin afnity regulation in response
bound, it stabilizes the extended conforma- to chemokines (14, 106108). Furthermore,
tion with the open headpiece. Swing-out of sensitive assays often demonstrate that phys-
the hybrid domain would favor, but seems iological stimuli normally induce markedly
unlikely to enforce, transmembrane domain less soluble ligand binding than Mn2+ (109),
separation because of the exibility of the which has been commonly employed as a pos-
lower leg (compare integrins with dashed itive control for afnity regulation.
and solid legs in panel 3 of Figure 6c or The idea of clustering as a mode of prim-
d ). In agreement, a disulde bond between ing implies proactive and directed mecha-
the exofacial portions of the and subunit nisms for lateral redistribution (110). Vesic-
transmembrane domains does not prevent ex- ular trafcking (111, 112) and Rap1- and
tracellular agents such as Mn2+ and antibod- RAPL-driven polarization of integrins to the
ies from activating ligand binding, although lamellapodium (89, 113) represent important
it does prevent intracellular signals from ac- active modes of integrin reorganization that
tivating ligand binding (97). It seems likely take place during cell migration. However,
that the stability of and subunit trans- mechanistic support for active reorganization
membrane and cytoplasmic domain associa- of integrins during the initial stages of prim-
tion is low, and that in the absence of close ing remains tenuous. On the basis of the ob-
association between the and ectodomain servation that peptides containing integrin
C-terminal segments, the transmembrane and and subunit transmembrane domains form
cytoplasmic domains spontaneously dissoci- homodimers and homotrimers in detergent,
ate and thereby transmit signals into the investigators have proposed that homomeric
cell. association between the transmembrane do-
mains can induce integrin clustering (114).
However, several subsequent studies in in-
Role and Regulation of Integrin tact cells have shown that homomeric -
Lateral Association or - association does not occur as a con-
As discussed above, conformational mecha- sequence of integrin priming and dissociation
nisms for regulating integrin afnity have be- of transmembrane domain heterodimers
come relatively well established. However, the (75, 98). Other studies have implied a role for
role and regulation of integrin lateral redistri- cholesterol-rich lipid rafts in driving integrin
bution on the cell surface, often referred to as clustering, but this has remained controversial
clustering, has remained unclear and contro- (104).
versial (104). Several early studies suggested Many of the ideas on activation-induced
a dominant and proactive role for integrin integrin clustering have been replaced by
redistribution in the initiation or priming of an emerging model of multivalent ligand-
cells for efcient ligand binding (105). In prac- dependent, mass-action-driven integrin re-
tice, such valency regulation (104) has usually distribution that is modulated by the
been inferred when activators promote cell cytoskeleton (104). On resting cells 2 inte-
adhesion without promoting detectable sol- grin mobility is conned by cytoskeletal inter-
uble ligand binding. However, this appears to actions with the cytoplasmic tail (115). Cell
reect a lack of sensitivity of assays to inter- activation by PMA or chemokine increases
mediate levels of afnity, rather than a lack of LFA-1 diffusion on the membrane (106,
afnity regulation. Recent improvements in 116). Moreover, articially increasing LFA-1

diffusiveness by actin cytoskeleton disruption measured by cell spreading, actin stress ber,
enhances both mobility and adhesion (116). focal adhesion formation, and focal adhe-
However, redistribution or clustering of inte- sion kinase activation ( J. Zhu, C. Carman,
grins was not induced directly by treatments M. Kim, M. Shimaoka, T.A. Springer & B.
that increased membrane mobility alone, and Luo, unpublished observations). These de-
redistribution was, instead, dependent on the fects in outside-in signaling were rescued
presence of multivalent ligand substrates, sug- by reduction of the intersubunit disulde
gesting a ligand- and mass-action-driven re- bond. Thus, separation of transmembrane do-
distribution model that functions in adhe- mains is an important component of integrin
sion strengthening rather than in priming (75, outside-in signal transduction. The role of
117). Complexity is added to this model by clustering might then be to facilitate inter-
ndings that LFA-1 (118, 119) and other in- actions among different integrin-bound and
tegrins (120) become conned upon ligand focal adhesionassociated kinases to promote
binding or stabilization of the open integrin transphosphorylation/activation events in a

conformation and that diffusion rates may de- fashion loosely analogous to the well-
pend on afnity states (121). characterized paradigm of receptor tyrosine
kinase subunit-subunit transactivation.
Integrin Outside-In Signaling

Binding of extracellular ligands by inte- CONCLUDING REMARKS
grins results in signal transduction across the Recent structural, biochemical, and biophys-
plasma membrane that regulates cell shape, ical studies have greatly advanced our un-
migration, growth, and survival, a process derstanding of the mechanisms of inte-
termed outside-in signaling. Details of the grin bidirectional signaling across the plasma
many signaling pathways emanating from in- membrane. Indeed, perhaps more is known
tegrins are beyond the scope of this review, about how integrins transmit signals across
and readers are referred to several recent the membrane than for any other receptors
and extensive reviews (122125). Investiga- with two transmembrane domains, includ-
tors widely believe that lateral association (i.e., ing receptor kinases. Accumulating evidence
clustering) of integrin heterodimers, which demonstrates that conformational afnity
occurs as a consequence of multivalent lig- regulation plays a dominant role in integrin
and binding (75, 117), plays a major role in priming (inside-out signaling), whereas lateral
outside-in signaling (see review in Reference redistribution (clustering) functions in adhe-
122). However, ligand binding can also di- sion strengthening, and both integrin confor-
rectly lead to and stabilize separation of inte- mational change and clustering are required
grin cytoplasmic domains (35). To character- for outside-in signaling. The many different
ize the role of integrin conformational change conformational states of integrins are in dy-
(e.g., separation of the transmembrane and namic equilibrium. Intracellular signals or lig-
cytoplasmic domain interfaces) in outside- and binding act by shifting the equilibrium,
in signaling, a mutant with an intersubunit not by locking integrins in one specic state.
disulde bond between the and trans- Furthermore, inside-out signals activate only
membrane domains (97) was studied. The a subset of integrin molecules on the cell
IIb3 mutant retains Mn2+ -stimulated lig- surface, and these may have a localized cell
and binding as described above and medi- surface distribution. Much more remains to
ates adhesion to brinogen substrates. How- be learned about integrin structure, dynam-
ever, this mutant exhibits a profound defect ics, and linkage to the cytoskeleton and both
in adhesion-induced outside-in signaling as downstream and upstream effectors.

ACKNOWLEDGMENTS
The authors are supported by the American Heart Association (0535403T to B.H.L.), the
Arthritis Foundation (C.V.C.), and NIH grants (HL48675, CA31798, and CA31799 to T.A.S.).
We thank Can Xie and Nori Nishida for help with gures.
LITERATURE CITED
1. Pribila JT, Quale AC, Mueller KL, Shimizu Y. 2004. Integrins and T cell-mediated
immunity. Annu. Rev. Immunol. 22:15780
2. Kinashi T. 2005. Intracellular signaling controlling integrin activation in lymphocytes.
Nat. Rev. Immunol. 5:54659
3. Anderson DC, Springer TA. 1987. Leukocyte adhesion deciency: an inherited defect in
the Mac-1, LFA-1, and p150,95 glycoproteins. Annu. Rev. Med. 38:17594
4. Lee JO, Rieu P, Arnaout MA, Liddington R. 1995. Crystal structure of the A domain
from the subunit of integrin CR3 (CD11b/CD18). Cell 80:63138
5. Emsley J, Knight CG, Farndale RW, Barnes MJ, Liddington RC. 2000. Structural basis
of collagen recognition by integrin 21. Cell 101:4756
6. Shimaoka M, Xiao T, Liu JH, Yang Y, Dong Y, et al. 2003. Structures of the L I domain
and its complex with ICAM-1 reveal a shape-shifting pathway for integrin regulation.
Cell 112:99111
7. Song G, Yang Y, Liu JH, Casasnovas JM, Shimaoka M, et al. 2005. An atomic resolution
view of ICAM recognition in a complex between the binding domains of ICAM-3 and
integrin L2. Proc. Natl. Acad. Sci. USA 102:336671
8. Lee JO, Bankston LA, Arnaout MA, Liddington RC. 1995. Two conformations of the
integrin A-domain (I-domain): a pathway for activation? Structure 3:133340
9. Shimaoka M, Shifman JM, Jing H, Takagi J, Mayo SL, Springer TA. 2000. Computational
design of an integrin I domain stabilized in the open, high afnity conformation. Nat.
Struct. Biol. 7:67478
10. Jin M, Song G, Kim YS, Astrof N, Shimaoka M, et al. 2006. Directed evolution to probe
protein allostery and integrin I domains of 200,000-fold higher afnity. Proc. Natl. Acad.
Sci. USA 103:575863
11. Shimaoka M, Lu C, Palframan RT, von Andrian UH, McCormack A, et al. 2001. Re-
versibly locking a protein fold in an active conformation with a disulde bond: integrin
L I domains with high afnity and antagonist activity in vivo. Proc. Natl. Acad. Sci. USA
98:600914
12. Jin M, Andricioaei I, Springer TA. 2004. Conversion between three conformational states
of integrin I domains with a C-terminal pull spring studied with molecular dynamics.
Structure 12:213747
13. Huang L, Shimaoka M, Rondon I, Roy I, Chang Q, et al. 2006. Identication and
characterization of a human monoclonal antagonistic antibody AL-57 that preferentially
binds the high-afnity form of lymphocyte function-associated antigen-1. J. Leukoc. Biol.
80:90514
14. Shimaoka M, Kim M, Cohen EH, Yang W, Astrof N, et al. 2006. AL-57, a ligand-
mimetic antibody to integrin LFA-1, reveals chemokine-induced afnity up-regulation
in lymphocytes. Proc. Natl. Acad. Sci. USA 103:1399196
15. Diamond MS, Garcia-Aguilar J, Bickford JK, Corbi AL, Springer TA. 1993. The I domain
is a major recognition site on the leukocyte integrin Mac-1 (CD11b/CD18) for four
distinct adhesion ligands. J. Cell Biol. 120:103143

16. Kallen J, Welzenbach K, Ramage P, Geyl D, Kriwacki R, et al. 1999. Structural basis for
LFA-1 inhibition upon lovastatin binding to the CD11a I-domain. J. Mol. Biol. 292:19
17. Last-Barney K, Davidson W, Cardozo M, Frye LL, Grygon CA, et al. 2001. Binding site
elucidation of hydantoin-based antagonists of LFA-1 using multidisciplinary technolo-
gies: evidence for the allosteric inhibition of a protein-protein interaction. J. Am. Chem.
Soc. 123:564350
18. Liu G, Huth JR, Olejniczak ET, Mendoza R, DeVries P, et al. 2001. Novel p-arylthio cin-
namides as antagonists of leukocyte function-associated antigen-1/intracellular adhesion
molecule-1 interaction. 2. Mechanism of inhibition and structure-based improvement of
pharmaceutical properties. J. Med. Chem. 44:120210
19. Lu C, Shimaoka M, Zang Q, Takagi J, Springer TA. 2001. Locking in alternate conforma-
tions of the integrin L2 I domain with disulde bonds reveals functional relationships
among integrin domains. Proc. Natl. Acad. Sci. USA 98:239398
20. Lu C, Shimaoka M, Ferzly M, Oxvig C, Takagi J, Springer TA. 2001. An isolated, surface-
expressed I domain of the integrin L2 is sufcient for strong adhesive function when
locked in the open conformation with a disulde. Proc. Natl. Acad. Sci. USA 98:238792
21. Gahmberg CG, Tolvanen M, Kotovuori P. 1997. Leukocyte adhesion: structure and
function of human leukocyte 2 -integrins and their cellular ligands. Eur. J. Biochem.
245:21532
22. Song G, Lazar GA, Kortemme T, Shimaoka M, Desjarlais JR, et al. 2006. Rational design
of ICAM-1 variants for antagonizing integrin LFA-1-dependent adhesion. J. Biol. Chem.
281:504249
23. Casasnovas JM, Springer TA, Liu J, Harrison SC, Wang J. 1997. The crystal structure
of ICAM-2 reveals a distinctive integrin recognition surface. Nature 387:31215
24. Yang Y, Jun CD, Liu J, Zhang RG, Jochimiak A, et al. 2004. Structural basis for dimer-
ization of ICAM-1 on the cell surface. Mol. Cell 14:26976
25. Reilly PL, Woska JR Jr, Jeanfavre DD, McNally E, Rothlein R, Bormann BJ. 1995. The
native structure of intercellular adhesion molecule-1 (ICAM-1) is a dimer: correlation
with binding to LFA-1. J. Immunol. 155:52932
26. Miller J, Knorr R, Ferrone M, Houdei R, Carron CP, Dustin ML. 1995. Intercellular
adhesion molecule-1 dimerization and its consequences for adhesion mediated by lym-
phocyte function associated-1. J. Exp. Med. 182:123141
27. Altieri DC, Morrissey JH, Edgington TS. 1988. Adhesive receptor Mac-1 coordinates
the activation of factor X on stimulated cells of monocytic and myeloid differentiation:
an alternative initiation of the coagulation protease cascade. Proc. Natl. Acad. Sci. USA
85:746266
28. Diamond MS, Staunton DE, de Fougerolles AR, Stacker SA, Garcia-Aguilar J, et al. 1990.
ICAM-1 (CD54): a counter-receptor for Mac-1 (CD11b/CD18). J. Cell Biol. 111:312939
29. Vorup-Jensen T, Carman CV, Shimaoka M, Schuck P, Svitel J, Springer TA. 2005.
Exposure of acidic residues as a danger signal for recognition of brinogen and other
macromolecules by integrin X2. Proc. Natl. Acad. Sci. USA 102:161419
30. Nermut MV, Green NM, Eason P, Yamada SS, Yamada KM. 1988. Electron microscopy
and structural model of human bronectin receptor. EMBO J. 7:409399
31. Xiong J-P, Stehle T, Diefenbach B, Zhang R, Dunker R, et al. 2001. Crystal structure of
the extracellular segment of integrin V3. Science 294:33945
32. Xiong J-P, Stehle T, Zhang R, Joachimiak A, Frech M, et al. 2002. Crystal structure
of the extracellular segment of integrin V3 in complex with an Arg-Gly-Asp ligand.
Science 296:15155

33. Takagi J, Petre BM, Walz T, Springer TA. 2002. Global conformational rearrangements
in integrin extracellular domains in outside-in and inside-out signaling. Cell 110:599
611
34. Takagi J, Strokovich K, Springer TA, Walz T. 2003. Structure of integrin 51 in com-
plex with bronectin. EMBO J. 22:460715
35. Kim M, Carman CV, Springer TA. 2003. Bidirectional transmembrane signaling by
cytoplasmic domain separation in integrins. Science 301:172025
36. Xiao T, Takagi J, Wang J, Coller BS, Springer TA. 2004. Structural basis for allostery in
integrins and binding of ligand-mimetic therapeutics to the platelet receptor for brino-
gen. Nature 432:5967
37. Springer TA. 1997. Folding of the N-terminal, ligand-binding region of integrin -
subunits into a -propeller domain. Proc. Natl. Acad. Sci. USA 94:6572
38. Bilsland CA, Diamond MS, Springer TA. 1994. The leukocyte integrin p150,95
(CD11c/CD18) as a receptor for iC3b: activation by a heterologous subunit and local-

ization of a ligand recognition site to the I domain. J. Immunol. 152:458289
39. Huang C, Zang Q, Takagi J, Springer TA. 2000. Structural and functional studies with
antibodies to the integrin 2 subunit: a model for the I-like domain. J. Biol. Chem.
275:2151424
40. Xiong J-P, Stehle T, Goodman SL, Arnaout MA. 2004. A novel adaptation of the integrin
PSI domain revealed from its crystal structure. J. Biol. Chem. 279:4025254
41. Shi M, Sundramurthy K, Liu B, Tan SM, Law SK, Lescar J. 2005. The crystal structure
of the plexin-semaphorin-integrin domain/hybrid domain/I-EGF1 segment from the
human integrin 2 subunit at 1.8-A resolution. J. Biol. Chem. 280:3058693
42. Takagi J, Springer TA. 2002. Integrin activation and structural rearrangement. Immunol.
Rev. 186:14163
43. Nishida N, Xie C, Shimaoka M, Cheng Y, Walz T, Springer TA. 2006. Activation of
leukocyte 2 integrins by conversion from bent to extended conformations. Immunity
25:58394
44. Honda S, Tomiyama Y, Pelletier AJ, Annis D, Honda Y, et al. 1995. Topography of
ligand-induced binding sites, including a novel cation-sensitive epitope (AP5) at the amino
terminus, of the human integrin 3 subunit. J. Biol. Chem. 270:1194754
45. Lu C, Ferzly M, Takagi J, Springer TA. 2001. Epitope mapping of antibodies to the
C-terminal region of the integrin 2 subunit reveals regions that become exposed upon
receptor activation. J. Immunol. 166:562937
46. Mould AP, Travis MA, Barton SJ, Hamilton JA, Askari JA, et al. 2005. Evidence that mon-
oclonal antibodies directed against the integrin subunit plexin/semaphorin/integrin
domain stimulate function by inducing receptor extension. J. Biol. Chem. 280:423846
47. Peterson JA, Nyree CE, Newman PJ, Aster RH. 2003. A site involving the hybrid and
PSI homology domains of GPIIIa (3-integrin subunit) is a common target for antibodies
associated with quinine-induced immune thrombocytopenia. Blood 101:93742
48. Xie C, Shimaoka M, Xiao T, Schwab P, Klickstein LB, Springer TA. 2004. The integrin
subunit leg extends at a Ca2+ -dependent epitope in the thigh/genu interface upon
activation. Proc. Natl. Acad. Sci. USA 101:1542227
49. Chen JF, Salas A, Springer TA. 2003. Bistable regulation of integrin adhesiveness by a
bipolar metal ion cluster. Nat. Struct. Biol. 10:9951001
50. Mould AP, Barton SJ, Askari JA, Craig SE, Humphries MJ. 2003. Role of ADMIDAS
cation-binding site in ligand recognition by integrin 51. J. Biol. Chem. 278:51622
29

51. Chen JF, Yang W, Kim M, Carman CV, Springer TA. 2006. Regulation of outside-in
signaling and afnity by the 2 I domain of integrin L2. Proc. Natl. Acad. Sci. USA
103:1399196
52. Huth JR, Olejniczak ET, Mendoza R, Liang H, Harris EA, et al. 2000. NMR and mutage-
nesis evidence for an I domain allosteric site that regulates lymphocyte function-associated
antigen 1 ligand binding. Proc. Natl. Acad. Sci. USA 97:523136
53. Alonso JL, Essa M, Xiong J-P, Stehle T, Arnaout MA. 2002. Does the integrin A
domain act as a ligand for its A domain? Curr. Biol. 12:R34042
54. Yang W, Shimaoka M, Salas A, Takagi J, Springer TA. 2004. Inter-subunit signal trans-
mission in integrins by a receptor-like interaction with a pull spring. Proc. Natl. Acad. Sci.
USA 101:290611
55. Welzenbach K, Hommel U, Weitz-Schmidt G. 2002. Small molecule inhibitors in-
duce conformational changes in the I domain and the I-like domain of lymphocyte
function-associated antigen-1: molecular insights into integrin inhibition. J. Biol. Chem.

277:1059098
56. Shimaoka M, Salas A, Yang W, Weitz-Schmidt G, Springer TA. 2003. Small molecule
integrin antagonists that bind to the 2 subunit I-like domain and activate signals in one
direction and block them in another. Immunity 19:391402
57. Salas A, Shimaoka M, Kogan AN, Harwood C, von Andrian UH, Springer TA. 2004.
Rolling adhesion through an extended conformation of integrin L2 and relation to
I and I-like domain interaction. Immunity 20:393406
58. Knorr R, Dustin ML. 1997. The lymphocyte function-associated antigen 1 I domain is a
transient binding module for intercellular adhesion molecule (ICAM)-1 and ICAM-1 in
hydrodynamic ow. J. Exp. Med. 186:71930
59. Salas A, Shimaoka M, Chen SQ, Carman CV, Springer TA. 2002. Transition from rolling
to rm adhesion is regulated by the conformation of the I domain of the integrin LFA-1.
J. Biol. Chem. 277:5025562
60. Iwasaki K, Mitsuoka K, Fujiyoshi Y, Fujisawa Y, Kikuchi M, et al. 2005. Electron tomog-
raphy reveals diverse conformations of integrin IIb3 in the active state. J. Struct. Biol.
150:25967
61. Mould AP, Symonds EJ, Buckley PA, Grossmann JG, McEwan PA, et al. 2003. Structure
of an integrin-ligand complex deduced from solution X-ray scattering and site-directed
mutagenesis. J. Biol. Chem. 278:3999399
62. Luo BH, Springer TA, Takagi J. 2003. Stabilizing the open conformation of the integrin
headpiece with a glycan wedge increases afnity for ligand. Proc. Natl. Acad. Sci. USA
100:24038
63. Luo BH, Strokovich K, Walz T, Springer TA, Takagi J. 2004. Allosteric 1 integrin
antibodies that stabilize the low afnity state by preventing the swing-out of the hybrid
domain. J. Biol. Chem. 279:2746671
64. Tng E, Tan SM, Ranganathan S, Cheng M, Law SK. 2004. The integrin L2 hybrid
domain serves as a link for the propagation of activation signal from its stalk regions to
the I-like domain. J. Biol. Chem. 279:5433439
65. Mould AP, Barton SJ, Askari JA, McEwan PA, Buckley PA, et al. 2003. Conformational
changes in the integrin A domain provide a mechanism for signal transduction via hybrid
domain movement. J. Biol. Chem. 278:1702835
66. Tang RH, Tng E, Law SK, Tan SM. 2005. Epitope mapping of monoclonal antibody to
integrin L2 hybrid domain suggests different requirements of afnity states for inter-
cellular adhesion molecules (ICAM)-1 and ICAM-3 binding. J. Biol. Chem. 280:2920816

67. Arnaout MA, Mahalingam B, Xiong J-P. 2005. Integrin structure, allostery, and bidirec-
tional signaling. Annu. Rev. Cell Dev. Biol. 21:381410
68. Luo BH, Springer TA. 2006. Integrin structures and conformational signaling. Curr.
Opin. Cell Biol. 18:57986
69. Adair BD, Xiong J-P, Maddock C, Goodman SL, Arnaout MA, Yeager M. 2005. Three-
dimensional EM structure of the ectodomain of integrin V3 in a complex with -
bronectin. J. Cell Biol. 168:110918
70. Robinson MK, Andrew D, Rosen H, Brown D, Ortlepp S, et al. 1992. Antibody against the
Leu-CAM -chain (CD18) promotes both LFA-1-and CR3-dependent adhesion events.
J. Immunol. 148:108085
71. Andrew D, Shock A, Ball E, Ortlepp S, Bell J, Robinson M. 1993. KIM185, a monoclonal
antibody to CD18 which induces a change in the conformation of CD18 and promotes
both LFA-1- and CR3-dependent adhesion. Eur. J. Immunol. 23:221722

72. Evangelista V, Manarini S, Rotondo S, Martelli N, Polischuk R, et al. 1996.
Platelet/polymorphonuclear leukocyte interaction in dynamic conditions: evidence of
adhesion cascade and cross talk between P-selectin and the 2 integrin CD11b/CD18.
Blood 88:418394
73. Takami M, Herrera R, Petruzzelli L. 2001. Mac-1-dependent tyrosine phosphorylation
during neutrophil adhesion. Am. J. Physiol. Cell Physiol. 280:C104556
74. Petruzzelli L, Maduzia L, Springer TA. 1995. Activation of LFA-1 (CD11a/CD18) and
Mac-1 (CD11b/CD18) mimicked by an antibody directed against CD18. J. Immunol.
155:85466
75. Kim M, Carman CV, Yang W, Salas A, Springer TA. 2004. The primacy of afnity over
clustering in regulation of adhesiveness of the integrin L2. J. Cell Biol. 167:124153
76. Shimaoka M, Lu C, Salas A, Xiao T, Takagi J, Springer TA. 2002. Stabilizing the integrin
M inserted domain in alternative conformations with a range of engineered disulde
bonds. Proc. Natl. Acad. Sci. USA 99:1673741
77. Takagi J, Beglova N, Yalamanchili P, Blacklow SC, Springer TA. 2001. Denition of EGF-
like, closely interacting modules that bear activation epitopes in integrin subunits. Proc.
Natl. Acad. Sci. USA 98:1117580
78. Yang W, Shimaoka M, Chen JF, Springer TA. 2004. Activation of integrin subunit I-like
domains by one-turn C-terminal -helix deletions. Proc. Natl. Acad. Sci. USA 101:233338
79. Takagi J, Erickson HP, Springer TA. 2001. C-terminal opening mimics inside-out
activation of integrin 51. Nat. Struct. Biol. 8:41216
80. OToole TE, Mandelman D, Forsyth J, Shattil SJ, Plow EF, Ginsberg MH. 1991. Mod-
ulation of the afnity of integrin IIb 3 (GPIIb-IIIa) by the cytoplasmic domain of IIb .
Science 254:84547
81. OToole TE, Katagiri Y, Faull RJ, Peter K, Tamura R, et al. 1994. Integrin cytoplasmic
domains mediate inside-out signal transduction. J. Cell Biol. 124:104759
82. Hughes PE, Diaz-Gonzalez F, Leong L, Wu C, McDonald JA, et al. 1996. Breaking the
integrin hinge. J. Biol. Chem. 271:657174
83. Lu C, Springer TA. 1997. The subunit cytoplasmic domain regulates the assembly and
adhesiveness of integrin lymphocyte function-associated antigen-1 (LFA-1). J. Immunol.
159:26878
84. Lu C, Takagi J, Springer TA. 2001. Association of the membrane-proximal regions of the
and subunit cytoplasmic domains constrains an integrin in the inactive state. J. Biol.
Chem. 276:1464248

85. Vinogradova O, Velyvis A, Velyviene A, Hu B, Haas TA, et al. 2002. A structural mech-
anism of integrin IIb 3 inside-out activation as regulated by its cytoplasmic face. Cell
110:58797
86. Weljie AM, Hwang PM, Vogel HJ. 2002. Solution structures of the cytoplasmic tail
complex from platelet IIb- and 3-subunits. Proc. Natl. Acad. Sci. USA 99:587883
87. Ulmer TS, Yaspan B, Ginsberg MH, Campbell ID. 2001. NMR analysis of structure
and dynamics of the cytosolic tails of integrin IIb3 in aqueous solution. Biochemistry
40:7498508
88. Vinogradova O, Vaynberg J, Kong X, Haas TA, Plow EF, Qin J. 2004. Membrane-
mediated structural transitions at the cytoplasmic face during integrin activation. Proc.
Natl. Acad. Sci. USA 101:409499
89. Katagiri K, Maeda A, Shimonaka M, Kinashi T. 2003. RAPL, a novel Rap1-binding
molecule, mediates Rap1-induced adhesion through spatial regulation of LFA-1. Nat.
Immunol. 4:74148
90. Calderwood DA, Zent R, Grant R, Rees DJ, Hynes RO, Ginsberg MH. 1999. The
talin head domain binds to integrin subunit cytoplasmic tails and regulates integrin
activation. J. Biol. Chem. 274:2807174

91. Garcia-Alvarez B, de Pereda JM, Calderwood DA, Ulmer TS, Critchley D, et al. 2003.
Structural determinants of integrin recognition by talin. Mol. Cell 11:4958
92. Tadokoro S, Shattil SJ, Eto K, Tai V, Liddington RC, et al. 2003. Talin binding to integrin
tails: a nal common step in integrin activation. Science 302:1036
93. Calderwood DA, Huttenlocher A, Kiosses WB, Rose DM, Woodside DG, et al. 2001.
Increased lamin binding to -integrin cytoplasmic domains inhibits cell migration. Nat.
Cell Biol. 3:106068
94. Chang DD, Hoang BQ, Liu J, Springer TA. 2002. Molecular basis for interaction between
Icap1 PTB domain and 1 integrin. J. Biol. Chem. 277:814045
95. Ulmer TS, Calderwood DA, Ginsberg MH, Campbell ID. 2003. Domain-specic in-
teractions of talin with the membrane-proximal region of the integrin 3 subunit.
Biochemistry 42:830712
96. Kiema T, Lad Y, Jiang P, Oxley CL, Baldassarre M, et al. 2006. The molecular basis of
lamin binding to integrins and competition with talin. Mol. Cell 21:33747
97. Luo BH, Springer TA, Takagi J. 2004. A specic interface between integrin transmem-
brane helices and afnity for ligand. PLoS Biol. 2:77686
98. Luo BH, Carman CV, Takagi J, Springer TA. 2005. Disrupting integrin transmembrane
domain heterodimerization increases ligand binding afnity, not valency or clustering.
Proc. Natl. Acad. Sci. USA 102:367984
99. Li W, Metcalf DG, Gorelik R, Li R, Mitra N, et al. 2005. A push-pull mechanism for
regulating integrin function. Proc. Natl. Acad. Sci. USA 102:142429
100. Partridge AW, Liu S, Kim S, Bowie JU, Ginsberg MH. 2005. Transmembrane domain
packing stabilizes integrin IIb3 in the low afnity state. J. Biol. Chem. 280:7294300
101. Bazzoni G, Ma L, Blue ML, Hemler ME. 1998. Divalent cations and ligands induce
conformational changes that are highly divergent among 1 integrins. J. Biol. Chem.
273:667078
102. Chen CS, Tan J, Tien J. 2004. Mechanotransduction at cell-matrix and cell-cell contacts.
Annu. Rev. Biomed. Eng. 6:275302
103. Astrof NS, Salas A, Shimaoka M, Chen JF, Springer TA. 2006. The importance of force
linkage in mechanochemistry of adhesion receptors. Biochemistry 45:1502028
104. Carman CV, Springer TA. 2003. Integrin avidity regulation: Are changes in afnity and
conformation underemphasized? Curr. Opin. Cell Biol. 15:54756

105. Bazzoni G, Hemler ME. 1998. Are changes in integrin afnity and conformation overem-
phasized? Trends Biochem. Sci. 23:3034
106. Constantin G, Majeed M, Giagulli C, Piccib L, Kim JY, et al. 2000. Chemokines trigger
immediate 2 integrin afnity and mobility changes: differential regulation and roles in
lymphocyte arrest under ow. Immunity 13:75969
107. Chan JR, Hyduk SJ, Cybulsky MI. 2001. Chemoattractants induce rapid and transient
upregulation of monocyte 4 integrin afnity for vascular adhesion molecule 1 which
mediates arrest: an early step in the process of emmigration. J. Exp. Med. 193:114958
108. Chan JR, Hyduk SJ, Cybulsky MI. 2003. Detecting rapid and transient upregulation
of leukocyte integrin afnity induced by chemokines and chemoattractants. J. Immunol.
Methods 273:4352
109. Lollo BA, Chan KW, Hanson EM, Moy VT, Brian AA. 1993. Direct evidence for two
afnity states for lymphocyte function-associated antigen 1 on activated T cells. J. Biol.
Chem. 268:21693700
110. van Kooyk Y, Figdor CG. 2000. Avidity regulation of integrins: the driving force in
leukocyte adhesion. Curr. Opin. Cell Biol. 12:54247
111. Lawson MA, Maxeld FR. 1995. Ca2+ - and calcineurin-dependent recycling of an inte-
grin to the front of migrating neutrophils. Nature 377:7579
112. Tohyama Y, Katagiri K, Pardi R, Lu C, Springer TA, Kinashi T. 2003. The critical
cytoplasmic regions of the L/2 integrin in Rap1-induced adhesion and migration.
Mol. Biol. Cell 14:257082
113. Shimonaka M, Katagiri K, Kakayama T, Fujita N, Tsuruo T, et al. 2003. Rap1 translates
chemokine signals to integrin activation, cell polarization, and motility across vascular
endothelium under ow. J. Cell Biol. 161:41727
114. Li R, Mitra N, Gratkowski H, Vilaire G, Litvinov SV, et al. 2003. Activation of integrin
IIb3 by modulation of transmembrane helix associations. Science 300:79598
115. Jin T, Li J. 2002. Dynamitin controls 2 integrin avidity by modulating cytoskeletal
constraint on integrin molecules. J. Biol. Chem. 277:3296369
116. Kucik DF, Dustin ML, Miller JM, Brown EJ. 1996. Adhesion-activating phorbol ester
increases the mobility of leukocyte integrin LFA-1 in cultured lymphocytes. J. Clin. Invest.
97:213944
117. Buensuceso C, De Virgilio M, Shattil SJ. 2003. Detection of integrin IIb3 clustering
in living cells. J. Biol. Chem. 278:1521724
118. Peters IM, van Kooyk Y, van Vliet SJ, de Grooth BG, Figdor CG, Greve J. 1999. 3D
single-particle tracking and optical trap measurements on adhesion proteins. Cytometry
36:18994
119. Smith A, Carrasco YR, Stanley P, Kieffer N, Batista FD, Hogg N. 2005. A talin-dependent
LFA-1 focal zone is formed by rapidly migrating T lymphocytes. J. Cell Biol. 170:14151
120. Felsenfeld DP, Choquet D, Sheetz MP. 1996. Ligand binding regulates the directed
movement of 1 integrins on broblasts. Nature 383:43840
121. Cairo CW, Mirchev R, Golan DE. 2006. Cytoskeletal regulation couples LFA-1 confor-
mational changes to receptor lateral mobility and clustering. Immunity 25:297308
122. Shattil SJ, Newman PJ. 2004. Integrins: dynamic scaffolds for adhesion and signaling in
platelets. Blood 104:160615
123. Grashoff C, Thievessen I, Lorenz K, Ussar S, Fassler R. 2004. Integrin-linked kinase:
integrins mysterious partner. Curr. Opin. Cell Biol. 16:56571
124. Ridley AJ, Schwartz MA, Burridge K, Firtel RA, Ginsberg MH, et al. 2003. Cell migration:
integrating signals from front to back. Science 302:17049

125. Guo W, Giancotti FG. 2004. Integrin signaling during tumor progression. Nat. Rev. Mol.
Cell Biol. 5:81626
126. Van der Vieren M, Crowe DT, Hoekstra D, Vazeux R, Hoffman PA, et al. 1999. The
leukocyte integrin D2 binds VCAM-1: evidence for a binding interface between I
domain and VCAM-1. J. Immunol. 163:198490
127. Schlossman SF, Boumsell L, Gilks W, Harlan JM, Kishimoto TK, et al. 1995. White Cell
Differentiation Antigens. New York: Oxford Univ. Press
128. Kishimoto TK, Kikutani H, von dem Borne AEGK, Goyert SM, Mason DY, et al. 1997.
White Cell Differentiation Antigens. New York: Garland

AR306-FM ARI 13 February 2007 11:22
Annual Review of
Immunology
Contents Volume 25, 2007
Frontispiece
Peter C. Doherty p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p x
Challenged by Complexity: My Twentieth Century in Immunology
Peter C. Doherty p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p1
The Impact of Glycosylation on the Biological Function and Structure
of Human Immunoglobulins
James N. Arnold, Mark R. Wormald, Robert B. Sim, Pauline M. Rudd,
and Raymond A. Dwek p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 21
The Multiple Roles of Osteoclasts in Host Defense: Bone Remodeling
and Hematopoietic Stem Cell Mobilization
Orit Kollet, Ayelet Dar, and Tsvee Lapidot p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 51
Flying Under the Radar: The Immunobiology of Hepatitis C
Lynn B. Dustin and Charles M. Rice p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p 71
Resolution Phase of Inammation: Novel Endogenous
Anti-Inammatory and Proresolving Lipid Mediators and Pathways
Charles N. Serhan p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p101
Immunobiology of Allogeneic Hematopoietic Stem Cell
Transplantation
Lisbeth A. Welniak, Bruce R. Blazar, and William J. Murphy p p p p p p p p p p p p p p p p p p p p p p p139
Effector and Memory CTL Differentiation
Matthew A. Williams and Michael J. Bevan p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p171
TSLP: An Epithelial Cell Cytokine that Regulates T Cell
Differentiation by Conditioning Dendritic Cell Maturation
Yong-Jun Liu, Vasilli Soumelis, Norihiko Watanabe, Tomoki Ito,
Yui-Hsi Wang, Rene de Waal Malefyt, Miyuki Omori, Baohua Zhou,
and Steven F. Ziegler p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p193
Discovery and Biology of IL-23 and IL-27: Related but Functionally
Distinct Regulators of Inammation
Robert A. Kastelein, Christopher A. Hunter, and Daniel J. Cua p p p p p p p p p p p p p p p p p p p p p p221
v
Improving T Cell Therapy for Cancer

Ann M. Leen, Cliona M. Rooney, and Aaron E. Foster p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p243
Immunosuppressive Strategies that are Mediated by Tumor Cells
Gabriel A. Rabinovich, Dmitry Gabrilovich, and Eduardo M. Sotomayor p p p p p p p p p p p p267
The Biology of NKT Cells
Albert Bendelac, Paul B. Savage, and Luc Teyton p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p297
Regulation of Cellular and Humoral Immune Responses by the SLAM
and SAP Families of Molecules
Cindy S. Ma, Kim E. Nichols, and Stuart G. Tangye p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p337
Mucosal Dendritic Cells

Akiko Iwasaki p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p381
Immunologically Active Autoantigens: The Role of Toll-Like

Receptors in the Development of Chronic Inammatory Disease
Ann Marshak-Rothstein and Ian R. Rifkin p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p419
The Immunobiology of SARS
Jun Chen and Kanta Subbarao p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p443
Nonreceptor Protein-Tyrosine Phosphatases in Immune Cell Signaling
Lily I. Pao, Karen Badour, Katherine A. Siminovitch, and Benjamin G. Neel p p p p p p p473
Fc Receptor-Like Molecules
Randall S. Davis p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p525
The Death Domain Superfamily in Intracellular Signaling of Apoptosis
and Inammation
Hyun Ho Park, Yu-Chih Lo, Su-Chang Lin, Liwei Wang, Jin Kuk Yang,
and Hao Wu p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p561
Cellular Responses to Viral Infection in Humans: Lessons from
Epstein-Barr Virus
Andrew D. Hislop, Graham S. Taylor, Delphine Sauce, and Alan B. Rickinson p p p p p p587
Structural Basis of Integrin Regulation and Signaling
Bing-Hao Luo, Christopher V. Carman, and Timothy A. Springer p p p p p p p p p p p p p p p p p p p619
Zoned Out: Functional Mapping of Stromal Signaling
Microenvironments in the Thymus
Howard T. Petrie and Juan Carlos Zuniga-Pucker p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p649
T Cells as a Self-Referential, Sensory Organ
Mark M. Davis, Michelle Krogsgaard, Morgan Huse, Johannes Huppa,
Bjoern F. Lillemeier, and Qi-jing Li p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p681
The Host Defense of Drosophila melanogaster
Bruno Lemaitre and Jules Hoffmann p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p697
vi Contents
Ontogeny of the Hematopoietic System

Ana Cumano and Isabelle Godin p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p745
Chemokine:Receptor Structure, Interactions, and Antagonism
Samantha J. Allen, Susan E. Crown, and Tracy M. Handel p p p p p p p p p p p p p p p p p p p p p p p p p p787
IL-17 Family Cytokines and the Expanding Diversity of Effector
T Cell Lineages
Casey T. Weaver, Robin D. Hatton, Paul R. Mangan, and Laurie E. Harrington p p p821
Indexes
Cumulative Index of Contributing Authors, Volumes 1525 p p p p p p p p p p p p p p p p p p p p p p p p853

Cumulative Index of Chapter Titles, Volumes 1525 p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p p860
Errata
An online log of corrections to Annual Review of Immunology chapters (if any, 1997 to
the present) may be found at http://immunol.annualreviews.org/errata.shtml
Contents vii

Structural Basis of Integrin Regulation and Signaling

Загружено:

Сведения о документе

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Structural Basis of Integrin Regulation and Signaling

Загружено:

Авторское право:

Доступные форматы

ANRV306-IY25-21 ARI 11 February 2007 13:4

Structural Basis of Integrin

and Timothy A. Springer

Medical School, Boston, Massachusetts 02115; email: luo@cbr.med.harvard.edu,

Annu. Rev. Immunol. 2007. 25:61947 Key Words

INTRODUCTION As adhesion molecules, integrins are

valently associated heterodimeric cell surface

(Figure 1). This diversity in subunit composi-

620 Luo Carman Springer

Table 1 Integrins on leukocytesa

11, VLA-1, CD49a/CD29 Long-term activated T lymphocytes, B Collagen

21, VLA-2, GPIa, CD49b/CD29 long-term activated T lymphocytes, B Collagen

www.annualreviews.org Integrin Regulation and Signaling 621

surrounding the metal (Figure 3). In the

esized to enhance the strength of the metal-

moves from the primary to the secondary co-

622 Luo Carman Springer

www.annualreviews.org Integrin Regulation and Signaling 623

624 Luo Carman Springer

are converted to a higher-afnity state. This This metal-coordination bond is surrounded

www.annualreviews.org Integrin Regulation and Signaling 625

interface in D1 has been revealed in differ- yielding a one-dimensional array of ICAM-

626 Luo Carman Springer

binding of M and especially of X I domains associated with a large-scale global confor-

INTEGRIN GLOBAL TOPOLOGY predicted (37) and later conrmed by crystal

www.annualreviews.org Integrin Regulation and Signaling 627

Closed headpiece, Closed headpiece, Open headpiece,

628 Luo Carman Springer

larger. Residues shown by mutagenesis to be (36, 40, 41).

www.annualreviews.org Integrin Regulation and Signaling 629

630 Luo Carman Springer

www.annualreviews.org Integrin Regulation and Signaling 631

Hybrid Hybrid Invariant Glu residue

subunit subunit subunit subunit

I domain I domain I domain

Hybrid Hybrid Hybrid

subunit subunit subunit subunit subunit subunit

L-E310C 2-A210C L-E310C / 2-A210C

632 Luo Carman Springer

/ I Domain Allosteric Antagonists where in the LFA-1 molecule. This mutant

the I domain MIDAS is mutated (56). Some

www.annualreviews.org Integrin Regulation and Signaling 633

X2 particles are predominantly in the bent

extended. For V3 and X2, about half of

634 Luo Carman Springer

www.annualreviews.org Integrin Regulation and Signaling 635

636 Luo Carman Springer

the transmembrane domains revealed a spe- to enable a substantial proportion of integrin

PLASMA MEMBRANE hesion. The distance in the subunit be-

www.annualreviews.org Integrin Regulation and Signaling 637

638 Luo Carman Springer

binding or stabilization of the open integrin transphosphorylation/activation events in a

kinase subunit-subunit transactivation.

Integrin Outside-In Signaling

www.annualreviews.org Integrin Regulation and Signaling 639

640 Luo Carman Springer

www.annualreviews.org Integrin Regulation and Signaling 641

(CD11c/CD18) as a receptor for iC3b: activation by a heterologous subunit and local-

642 Luo Carman Springer

function-associated antigen-1: molecular insights into integrin inhibition. J. Biol. Chem.

www.annualreviews.org Integrin Regulation and Signaling 643

both LFA-1- and CR3-dependent adhesion. Eur. J. Immunol. 23:221722

644 Luo Carman Springer

activation. J. Biol. Chem. 274:2807174