SlideShare a Scribd company logo
1 of 6
Download to read offline
Cytoskeletal protein radixin activates integrin aMb2 by binding to
its cytoplasmic tail
Pingtao Tang, Chunzhang Cao, Min Xu, Li Zhang*
Center for Vascular and Inflammatory Diseases, Department of Physiology, University of Maryland School of Medicine,
800 W Baltimore Street, Baltimore, MD 21201, USA
Received 1 December 2006; revised 4 February 2007; accepted 6 February 2007
Available online 15 February 2007
Edited by Gianni Cesareni
Abstract Talin binding of integrins, via its band 4.1, ezrin, ra-
dixin, and moesin (FERM)-homologous domain, directly acti-
vates the integrin receptor. However, it is not known whether
other FERM-containing proteins also possess such an integrin
activating capability. We report here that radixin, one of the ori-
ginal FERM-domain proteins, binds to the membrane-proximal
region of the integrin b2 but not aM cytoplasmic tail. Impor-
tantly, we show that radixin binding significantly enhances the
adhesive activity of integrin aMb2. Given the distinct biological
activities of radixin and talin, radixin may represent a novel
talin-independent pathway for integrin activation under specific
settings.
Ó 2007 Federation of European Biochemical Societies. Pub-
lished by Elsevier B.V. All rights reserved.
Keywords: Cell adhesion; Integrin; The band 4.1, ezrin,
radixin, and moesin-homologous domain; Cytoskeleton
1. Introduction
Recent studies suggest that talin binding to the cytoplasmic
domain of integrin receptors represents the final step in inte-
grin activation (‘‘inside-out’’ signaling) [1]. Talin comprises
of an N-terminal 50-kDa head region that is homologous to
the FERM domain, and a C-terminal 205-kDa rod region,
which regulates the function of the N-terminal head via intra-
and inter-molecular interactions. Talin can be activated by cal-
pain-mediated proteolysis [2] or by phosphoinositide binding
[3], and once activated, it binds, through the phosphotyrosine
binding (PTB) domain located within its FERM-homologous
region, to a highly conserved NPxY motif within the integrin
b cytoplasmic tail [4,5]. Integrin engagement by talin subse-
quently disrupts a salt bridge between its a and b cytoplasmic
domains [6], leading to its activation [5]. Interestingly,
although a number of proteins, including syk, Numb, and
Dok, contain a PTB domain that interacts with the integrin
b cytoplasmic tail with high affinity, however, none of them
was capable of activating integrin receptor [5]. Thus, the talin
FERM domain seems very unique in its integrin activating
capability.
In this study, we searched for additional proteins that are
capable of activating integrins. We report here the identifica-
tion of another FERM protein radixin that possesses similar
integrin activating activities. Using aMb2 (Mac-1, CD11b/
CD18) as a representative of integrin receptors, we demon-
strate that radixin binds to the membrane-proximal region of
the integrin b cytoplasmic tail via its N-terminal ezrin–radix-
in–moesin family (ERM) associated domain (N-ERMAD).
We show that radixin binding to integrin aMb2 enhances its
adhesive activity. Together, our study identifies radixin as an
important player in integrin ‘‘inside-out’’ signaling in addition
to talin. As radixin recognizes the integrin cytoplasmic tail dif-
ferently than talin, radixin may represent a novel talin-inde-
pendent pathway for integrin activation under specific settings.
2. Materials and methods
2.1. Materials
MGIN [7] was provided by Dr. R.G. Hawley (American Red Cross,
Rockville, MD). IB4, OKM1, and TS1/18 were from ATCC (Rock-
ville, MD); mAb 44 and 2LPM19c were from Sigma (St. Louis,
MO). mAb 24 [8] was provided by Dr. Nancy Hogg (Imperial Cancer
Research Fund, London, UK). Recombinant c-module of fibrinogen
(Fg) was provided by Dr. L. Medved (University of Maryland,
MD). Neutrophil inhibitor factor (NIF) [9] was provided by Dr. E.
Plow (Cleveland Clinic Foundation, OH). All other reagents were of
the highest grade available from Sigma Chemical Co. (St Louis,
MO) unless otherwise noted.
2.2. Expression of recombinant radixin in E. coli
To express the radixin N-ERMAD (1–298) and C-ERMAD (281–
584) in E. coli, their corresponding cDNAs were amplified using the fol-
lowing pairs of primers: 50
-AAAGGATCC-ATGCCGAAGCCAAT-
CAATGTAAGAGT-30
and 50
-TCAGTCGACTT-AATCAGGCTTT-
CTCCTTCGCATGTACAG-30
, and 50
-AAAGGATCCTTGGCCT-
TATGTATGG-GAAACATGAA-30
and 50
-GAAGTCGACTCA-
CATGGCTTCAAACTCATCGATGCG-30
, and then cloned into
pGEX-4T-1 (for glutathione-s-transferase (GST) fusion; Amersham,
Piscataway, NJ) and pET-32a (for histidine-tag fusion; Novagen, Mad-
ison, WI). Cytohesin-1 was expressed as a histidine-tagged protein using
pQE-31 (QIAgen). Generation of recombinant proteins in E. coli
BL21(DE3) and their purification using the glutathione–Sepharose or
Ni–NTA column have been described previously [9]. The concentrations
of the purified proteins were determined spectrophotometrically using
Abbreviations: C-ERMAD, C-terminal ezrin–radixin–moesin family
(ERM) associated domain; CHO, Chinese hamster cells; ERM, the
ezrin–radixin–moesin protein family; ERMAD, ERM-associated
domain; FERM, the band 4.1, ezrin, radixin, and moesin-homologous
domain; Fg, fibrinogen; GFP, green fluorescent protein; HRP, horse-
radish peroxidase; ICAM-1, intercellular adhesion molecule-1; N-E-
RMAD, N-terminal ezrin–radixin–moesin family (ERM) associated
domain; PBS, phosphate buffered saline; PTB, phosphotyrosine bind-
ing domain; GST, glutathione-s-transferase; SDS–PAGE, sodium
dodecyl sulfate polyacrylamide gel electrophoresis
*
Corresponding author. Fax: +1 410 706 8121.
E-mail address: lizhang@som.umaryland.edu (L. Zhang).
0014-5793/$32.00 Ó 2007 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.febslet.2007.02.013
FEBS Letters 581 (2007) 1103–1108
the equation C (mg/ml) = (A280–A320)/1.45 and the concentrations
were further verified by Coomassie blue-stained SDS–PAGE using
BSA as a standard.
2.3. Binding assays by ELISA
Ninety six-well polystyrene microtiter plates (Costar, Corning, NY)
were coated with different synthetic peptides (1 mg/ml) overnight at
4 °C and post-coated with 10% milk for 1 h at room temperature.
GST-tagged radixin fragments or His-tagged cytohesin-1 (both at
150 nM) in phosphate buffered saline (PBS) plus 0.1%BSA were added
and incubated for 40 min at room temperature. After washing with
PBS, bound GST-radixin was detected using a horseradish peroxidase
(HRP)-anti-GST conjugate, and an HRP substrate 3,30
,5,50
-tetrameth-
ylbenzidine, measuring absorption at 550 nm (A550).
2.4. Pull down assays
Cells expressing aMb2, which were established previously [10], were
lysed in 20 mM Tris–HCl, 150 mM NaCl, pH 7.4, 0.25% deoxycholate,
1% p-nitrophenyl-p-guanidinobezoate, 2 mM EGTA, 1 mM PMSF,
10 mM benzamidine, 25 lg/ml SBTI, and 20 lg/ml leupeptin. After
centrifugation, cell lysates were incubated with GST-radixin N-ER-
MAD that was immobilized onto a glutathione agarose column. After
washing, bound proteins were eluted and then analyzed by 10% SDS–
PAGE. The presence of aMb2 was detected by immunobloting using an
anti-b2 cytoplasmic domain antibody [11] and visualized using an HRP
conjugate of goat anti-rabbit IgG.
2.5. Establishment of Chinese hamster cells (CHO) expressing both
integrin aMb2 and radixin
To express aMb2 and radixin simultaneously, cDNAs encoding N-
ERMAD and C-ERMAD were inserted individually into the expres-
sion vector MGIN [7], in which radixin expression is tightly linked
to green fluorescent protein (GFP) expression via an internal ribosome
entry site (IRES). These expression plasmids were transfected, together
with a expression vector containing the hygromycin cDNA, into the
aMb2-expressing CHO cells that we previously established [10]. After
selection with 200 lg/ml hygromycin, stable clones that express equiv-
alent amount of aMb2 and radixin were established by dual-color cell
sorting using a mAb anti-aM (OKM1) and GFP as cell sorting mark-
ers.
2.6. Cell adhesion assays
Cell adhesion was conducted using Fg or its aMb2-recognition do-
main, the c-module [10,12], based on our published procedures [13].
3. Results
3.1. The N-terminal FERM domain of radixin interacts with the
b2 cytoplasmic tail
To see if talin is the only FERM domain-containing protein
that is capable of activating integrins, we tested the ability of
radixin to interact with the b2 integrin cytoplasmic tail. Given
that radixin exists in an inactive state due to an intramolecular
interaction between its N- and C-terminal ERMADs [14], we
expressed N-ERMAD (residues 1–298) and C-ERMAD (resi-
dues 281–584) separately as either GST- or His-tagged proteins.
All of the recombinant proteins exhibited expected molecular
sizes and were more than 90% pure (data not shown).
Interactions between radixin fragments and integrin cyto-
plasmic tails were assessed by ELISA. Thus, 96-well microtiter
plates were coated with aM (CYKLGFFKRQYKDMM-
SEGGPPGAEPQ) or b2 (CWKALIHLSDLREYRRFEK-
EKLKSQWNNDNPLFKSATTTVMNPKFAES) cytoplas-
mic peptides. Different concentrations of radixin fragments
were added and bound proteins were quantified. As shown
in Fig. 1A, N-ERMAD bound the b2 cytoplasmic peptide
dose-dependently with a Kd of 110 ± 27 nM, whereas no signif-
icant binding activity was observed for C-ERMAD. Moreover,
neither radixin fragments interacted with the aM cytoplasmic
peptide. Furthermore, no binding to an irrelevant peptide
SW63 (CSTHFPGGGSISHSGHF) or a scrambled peptide
sCTD31 (CLYRLEWFHAILRSRKD) was observed for N-
ERMAD (data not shown), verifying the specificity of the
binding assay. These data demonstrated that radixin recog-
nizes the b2 integrin cytoplasmic tail via its N-terminal FERM
domain.
3.2. Inhibition of N-ERMAD binding to integrin cytoplasmic
domain by C-ERMAD
To see if N-ERMAD binding to the integrin b2 cytoplasmic
domain is controlled by C-ERMAD, we conducted the above
binding assays in the presence of His-tagged C-ERMAD.
Addition of C-ERMAD at ratios of 2:1 and 4:1 (C-ER-
MAD:N-ERMAD) both led to significant inhibition on N-
ERMAD binding to the b2 cytoplasmic peptide (Fig. 1B). As
expected, N-ERMAD did not interact with the aM cytoplasmic
tail with or without C-ERMAD.
BoundradixinN-ERMAD(A550)
0.0
0.1
0.2
0.3
0.4
0.5
0.6
C/N Ratio: 0:1 2:1 4:1
the β2 peptide
the αM peptide
Radixin fragment added (nM)
0 50 100 150 200 250
Boundradixinfragments(A550)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
B
A
Fig. 1. Radixin N-ERMAD interacts with the integrin b cytoplasmic
tail. (A) Radixin N-ERMAD recognizes the b2 cytoplasmic domain. A
96-well plate was coated with either the aM (s, n) or b2 cytoplasmic
peptides (d, m), blocked, and then incubated with GST-radixin N-
ERMAD (s, d) or GST-C-ERMAD (n, m). After washing, bound
proteins were determined. (B) Radixin C-ERMAD blocks N-ERMAD
binding to the b2 cytoplasmic tail. Binding of radixin N-ERMAD to
the aM (filled bar) or b2 (open bar) peptides were conducted as
described above, except that different concentrations of His-tagged
radixin C-ERMAD were added. Data are means ± S.D. of duplicate
samples and are representative of three independent experiments.
1104 P. Tang et al. / FEBS Letters 581 (2007) 1103–1108
3.3. Radixin binds to the membrane-proximal region of the
integrin b2 cytoplasmic tail
To localize the functional sequence within the integrin b2
cytoplasmic tail for radixin recognition, we tested N-ERMAD
binding to truncated b2 cytoplasmic peptides (Fig. 2A). We
found that removal of NPxY/F, the sequence recognized by ta-
lin, did not significantly affect N-ERMAD binding as both
truncated b2 peptides CTD19 and CTD31 bound N-ERMAD
well (Fig. 2B). In fact, they exhibited higher binding activity
toward N-ERMAD than the full length peptide. Verifying
specificity, N-ERMAD did not bind to either the scrambled
peptide sCTD31 or BSA-coated wells.
To further confirm the involvement of the membrane-prox-
imal region of the b2 cytoplasmic tail in radixin recognition,
the ability of soluble CTD31 to inhibit N-ERMAD binding
of the full length b2 cytoplasmic peptide was tested. Thus,
N-ERMAD was added to b2 peptide-coated 96-well plates in
the presence of different concentrations of CTD31. After wash-
ing, bound N-ERMAD was determined as above. Fig. 2C
shows that CTD31, which contains only the membrane-proxi-
mal 16 residues, competed effectively with the full length b2
peptide for radixin binding with a C50 of 25 lM, whereas its
scrambled control peptide sCTD31 or DMSO had no effect.
Together, these data demonstrated that the NPxY/F sequence
within the b2 cytoplasmic domain, which is critical to talin rec-
ognition, is not required for radixin binding.
3.4. Radixin and cytohesin-1 recognize overlapping binding sites
within the b2 cytoplasmic tail
Cytohesin-1 activates aLb2 by binding to residues 723–731 of
b2 [15]. To see if cytohesin-1 and radixin share a common rec-
ognition site within b2, we conducted additional binding exper-
iments. Fig. 3A shows that cytohesin-1 binding to the full
length b2 peptide was effectively inhibited by soluble CTD31
in a dose- dependent manner, whereas additions of similar con-
centrations of DMSO or an irrelevant peptide SW63 had no
effect and addition of (sCTD31) yielded much weaker inhibi-
tion.
To further confirm that radixin and cytohesin-1 recognize an
overlapping site within the b2 cytoplasmic tail, we tested if
CWKALIHLSDLREYRRFEKEKLKSQWNNDNPLFKSATTTVMNPKFAESWT:
CWKALIHLSDLREYRRFEKEKLKSQWNNDCTΔ19:
CLYRLEWFHAILRSRKDSCTΔ31:
CWKALIHLSDLREYRRFCTΔ31:
723
|
750
|
738
|
Figure 2
BoundradixinN-ERMAD(A550)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
WT CTΔ19 CTΔ31 sCTΔ31 BSA
Peptide added (μM)
0 100 200 300 400
BoundradixinN-ERMAD(A550)
0.0
0.5
1.0
1.5
2.0
DMSO
CTΔ31
GST
s CTΔ31
A
B
C
Fig. 2. Radixin recognizes the membrane-proximal region of the
integrin b2 cytoplasmic tail. (A) The sequence of the b2 cytoplasmic
peptides. (B) Radixin N-ERMAD binds to the membrane-proximal
region of the b2 cytoplasmic tail. Interaction between the different b2
cytoplasmic peptides and radixin N-ERMAD were determined by
ELISA as above. (C) Inhibition of N-ERMAD binding to the b2
cytoplasmic peptide by CTD31. Radixin N-ERMAD binding to the b2
cytoplasmic tail was performed as above, except in the presence
different soluble peptides. Data are means ± S.D. of duplicate samples
and are representative of two independent experiments. sCTD31,
scrambled CTD31.
Cytohesin-1 (μM)
0 2 4 6 8 10 12
BoundradixinN-ERMAD(A550)
0.0
0.1
0.2
0.3
0.4
0.5
0.6
Peptide added (μM)
0 20 40 60 80 100
Boundcytohesin-1(A550)
0.0
0.5
1.0
1.5
2.0
2.5
DMSO
CTΔ31
GST
sCTΔ31
SW63
B
A
Fig. 3. Cytohesin-1 and radixin N-ERMAD recognize overlapping
regions of the b2 cytoplasmic tail. (A) Cytohesin-1 binding to the b2
cytoplasmic tail. His-cytohesin-1 binding to the b2 cytoplasmic tail in
the presence of different b2 peptides was performed as above. (B)
Competition between Cytohesin-1 and radixin. Binding of GST-
radixin N-ERMAD to the b2 cytoplasmic tail was conducted as in
Fig. 1 except in the presence of His-tagged cytohesin-1 (h) or a control
protein BSA (s). Data are means ± S.D. of duplicate samples and are
representative of two independent experiments.
P. Tang et al. / FEBS Letters 581 (2007) 1103–1108 1105
ActivationIndex
0.00
0.02
0.04
0.06
0.08
0.10
0.12
GFP N-ERMAD
clone 6
*
Celladhesion(A570)
0
1
2
3
4
5
G
FPN
-ER
M
AD
clone
6N
-ER
M
AD
poolC
-ER
M
AD
poolN
-ER
M
AD
pool+ED
TA
N
-ER
M
AD
clone
6+N
IF
m
ock
C
-ER
M
A
D
pool+N
IF
A
B
C
D
Fig. 4. Expression of the N-terminal radixin promotes integrin adhesive activity. (A) Pull-down assays. Cell lysates from aMb2-expressing CHO cells
were passed through either GST-N-ERMAD (lanes 1 and 3) or GST (lanes 2 and 4)-column. After washing, the bound proteins were analyzed by
SDS–PAGE and immunoblotted with anti-b2 cytoplasmic tail (lanes 1 and 2) or anti-GST (lanes 3 and 4). (B) The effect of radixin N-ERMAD on
cell adhesion. Different radixin fragments were expressed in aMb2/CHO cells and their adhesion to the Fg c-module (10 lg/ml) was determined in the
presence or absence of an aMb2-specific inhibitor NIF (10 nM). After washing, representative adherent cells were photographed with 20· differential
interference contrast (DIC) lens using the Zeiss Axiovert 200M system. Scale bars: 20 lm. (C) The numbers of adherent cells in (B) were quantified by
staining with Crystal violet and measuring absorption at 570 nm. The specificity of the adhesion assay was confirmed by lack of cell adhesion of mock
transfected cells. (D) Conformational changes within the aMb2 receptor. aMb2/CHO cells expressing either N-ERMAD or GFP were incubated
with 1 lg of mAb 24 at 37 °C or 2LPM19c in the presence of 1 mM Ca2+
/1 mM Mg2+
for 30 min. Bound mAb 24 or 2LPM19c was determined by
flow cytometry using R-Phycoerythrin conjugated anti-mouse IgG. An isotype matched non-immune IgG was used as a control (not shown). Mean
fluorescence intensity of mAb 24, calculated using a FACScan program (BD Biosciences, Mountain View, CA) and normalized by surface expression
of the receptor (based on 2LPM19c staining), was used as an activation index. N-ERMAD-expressing aMb2/CHO cells exhibited significantly higher
mAb 24 binding compared to those expressing GFP (*, P = 0.002). Data are means ± S.D. of duplicate samples and are representative of three
independent experiments.
1106 P. Tang et al. / FEBS Letters 581 (2007) 1103–1108
radixin and cytohesin-1 could compete directly with each other
for binding to the b2 cytoplasmic peptide. We found that in-
deed cytohesin-1 competed with radixin N-ERMAD effectively
with a C50 of 0.2 lM, whereas addition of BSA did not have
significant effect (Fig. 3B).
3.5. Radixin binds integrin aMb2 in total cell lysates
To test if radixin interacts with integrin receptors in whole
cell lysates, we carried out pull-down experiments between ra-
dixin N-ERMAD and aMb2, a major b2 integrin expressed on
leukocytes. Fig. 4A shows that radixin N-ERMAD but not
GST, both of which were immobilized onto glutathione–Se-
pharose resin, could effectively pull down aMb2, as demon-
strated by the presence of a 95 kDa protein by immunoblot
using anti-b2 polyclonal antibody. Additional immunoblot
analyses of the pull-down products with anti-GST antibody
verified that equally amount of GST and GST-N-ERMAD
were present on the resin.
3.6. Radixin promotes integrin adhesive activity
To test if radixin binding to the b2 cytoplasmic domain
could result in integrin activation, we transfected N-ERMAD
or C-ERMAD of radixin into aMb2-expressing CHO cells,
using the retroviral vector MGIN [7]. Stable cell lines that
express equivalent levels of aMb2 on the cell surface and similar
levels of radixin were established by cell sorting using both the
aM-specific mAb OKM1 and GFP as markers. Surface expres-
sion of aMb2 by the different clones and also pools of the aMb2/
radixin/CHO cells used in this study differed by less than two-
fold (mean fluorescence intensity: 1594 for GFP, 1433 for C-
ERMAD pool, 1035 for N-ERMAD clone 6, and 2555 for
N-ERMAD pool). Cell adhesion experiments demonstrated
that expression of radixin N-ERMAD in aMb2/CHO cells sig-
nificantly increased the adhesive activity of aMb2 toward the
Fg c-module, a representative aMb2 ligand (Fig. 4B). Expres-
sion of radixin N-ERMAD increased the number of adherent
aMb2/CHO cells more than threefold when compared to that
of GFP (radixin vs. GFP, P < 0.005), whereas expression of ra-
dixin C-ERMAD had no effect (Fig. 4C). Verifying specificity
of the adhesion assays, addition of EDTA or NIF, a high affin-
ity aMb2-specific inhibitor [9], completely blocked the aMb2-
mediated cell adhesion. Moreover, no significant adhesion to
the Fg c-module was observed for mock-transfected CHO cells
(Fig. 4C). Finally, to investigate whether radixin N-ERMAD
promoted the adhesive activity of aMb2 by changing its confor-
mation, we conducted FACS analyses using an activation-spe-
cific mAb (mAb 24), which has been used previously by us
[16,17] and others [8,18] to probe the activation status of
aMb2. Fig. 4D shows that expression of radixin N-ERMAD
significantly increased mAb 24 binding to aMb2 (P = 0.002),
suggesting that in the presence of radixin N-ERMAD aMb2 ex-
ists in an active conformation. Thus, radixin N-ERMAD most
likely enhanced aMb2-mediated cell adhesion by activating the
receptor.
4. Discussion
Talin is the only known FERM-containing protein capable
of integrin activation [1]. In this study, we identified another
FERM domain-containing protein radixin that exhibits similar
integrin activation capability. Compared to talin, radixin
exhibits two distinct features: first, unlike talin, radixin recog-
nizes the membrane-proximal region but not the NPxY/F se-
quence within the integrin b cytoplasmic tail; and second,
talin but not radixin, is sensitive to calpain proteolysis and
β2 CT: WKALIHLSDLREYRRFEKEKLKSQWNNDNPLFKSATTTVMNPKFAES
ICAM-2 CT: WHRRRTGTYGVLAAWRRL
A
B
Fig. 5. The integrin b2 cytoplasmic domain binding site within the radixin FERM domain. (A) Sequence alignment between the cytoplasmic domains
of ICAM-2 and integrin b2. Functional residues that are conserved between these two proteins are shown in boxes. (B) The b2 binding site within the
radixin FERM domain. A schematic drawing of the b2 cytoplasmic tail/radixin FERM complex was prepared from the crystal structure of the
radixin/ICAM-2 complex [22]. The b2 peptide is represented in red. The radixin FERM domain is composed of three subdomains: Fl (in cyan), F2 (in
orange) and F3 (in green). The conserved residues are shown in balls and sticks.
P. Tang et al. / FEBS Letters 581 (2007) 1103–1108 1107
activation [19]. Thus, radixin may function as a talin-indepen-
dent integrin activation pathway under specific settings.
Radixin interacts with the cytoplasmic domains of a number
of cell surface proteins, including ICAM-1, ICAM-2, ICAM-3,
CD44, and CD43 [20,21]. Though no consensus sequences
among these different cytoplasmic domains have been identi-
fied for radixin binding, most of the known radixin recognition
sequences reside within the membrane-proximal regions and
contain clusters of hydrophobic and positively charged resi-
dues [21]. Indeed, results from this study demonstrate that
radixin binds to the membrane-proximal region of integrin
b2, which shares a significant homology with the radixin-recog-
nition sequence within ICAM-2, especially surrounding those
resides (shown in boxes, Fig. 5A) that are in direct contacts
with the PTB/F3 domain of radixin (Fig. 5B) [22]. Accord-
ingly, we hypothesize that the LxxLxxYRRF sequence within
the b2 cytoplasmic domain, a sequence that is homologous to
the radixin-binding site (YxxLxxWRRL) within ICAM-2, rep-
resents the major recognition site for radixin (Fig. 5B, con-
served residues are shown in balls and sticks). Experiments
to test this hypothesis are under way.
In summary, using a number of different approaches, we have
established radixin as an important player in integrin activation.
Our data demonstrate that radixin is capable of binding and
activating the b2 integrin receptor, thus identifying radixin as
the second FERM-containing protein, in addition to talin, that
supports integrin ‘‘inside-out’’ signaling. As radixin and talin
recognize different regions of the integrin cytoplasmic tail, and
are activated by independent intracellular signaling pathways,
for example, radixin is activated by Rho-dependent kinase
(ROCK) [23] and phosphatidylinositol 4-phosphate 5-kinase
[24], whereas talin is primarily activated by calpain-mediated
cleavage [2], radixin may represent a novel talin-independent
pathway for integrin activation under specific settings.
Acknowledgments: We thank Drs. Hawley, Medved, Hogg and Plow
for providing valuable reagents. This work was supported in part by
grants (R01 HL61589 and P01 HL54710) from the National Institutes
of Health.
References
[1] Tadokoro, S., Shattil, S.J., Eto, K., Tai, V., Liddington, R.C., de
Pereda, J.M., Ginsberg, M.H. and Calderwood, D.A. (2003) Talin
binding to integrin beta tails: a final common step in integrin
activation. Science 302, 103–106.
[2] Yan, B., Calderwood, D.A., Yaspan, B. and Ginsberg, M.H.
(2001) Calpain cleavage promotes talin binding to the beta 3
integrin cytoplasmic domain. J. Biol. Chem. 276, 28164–28170.
[3] Martel, V., Racaud-Sultan, C., Dupe, S., Marie, C., Paulhe, F.,
Galmiche, A., Block, M.R. and Albiges-Rizo, C. (2001) Confor-
mation, localization, and integrin binding of talin depend on its
interaction with phosphoinositides. J. Biol. Chem. 276, 21217–
21227.
[4] Rees, D.J., Ades, S.E., Singer, S.J. and Hynes, R.O. (1990)
Sequence and domain structure of talin. Nature 347, 685–689.
[5] Calderwood, D.A., Yan, B., de Pereda, J.M., Alvarez, B.G.,
Fujioka, Y., Liddington, R.C. and Ginsberg, M.H. (2002) The
phosphotyrosine binding-like domain of talin activates integrins.
J. Biol. Chem. 277, 21749–21758.
[6] Hughes, P.E., Diaz-Gonzalez, F., Leong, L., Wu, C., McDonald,
J.A., Shattil, S.J. and Ginsberg, M.H. (1996) Breaking the integrin
hinge. A defined structural constraint regulates integrin signaling.
J. Biol. Chem. 271, 6571–6574.
[7] Cheng, L., Du, C., Murray, D., Tong, X., Zhang, Y.A., Chen,
B.P. and Hawley, R.G. (1997) A GFP reporter system to assess
gene transfer and expression in human hematopoietic progenitor
cells. Gene Ther. 4, 1013–1022.
[8] Dransfield, I. and Hogg, N. (1989) Regulated expression of Mg2+
binding epitope on leukocyte integrin alpha subunits. EMBO J. 8,
3759–3765.
[9] Muchowski, P.J., Zhang, L., Chang, E.R., Soule, H.R., Plow,
E.F. and Moyle, M. (1994) Functional interaction between the
integrin antagonist neutrophil inhibitory factor and the I domain
of CD1 1b/CD18. J. Biol. Chem. 269, 26419–26423.
[10] Zhang, L. and Plow, E.F. (1996) Overlapping, but not identical,
sites are involved in the recognition of C3bi, neutrophil inhibitory
factor, and adhesive ligands by the alphaMbeta2 integrin. J. Biol.
Chem. 271, 18211–18216.
[11] Cao, C., Lawrence, D.A., Li, Y., von Arnim, C.A., Herz, J., Su,
E.J., Makarova, A., Hyman, B.T., Strickland, D.K. and Zhang,
L. (2006) Endocytic receptor LRP together with tPA and PAI-1
coordinates Mac-1-dependent macrophage migration. EMBO J.
25, 1860–1870.
[12] Ugarova, T.P., Solovjov, D.A., Zhang, L., Loukinov, D.I., Yee,
V.C., Medved, L.V. and Plow, E.F. (1998) Identification of a
novel recognition sequence for integrin alphaM beta2 within the
gamma-chain of fibrinogen. J. Biol. Chem. 273, 22519–22527.
[13] Li, Y., Lawrence, D.A. and Zhang, L. (2003) Sequences within
domain II of the urokinase receptor critical for differential ligand
recognition. J. Biol. Chem. 278, 29925–29932.
[14] Magendantz, M., Henry, M.D., Lander, A. and Solomon, F.
(1995) Interdomain interactions of radixin in vitro. J. Biol. Chem.
270, 25324–25327.
[15] Geiger, C., Nagel, W., Boehm, T., van Kooyk, Y., Figdor, C.G.,
Kremmer, E., Hogg, N., Zeitlmann, L., Dierks, H., Weber, K.S.
and Kolanus, W. (2000) Cytohesin-1 regulates beta-2 integrin-
mediated adhesion through both ARF-GEF function and inter-
action with LFA-1. EMBO J. 19, 2525–2536, JID-8208664.
[16] Ehirchiou, D., Xiong, Y.M., Li, Y., Brew, S. and Zhang, L. (2005)
Dual function for a unique site within the beta2I domain of
integrin alphaMbeta2. J. Biol. Chem. 280, 8324–8331.
[17] Xiong, Y.M., Chen, J. and Zhang, L. (2003) Modulation of
CD11b/CD18 adhesive activity by its extracellular, membrane-
proximal regions. J. Immunol. 171, 1042–1050.
[18] Lu, C., Shimaoka, M., Zang, Q., Takagi, J. and Springer, T.A.
(2001) Locking in alternate conformations of the integrin alpha
Lbeta 2 I domain with disulfide bonds reveals functional
relationships among integrin domains. Proc. Natl. Acad. Sci.
USA 98, 2393–2398.
[19] Shcherbina, A., Bretscher, A., Kenney, D.M. and Remold-
O’Donnell, E. (1999) Moesin, the major ERM protein of
lymphocytes and platelets, differs from ezrin in its insensitivity
to calpain. FEBS Lett. 443, 31–36.
[20] Tsukita, S., Oishi, K., Sato, N., Sagara, J., Kawai, A. and
Tsukita, S. (1994) ERM family members as molecular linkers
between the cell surface glycoprotein CD44 and actin-based
cytoskeletons. J. Cell Biol. 126, 391–401.
[21] Yonemura, S., Hirao, M., Doi, Y., Takahashi, N., Kondo, T.,
Tsukita, S. and Tsukita, S. (1998) Ezrin/radixin/moesin (ERM)
proteins bind to a positively charged amino acid cluster in the
juxta-membrane cytoplasmic domain of CD44, CD43, and
ICAM-2. J. Cell Biol. 140, 885–895.
[22] Hamada, K., Shimizu, T., Yonemura, S., Tsukita, S., Tsukita, S.
and Hakoshima, T. (2003) Structural basis of adhesion-molecule
recognition by ERM proteins revealed by the crystal structure of
the radixin-ICAM-2 complex. EMBO J. 22, 502–514.
[23] Matsui, T., Maeda, M., Doi, Y., Yonemura, S., Amano, M.,
Kaibuchi, K. and Tsukita, S. (1998) Rho-kinase phosphorylates
COOH-terminal threonines of ezrin/radixin/moesin (ERM) pro-
teins and regulates their head-to-tail association. J. Cell. Biol. 140,
647–657.
[24] Matsui, T., Yonemura, S., Tsukita, S. and Tsukita, S. (1999)
Activation of ERM proteins in vivo by Rho involves phospha-
tidyl-inositol 4-phosphate 5-kinase and not ROCK kinases. Curr.
Biol. 9, 1259–1262.
1108 P. Tang et al. / FEBS Letters 581 (2007) 1103–1108

More Related Content

What's hot

Tomasz Pawlak, TIGP SI poster, 2015
Tomasz Pawlak, TIGP SI poster, 2015Tomasz Pawlak, TIGP SI poster, 2015
Tomasz Pawlak, TIGP SI poster, 2015Tomasz Pawlak
 
PPAR alpha final research paper
PPAR alpha final research paperPPAR alpha final research paper
PPAR alpha final research paperEric Cobb
 
NEDD_NatureCommunications2016
NEDD_NatureCommunications2016NEDD_NatureCommunications2016
NEDD_NatureCommunications2016Je-Hyun Baek
 
Advanced Medicinal Chemistry of GPCR Receptor
Advanced Medicinal Chemistry of GPCR ReceptorAdvanced Medicinal Chemistry of GPCR Receptor
Advanced Medicinal Chemistry of GPCR Receptorsaurabh gupta
 
Cell signaling- G protein
Cell signaling- G proteinCell signaling- G protein
Cell signaling- G proteinAly Barakat
 
Biochemj00056 0080
Biochemj00056 0080Biochemj00056 0080
Biochemj00056 0080Ana Bruno
 
SmartScreen Technology for Building a Better Assay
SmartScreen Technology for Building a Better AssaySmartScreen Technology for Building a Better Assay
SmartScreen Technology for Building a Better AssayKristin Rider
 
Post-translational Modifications in Crop Improvement
Post-translational Modifications in Crop ImprovementPost-translational Modifications in Crop Improvement
Post-translational Modifications in Crop ImprovementFaraZ AhmaD
 
Post Translational Modification
Post Translational ModificationPost Translational Modification
Post Translational ModificationLubnaKhayal
 

What's hot (20)

Tomasz Pawlak, TIGP SI poster, 2015
Tomasz Pawlak, TIGP SI poster, 2015Tomasz Pawlak, TIGP SI poster, 2015
Tomasz Pawlak, TIGP SI poster, 2015
 
SHMT_BBA_1994
SHMT_BBA_1994SHMT_BBA_1994
SHMT_BBA_1994
 
J med chem
J med chemJ med chem
J med chem
 
Biotech 2012 spring_8_post-trans
Biotech 2012 spring_8_post-transBiotech 2012 spring_8_post-trans
Biotech 2012 spring_8_post-trans
 
PPAR alpha final research paper
PPAR alpha final research paperPPAR alpha final research paper
PPAR alpha final research paper
 
NEDD_NatureCommunications2016
NEDD_NatureCommunications2016NEDD_NatureCommunications2016
NEDD_NatureCommunications2016
 
J med chem-2
J med chem-2J med chem-2
J med chem-2
 
Advanced Medicinal Chemistry of GPCR Receptor
Advanced Medicinal Chemistry of GPCR ReceptorAdvanced Medicinal Chemistry of GPCR Receptor
Advanced Medicinal Chemistry of GPCR Receptor
 
Cell signaling- G protein
Cell signaling- G proteinCell signaling- G protein
Cell signaling- G protein
 
Biochemj00056 0080
Biochemj00056 0080Biochemj00056 0080
Biochemj00056 0080
 
769.full
769.full769.full
769.full
 
G- Protein Coupled Receptors
G- Protein Coupled ReceptorsG- Protein Coupled Receptors
G- Protein Coupled Receptors
 
SHMT_ABB_1996
SHMT_ABB_1996SHMT_ABB_1996
SHMT_ABB_1996
 
SmartScreen Technology for Building a Better Assay
SmartScreen Technology for Building a Better AssaySmartScreen Technology for Building a Better Assay
SmartScreen Technology for Building a Better Assay
 
Post-translational Modifications in Crop Improvement
Post-translational Modifications in Crop ImprovementPost-translational Modifications in Crop Improvement
Post-translational Modifications in Crop Improvement
 
Post Translational Modification
Post Translational ModificationPost Translational Modification
Post Translational Modification
 
G protein coupled receptors copy
G protein coupled receptors   copyG protein coupled receptors   copy
G protein coupled receptors copy
 
Blaiser poster
Blaiser posterBlaiser poster
Blaiser poster
 
Ferrara 2012
Ferrara 2012Ferrara 2012
Ferrara 2012
 
CcP2APX_Biochem_2008
CcP2APX_Biochem_2008CcP2APX_Biochem_2008
CcP2APX_Biochem_2008
 

Viewers also liked

Blake Whittaker Resume-2
Blake Whittaker Resume-2Blake Whittaker Resume-2
Blake Whittaker Resume-2Blake Whittaker
 
“Advertisement in the Economic Community of West African States (ECOWAS)”: Le...
“Advertisement in the Economic Community of West African States (ECOWAS)”: Le...“Advertisement in the Economic Community of West African States (ECOWAS)”: Le...
“Advertisement in the Economic Community of West African States (ECOWAS)”: Le...Edmund Siah-Armah
 
Let's reading OSS's source
Let's reading OSS's sourceLet's reading OSS's source
Let's reading OSS's sourcetemama
 
Finance Academy Recruiting Slideshow
Finance Academy Recruiting Slideshow Finance Academy Recruiting Slideshow
Finance Academy Recruiting Slideshow HHSBizFinanceAcademy
 
Advertisement in the Economic Community of West African States (ECOWAS): Lear...
Advertisement in the Economic Community of West African States (ECOWAS): Lear...Advertisement in the Economic Community of West African States (ECOWAS): Lear...
Advertisement in the Economic Community of West African States (ECOWAS): Lear...Edmund Siah-Armah
 
Using a learning management system to improve and monetize your website
Using a learning management system to improve and monetize your websiteUsing a learning management system to improve and monetize your website
Using a learning management system to improve and monetize your websiteH. Trevor Johnson-Steigelman
 
Mobile app development Company india
Mobile app development Company indiaMobile app development Company india
Mobile app development Company indiaMobileAPPtelligence
 
13а Karina Finke
13а Karina Finke13а Karina Finke
13а Karina FinkeKarinaFinke
 

Viewers also liked (15)

Blake Whittaker Resume-2
Blake Whittaker Resume-2Blake Whittaker Resume-2
Blake Whittaker Resume-2
 
“Advertisement in the Economic Community of West African States (ECOWAS)”: Le...
“Advertisement in the Economic Community of West African States (ECOWAS)”: Le...“Advertisement in the Economic Community of West African States (ECOWAS)”: Le...
“Advertisement in the Economic Community of West African States (ECOWAS)”: Le...
 
How to create an app
How to create an appHow to create an app
How to create an app
 
Let's reading OSS's source
Let's reading OSS's sourceLet's reading OSS's source
Let's reading OSS's source
 
Xie X et al 2014
Xie X et al 2014Xie X et al 2014
Xie X et al 2014
 
Finance Academy Recruiting Slideshow
Finance Academy Recruiting Slideshow Finance Academy Recruiting Slideshow
Finance Academy Recruiting Slideshow
 
Advertisement in the Economic Community of West African States (ECOWAS): Lear...
Advertisement in the Economic Community of West African States (ECOWAS): Lear...Advertisement in the Economic Community of West African States (ECOWAS): Lear...
Advertisement in the Economic Community of West African States (ECOWAS): Lear...
 
колом
коломколом
колом
 
Using a learning management system to improve and monetize your website
Using a learning management system to improve and monetize your websiteUsing a learning management system to improve and monetize your website
Using a learning management system to improve and monetize your website
 
Mobile app development Company india
Mobile app development Company indiaMobile app development Company india
Mobile app development Company india
 
resume_GAUTAMKUMAR_(1)
resume_GAUTAMKUMAR_(1)resume_GAUTAMKUMAR_(1)
resume_GAUTAMKUMAR_(1)
 
How to make an app
How to make an appHow to make an app
How to make an app
 
13а Karina Finke
13а Karina Finke13а Karina Finke
13а Karina Finke
 
портфоліо коломицкая
портфоліо коломицкаяпортфоліо коломицкая
портфоліо коломицкая
 
A NEW POLITICAL SCIENCE.4
A NEW POLITICAL SCIENCE.4A NEW POLITICAL SCIENCE.4
A NEW POLITICAL SCIENCE.4
 

Similar to FEBS Letters 2007 Tang

The 5' terminal uracil of let-7a is critical for the recruitment of mRNA to A...
The 5' terminal uracil of let-7a is critical for the recruitment of mRNA to A...The 5' terminal uracil of let-7a is critical for the recruitment of mRNA to A...
The 5' terminal uracil of let-7a is critical for the recruitment of mRNA to A...David W. Salzman
 
Receptors with intrinsic protein kinase activity
Receptors with intrinsic protein kinase activityReceptors with intrinsic protein kinase activity
Receptors with intrinsic protein kinase activityDr.M.Prasad Naidu
 
Regulation of pten activity by its carboxyl terminal autoinhibitory
Regulation of pten activity by its carboxyl terminal autoinhibitoryRegulation of pten activity by its carboxyl terminal autoinhibitory
Regulation of pten activity by its carboxyl terminal autoinhibitoryChau Chan Lao
 
adenylate_cyclase_poster
adenylate_cyclase_posteradenylate_cyclase_poster
adenylate_cyclase_posterKelly Thompson
 
project- Liron Schwartz
project- Liron Schwartzproject- Liron Schwartz
project- Liron SchwartzLiron Schwartz
 
RAGE-Mediated Cell Signaling.pdf
RAGE-Mediated Cell Signaling.pdfRAGE-Mediated Cell Signaling.pdf
RAGE-Mediated Cell Signaling.pdfNazmunNahar479158
 
Evaluation of the lacZ gene in Escherichia coli mutagenesis using pBluescript...
Evaluation of the lacZ gene in Escherichia coli mutagenesis using pBluescript...Evaluation of the lacZ gene in Escherichia coli mutagenesis using pBluescript...
Evaluation of the lacZ gene in Escherichia coli mutagenesis using pBluescript...Emilio Solomon
 
Senior Thesis-Analyzing the interactions between MYOGEF and a component of er...
Senior Thesis-Analyzing the interactions between MYOGEF and a component of er...Senior Thesis-Analyzing the interactions between MYOGEF and a component of er...
Senior Thesis-Analyzing the interactions between MYOGEF and a component of er...Dougan McGrath
 
Mol Cell Proteomics-2013-Koytiger-1204-13
Mol Cell Proteomics-2013-Koytiger-1204-13Mol Cell Proteomics-2013-Koytiger-1204-13
Mol Cell Proteomics-2013-Koytiger-1204-13Greg Koytiger
 
Genetic Dna And Bioinformatics ( Accession No. Xp Essay
Genetic Dna And Bioinformatics ( Accession No. Xp EssayGenetic Dna And Bioinformatics ( Accession No. Xp Essay
Genetic Dna And Bioinformatics ( Accession No. Xp EssayJessica Deakin
 
A high copy number screen of the trs23Δ99C mutant
A high copy number screen of the trs23Δ99C mutantA high copy number screen of the trs23Δ99C mutant
A high copy number screen of the trs23Δ99C mutantGabriel Bendavit
 
Fussion protein and affinity Purification
Fussion protein and affinity PurificationFussion protein and affinity Purification
Fussion protein and affinity PurificationIndraSaputra493101
 

Similar to FEBS Letters 2007 Tang (20)

The 5' terminal uracil of let-7a is critical for the recruitment of mRNA to A...
The 5' terminal uracil of let-7a is critical for the recruitment of mRNA to A...The 5' terminal uracil of let-7a is critical for the recruitment of mRNA to A...
The 5' terminal uracil of let-7a is critical for the recruitment of mRNA to A...
 
J. Biol. Chem.-2015-Maganti-9812-22
J. Biol. Chem.-2015-Maganti-9812-22J. Biol. Chem.-2015-Maganti-9812-22
J. Biol. Chem.-2015-Maganti-9812-22
 
BIOS 5260 Term Paper
BIOS 5260 Term PaperBIOS 5260 Term Paper
BIOS 5260 Term Paper
 
Receptors with intrinsic protein kinase activity
Receptors with intrinsic protein kinase activityReceptors with intrinsic protein kinase activity
Receptors with intrinsic protein kinase activity
 
Regulation of pten activity by its carboxyl terminal autoinhibitory
Regulation of pten activity by its carboxyl terminal autoinhibitoryRegulation of pten activity by its carboxyl terminal autoinhibitory
Regulation of pten activity by its carboxyl terminal autoinhibitory
 
adenylate_cyclase_poster
adenylate_cyclase_posteradenylate_cyclase_poster
adenylate_cyclase_poster
 
CELL SIGNALING
CELL SIGNALINGCELL SIGNALING
CELL SIGNALING
 
project- Liron Schwartz
project- Liron Schwartzproject- Liron Schwartz
project- Liron Schwartz
 
RAGE-Mediated Cell Signaling.pdf
RAGE-Mediated Cell Signaling.pdfRAGE-Mediated Cell Signaling.pdf
RAGE-Mediated Cell Signaling.pdf
 
Evaluation of the lacZ gene in Escherichia coli mutagenesis using pBluescript...
Evaluation of the lacZ gene in Escherichia coli mutagenesis using pBluescript...Evaluation of the lacZ gene in Escherichia coli mutagenesis using pBluescript...
Evaluation of the lacZ gene in Escherichia coli mutagenesis using pBluescript...
 
Senior Thesis-Analyzing the interactions between MYOGEF and a component of er...
Senior Thesis-Analyzing the interactions between MYOGEF and a component of er...Senior Thesis-Analyzing the interactions between MYOGEF and a component of er...
Senior Thesis-Analyzing the interactions between MYOGEF and a component of er...
 
Mol Cell Proteomics-2013-Koytiger-1204-13
Mol Cell Proteomics-2013-Koytiger-1204-13Mol Cell Proteomics-2013-Koytiger-1204-13
Mol Cell Proteomics-2013-Koytiger-1204-13
 
srep43028
srep43028srep43028
srep43028
 
Genetic Dna And Bioinformatics ( Accession No. Xp Essay
Genetic Dna And Bioinformatics ( Accession No. Xp EssayGenetic Dna And Bioinformatics ( Accession No. Xp Essay
Genetic Dna And Bioinformatics ( Accession No. Xp Essay
 
( Isi, 2016) international journal of biological macromolecules gst kappa 201...
( Isi, 2016) international journal of biological macromolecules gst kappa 201...( Isi, 2016) international journal of biological macromolecules gst kappa 201...
( Isi, 2016) international journal of biological macromolecules gst kappa 201...
 
A high copy number screen of the trs23Δ99C mutant
A high copy number screen of the trs23Δ99C mutantA high copy number screen of the trs23Δ99C mutant
A high copy number screen of the trs23Δ99C mutant
 
MS Poster
MS PosterMS Poster
MS Poster
 
StAR
StARStAR
StAR
 
review
reviewreview
review
 
Fussion protein and affinity Purification
Fussion protein and affinity PurificationFussion protein and affinity Purification
Fussion protein and affinity Purification
 

FEBS Letters 2007 Tang

  • 1. Cytoskeletal protein radixin activates integrin aMb2 by binding to its cytoplasmic tail Pingtao Tang, Chunzhang Cao, Min Xu, Li Zhang* Center for Vascular and Inflammatory Diseases, Department of Physiology, University of Maryland School of Medicine, 800 W Baltimore Street, Baltimore, MD 21201, USA Received 1 December 2006; revised 4 February 2007; accepted 6 February 2007 Available online 15 February 2007 Edited by Gianni Cesareni Abstract Talin binding of integrins, via its band 4.1, ezrin, ra- dixin, and moesin (FERM)-homologous domain, directly acti- vates the integrin receptor. However, it is not known whether other FERM-containing proteins also possess such an integrin activating capability. We report here that radixin, one of the ori- ginal FERM-domain proteins, binds to the membrane-proximal region of the integrin b2 but not aM cytoplasmic tail. Impor- tantly, we show that radixin binding significantly enhances the adhesive activity of integrin aMb2. Given the distinct biological activities of radixin and talin, radixin may represent a novel talin-independent pathway for integrin activation under specific settings. Ó 2007 Federation of European Biochemical Societies. Pub- lished by Elsevier B.V. All rights reserved. Keywords: Cell adhesion; Integrin; The band 4.1, ezrin, radixin, and moesin-homologous domain; Cytoskeleton 1. Introduction Recent studies suggest that talin binding to the cytoplasmic domain of integrin receptors represents the final step in inte- grin activation (‘‘inside-out’’ signaling) [1]. Talin comprises of an N-terminal 50-kDa head region that is homologous to the FERM domain, and a C-terminal 205-kDa rod region, which regulates the function of the N-terminal head via intra- and inter-molecular interactions. Talin can be activated by cal- pain-mediated proteolysis [2] or by phosphoinositide binding [3], and once activated, it binds, through the phosphotyrosine binding (PTB) domain located within its FERM-homologous region, to a highly conserved NPxY motif within the integrin b cytoplasmic tail [4,5]. Integrin engagement by talin subse- quently disrupts a salt bridge between its a and b cytoplasmic domains [6], leading to its activation [5]. Interestingly, although a number of proteins, including syk, Numb, and Dok, contain a PTB domain that interacts with the integrin b cytoplasmic tail with high affinity, however, none of them was capable of activating integrin receptor [5]. Thus, the talin FERM domain seems very unique in its integrin activating capability. In this study, we searched for additional proteins that are capable of activating integrins. We report here the identifica- tion of another FERM protein radixin that possesses similar integrin activating activities. Using aMb2 (Mac-1, CD11b/ CD18) as a representative of integrin receptors, we demon- strate that radixin binds to the membrane-proximal region of the integrin b cytoplasmic tail via its N-terminal ezrin–radix- in–moesin family (ERM) associated domain (N-ERMAD). We show that radixin binding to integrin aMb2 enhances its adhesive activity. Together, our study identifies radixin as an important player in integrin ‘‘inside-out’’ signaling in addition to talin. As radixin recognizes the integrin cytoplasmic tail dif- ferently than talin, radixin may represent a novel talin-inde- pendent pathway for integrin activation under specific settings. 2. Materials and methods 2.1. Materials MGIN [7] was provided by Dr. R.G. Hawley (American Red Cross, Rockville, MD). IB4, OKM1, and TS1/18 were from ATCC (Rock- ville, MD); mAb 44 and 2LPM19c were from Sigma (St. Louis, MO). mAb 24 [8] was provided by Dr. Nancy Hogg (Imperial Cancer Research Fund, London, UK). Recombinant c-module of fibrinogen (Fg) was provided by Dr. L. Medved (University of Maryland, MD). Neutrophil inhibitor factor (NIF) [9] was provided by Dr. E. Plow (Cleveland Clinic Foundation, OH). All other reagents were of the highest grade available from Sigma Chemical Co. (St Louis, MO) unless otherwise noted. 2.2. Expression of recombinant radixin in E. coli To express the radixin N-ERMAD (1–298) and C-ERMAD (281– 584) in E. coli, their corresponding cDNAs were amplified using the fol- lowing pairs of primers: 50 -AAAGGATCC-ATGCCGAAGCCAAT- CAATGTAAGAGT-30 and 50 -TCAGTCGACTT-AATCAGGCTTT- CTCCTTCGCATGTACAG-30 , and 50 -AAAGGATCCTTGGCCT- TATGTATGG-GAAACATGAA-30 and 50 -GAAGTCGACTCA- CATGGCTTCAAACTCATCGATGCG-30 , and then cloned into pGEX-4T-1 (for glutathione-s-transferase (GST) fusion; Amersham, Piscataway, NJ) and pET-32a (for histidine-tag fusion; Novagen, Mad- ison, WI). Cytohesin-1 was expressed as a histidine-tagged protein using pQE-31 (QIAgen). Generation of recombinant proteins in E. coli BL21(DE3) and their purification using the glutathione–Sepharose or Ni–NTA column have been described previously [9]. The concentrations of the purified proteins were determined spectrophotometrically using Abbreviations: C-ERMAD, C-terminal ezrin–radixin–moesin family (ERM) associated domain; CHO, Chinese hamster cells; ERM, the ezrin–radixin–moesin protein family; ERMAD, ERM-associated domain; FERM, the band 4.1, ezrin, radixin, and moesin-homologous domain; Fg, fibrinogen; GFP, green fluorescent protein; HRP, horse- radish peroxidase; ICAM-1, intercellular adhesion molecule-1; N-E- RMAD, N-terminal ezrin–radixin–moesin family (ERM) associated domain; PBS, phosphate buffered saline; PTB, phosphotyrosine bind- ing domain; GST, glutathione-s-transferase; SDS–PAGE, sodium dodecyl sulfate polyacrylamide gel electrophoresis * Corresponding author. Fax: +1 410 706 8121. E-mail address: lizhang@som.umaryland.edu (L. Zhang). 0014-5793/$32.00 Ó 2007 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2007.02.013 FEBS Letters 581 (2007) 1103–1108
  • 2. the equation C (mg/ml) = (A280–A320)/1.45 and the concentrations were further verified by Coomassie blue-stained SDS–PAGE using BSA as a standard. 2.3. Binding assays by ELISA Ninety six-well polystyrene microtiter plates (Costar, Corning, NY) were coated with different synthetic peptides (1 mg/ml) overnight at 4 °C and post-coated with 10% milk for 1 h at room temperature. GST-tagged radixin fragments or His-tagged cytohesin-1 (both at 150 nM) in phosphate buffered saline (PBS) plus 0.1%BSA were added and incubated for 40 min at room temperature. After washing with PBS, bound GST-radixin was detected using a horseradish peroxidase (HRP)-anti-GST conjugate, and an HRP substrate 3,30 ,5,50 -tetrameth- ylbenzidine, measuring absorption at 550 nm (A550). 2.4. Pull down assays Cells expressing aMb2, which were established previously [10], were lysed in 20 mM Tris–HCl, 150 mM NaCl, pH 7.4, 0.25% deoxycholate, 1% p-nitrophenyl-p-guanidinobezoate, 2 mM EGTA, 1 mM PMSF, 10 mM benzamidine, 25 lg/ml SBTI, and 20 lg/ml leupeptin. After centrifugation, cell lysates were incubated with GST-radixin N-ER- MAD that was immobilized onto a glutathione agarose column. After washing, bound proteins were eluted and then analyzed by 10% SDS– PAGE. The presence of aMb2 was detected by immunobloting using an anti-b2 cytoplasmic domain antibody [11] and visualized using an HRP conjugate of goat anti-rabbit IgG. 2.5. Establishment of Chinese hamster cells (CHO) expressing both integrin aMb2 and radixin To express aMb2 and radixin simultaneously, cDNAs encoding N- ERMAD and C-ERMAD were inserted individually into the expres- sion vector MGIN [7], in which radixin expression is tightly linked to green fluorescent protein (GFP) expression via an internal ribosome entry site (IRES). These expression plasmids were transfected, together with a expression vector containing the hygromycin cDNA, into the aMb2-expressing CHO cells that we previously established [10]. After selection with 200 lg/ml hygromycin, stable clones that express equiv- alent amount of aMb2 and radixin were established by dual-color cell sorting using a mAb anti-aM (OKM1) and GFP as cell sorting mark- ers. 2.6. Cell adhesion assays Cell adhesion was conducted using Fg or its aMb2-recognition do- main, the c-module [10,12], based on our published procedures [13]. 3. Results 3.1. The N-terminal FERM domain of radixin interacts with the b2 cytoplasmic tail To see if talin is the only FERM domain-containing protein that is capable of activating integrins, we tested the ability of radixin to interact with the b2 integrin cytoplasmic tail. Given that radixin exists in an inactive state due to an intramolecular interaction between its N- and C-terminal ERMADs [14], we expressed N-ERMAD (residues 1–298) and C-ERMAD (resi- dues 281–584) separately as either GST- or His-tagged proteins. All of the recombinant proteins exhibited expected molecular sizes and were more than 90% pure (data not shown). Interactions between radixin fragments and integrin cyto- plasmic tails were assessed by ELISA. Thus, 96-well microtiter plates were coated with aM (CYKLGFFKRQYKDMM- SEGGPPGAEPQ) or b2 (CWKALIHLSDLREYRRFEK- EKLKSQWNNDNPLFKSATTTVMNPKFAES) cytoplas- mic peptides. Different concentrations of radixin fragments were added and bound proteins were quantified. As shown in Fig. 1A, N-ERMAD bound the b2 cytoplasmic peptide dose-dependently with a Kd of 110 ± 27 nM, whereas no signif- icant binding activity was observed for C-ERMAD. Moreover, neither radixin fragments interacted with the aM cytoplasmic peptide. Furthermore, no binding to an irrelevant peptide SW63 (CSTHFPGGGSISHSGHF) or a scrambled peptide sCTD31 (CLYRLEWFHAILRSRKD) was observed for N- ERMAD (data not shown), verifying the specificity of the binding assay. These data demonstrated that radixin recog- nizes the b2 integrin cytoplasmic tail via its N-terminal FERM domain. 3.2. Inhibition of N-ERMAD binding to integrin cytoplasmic domain by C-ERMAD To see if N-ERMAD binding to the integrin b2 cytoplasmic domain is controlled by C-ERMAD, we conducted the above binding assays in the presence of His-tagged C-ERMAD. Addition of C-ERMAD at ratios of 2:1 and 4:1 (C-ER- MAD:N-ERMAD) both led to significant inhibition on N- ERMAD binding to the b2 cytoplasmic peptide (Fig. 1B). As expected, N-ERMAD did not interact with the aM cytoplasmic tail with or without C-ERMAD. BoundradixinN-ERMAD(A550) 0.0 0.1 0.2 0.3 0.4 0.5 0.6 C/N Ratio: 0:1 2:1 4:1 the β2 peptide the αM peptide Radixin fragment added (nM) 0 50 100 150 200 250 Boundradixinfragments(A550) 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 B A Fig. 1. Radixin N-ERMAD interacts with the integrin b cytoplasmic tail. (A) Radixin N-ERMAD recognizes the b2 cytoplasmic domain. A 96-well plate was coated with either the aM (s, n) or b2 cytoplasmic peptides (d, m), blocked, and then incubated with GST-radixin N- ERMAD (s, d) or GST-C-ERMAD (n, m). After washing, bound proteins were determined. (B) Radixin C-ERMAD blocks N-ERMAD binding to the b2 cytoplasmic tail. Binding of radixin N-ERMAD to the aM (filled bar) or b2 (open bar) peptides were conducted as described above, except that different concentrations of His-tagged radixin C-ERMAD were added. Data are means ± S.D. of duplicate samples and are representative of three independent experiments. 1104 P. Tang et al. / FEBS Letters 581 (2007) 1103–1108
  • 3. 3.3. Radixin binds to the membrane-proximal region of the integrin b2 cytoplasmic tail To localize the functional sequence within the integrin b2 cytoplasmic tail for radixin recognition, we tested N-ERMAD binding to truncated b2 cytoplasmic peptides (Fig. 2A). We found that removal of NPxY/F, the sequence recognized by ta- lin, did not significantly affect N-ERMAD binding as both truncated b2 peptides CTD19 and CTD31 bound N-ERMAD well (Fig. 2B). In fact, they exhibited higher binding activity toward N-ERMAD than the full length peptide. Verifying specificity, N-ERMAD did not bind to either the scrambled peptide sCTD31 or BSA-coated wells. To further confirm the involvement of the membrane-prox- imal region of the b2 cytoplasmic tail in radixin recognition, the ability of soluble CTD31 to inhibit N-ERMAD binding of the full length b2 cytoplasmic peptide was tested. Thus, N-ERMAD was added to b2 peptide-coated 96-well plates in the presence of different concentrations of CTD31. After wash- ing, bound N-ERMAD was determined as above. Fig. 2C shows that CTD31, which contains only the membrane-proxi- mal 16 residues, competed effectively with the full length b2 peptide for radixin binding with a C50 of 25 lM, whereas its scrambled control peptide sCTD31 or DMSO had no effect. Together, these data demonstrated that the NPxY/F sequence within the b2 cytoplasmic domain, which is critical to talin rec- ognition, is not required for radixin binding. 3.4. Radixin and cytohesin-1 recognize overlapping binding sites within the b2 cytoplasmic tail Cytohesin-1 activates aLb2 by binding to residues 723–731 of b2 [15]. To see if cytohesin-1 and radixin share a common rec- ognition site within b2, we conducted additional binding exper- iments. Fig. 3A shows that cytohesin-1 binding to the full length b2 peptide was effectively inhibited by soluble CTD31 in a dose- dependent manner, whereas additions of similar con- centrations of DMSO or an irrelevant peptide SW63 had no effect and addition of (sCTD31) yielded much weaker inhibi- tion. To further confirm that radixin and cytohesin-1 recognize an overlapping site within the b2 cytoplasmic tail, we tested if CWKALIHLSDLREYRRFEKEKLKSQWNNDNPLFKSATTTVMNPKFAESWT: CWKALIHLSDLREYRRFEKEKLKSQWNNDCTΔ19: CLYRLEWFHAILRSRKDSCTΔ31: CWKALIHLSDLREYRRFCTΔ31: 723 | 750 | 738 | Figure 2 BoundradixinN-ERMAD(A550) 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 WT CTΔ19 CTΔ31 sCTΔ31 BSA Peptide added (μM) 0 100 200 300 400 BoundradixinN-ERMAD(A550) 0.0 0.5 1.0 1.5 2.0 DMSO CTΔ31 GST s CTΔ31 A B C Fig. 2. Radixin recognizes the membrane-proximal region of the integrin b2 cytoplasmic tail. (A) The sequence of the b2 cytoplasmic peptides. (B) Radixin N-ERMAD binds to the membrane-proximal region of the b2 cytoplasmic tail. Interaction between the different b2 cytoplasmic peptides and radixin N-ERMAD were determined by ELISA as above. (C) Inhibition of N-ERMAD binding to the b2 cytoplasmic peptide by CTD31. Radixin N-ERMAD binding to the b2 cytoplasmic tail was performed as above, except in the presence different soluble peptides. Data are means ± S.D. of duplicate samples and are representative of two independent experiments. sCTD31, scrambled CTD31. Cytohesin-1 (μM) 0 2 4 6 8 10 12 BoundradixinN-ERMAD(A550) 0.0 0.1 0.2 0.3 0.4 0.5 0.6 Peptide added (μM) 0 20 40 60 80 100 Boundcytohesin-1(A550) 0.0 0.5 1.0 1.5 2.0 2.5 DMSO CTΔ31 GST sCTΔ31 SW63 B A Fig. 3. Cytohesin-1 and radixin N-ERMAD recognize overlapping regions of the b2 cytoplasmic tail. (A) Cytohesin-1 binding to the b2 cytoplasmic tail. His-cytohesin-1 binding to the b2 cytoplasmic tail in the presence of different b2 peptides was performed as above. (B) Competition between Cytohesin-1 and radixin. Binding of GST- radixin N-ERMAD to the b2 cytoplasmic tail was conducted as in Fig. 1 except in the presence of His-tagged cytohesin-1 (h) or a control protein BSA (s). Data are means ± S.D. of duplicate samples and are representative of two independent experiments. P. Tang et al. / FEBS Letters 581 (2007) 1103–1108 1105
  • 4. ActivationIndex 0.00 0.02 0.04 0.06 0.08 0.10 0.12 GFP N-ERMAD clone 6 * Celladhesion(A570) 0 1 2 3 4 5 G FPN -ER M AD clone 6N -ER M AD poolC -ER M AD poolN -ER M AD pool+ED TA N -ER M AD clone 6+N IF m ock C -ER M A D pool+N IF A B C D Fig. 4. Expression of the N-terminal radixin promotes integrin adhesive activity. (A) Pull-down assays. Cell lysates from aMb2-expressing CHO cells were passed through either GST-N-ERMAD (lanes 1 and 3) or GST (lanes 2 and 4)-column. After washing, the bound proteins were analyzed by SDS–PAGE and immunoblotted with anti-b2 cytoplasmic tail (lanes 1 and 2) or anti-GST (lanes 3 and 4). (B) The effect of radixin N-ERMAD on cell adhesion. Different radixin fragments were expressed in aMb2/CHO cells and their adhesion to the Fg c-module (10 lg/ml) was determined in the presence or absence of an aMb2-specific inhibitor NIF (10 nM). After washing, representative adherent cells were photographed with 20· differential interference contrast (DIC) lens using the Zeiss Axiovert 200M system. Scale bars: 20 lm. (C) The numbers of adherent cells in (B) were quantified by staining with Crystal violet and measuring absorption at 570 nm. The specificity of the adhesion assay was confirmed by lack of cell adhesion of mock transfected cells. (D) Conformational changes within the aMb2 receptor. aMb2/CHO cells expressing either N-ERMAD or GFP were incubated with 1 lg of mAb 24 at 37 °C or 2LPM19c in the presence of 1 mM Ca2+ /1 mM Mg2+ for 30 min. Bound mAb 24 or 2LPM19c was determined by flow cytometry using R-Phycoerythrin conjugated anti-mouse IgG. An isotype matched non-immune IgG was used as a control (not shown). Mean fluorescence intensity of mAb 24, calculated using a FACScan program (BD Biosciences, Mountain View, CA) and normalized by surface expression of the receptor (based on 2LPM19c staining), was used as an activation index. N-ERMAD-expressing aMb2/CHO cells exhibited significantly higher mAb 24 binding compared to those expressing GFP (*, P = 0.002). Data are means ± S.D. of duplicate samples and are representative of three independent experiments. 1106 P. Tang et al. / FEBS Letters 581 (2007) 1103–1108
  • 5. radixin and cytohesin-1 could compete directly with each other for binding to the b2 cytoplasmic peptide. We found that in- deed cytohesin-1 competed with radixin N-ERMAD effectively with a C50 of 0.2 lM, whereas addition of BSA did not have significant effect (Fig. 3B). 3.5. Radixin binds integrin aMb2 in total cell lysates To test if radixin interacts with integrin receptors in whole cell lysates, we carried out pull-down experiments between ra- dixin N-ERMAD and aMb2, a major b2 integrin expressed on leukocytes. Fig. 4A shows that radixin N-ERMAD but not GST, both of which were immobilized onto glutathione–Se- pharose resin, could effectively pull down aMb2, as demon- strated by the presence of a 95 kDa protein by immunoblot using anti-b2 polyclonal antibody. Additional immunoblot analyses of the pull-down products with anti-GST antibody verified that equally amount of GST and GST-N-ERMAD were present on the resin. 3.6. Radixin promotes integrin adhesive activity To test if radixin binding to the b2 cytoplasmic domain could result in integrin activation, we transfected N-ERMAD or C-ERMAD of radixin into aMb2-expressing CHO cells, using the retroviral vector MGIN [7]. Stable cell lines that express equivalent levels of aMb2 on the cell surface and similar levels of radixin were established by cell sorting using both the aM-specific mAb OKM1 and GFP as markers. Surface expres- sion of aMb2 by the different clones and also pools of the aMb2/ radixin/CHO cells used in this study differed by less than two- fold (mean fluorescence intensity: 1594 for GFP, 1433 for C- ERMAD pool, 1035 for N-ERMAD clone 6, and 2555 for N-ERMAD pool). Cell adhesion experiments demonstrated that expression of radixin N-ERMAD in aMb2/CHO cells sig- nificantly increased the adhesive activity of aMb2 toward the Fg c-module, a representative aMb2 ligand (Fig. 4B). Expres- sion of radixin N-ERMAD increased the number of adherent aMb2/CHO cells more than threefold when compared to that of GFP (radixin vs. GFP, P < 0.005), whereas expression of ra- dixin C-ERMAD had no effect (Fig. 4C). Verifying specificity of the adhesion assays, addition of EDTA or NIF, a high affin- ity aMb2-specific inhibitor [9], completely blocked the aMb2- mediated cell adhesion. Moreover, no significant adhesion to the Fg c-module was observed for mock-transfected CHO cells (Fig. 4C). Finally, to investigate whether radixin N-ERMAD promoted the adhesive activity of aMb2 by changing its confor- mation, we conducted FACS analyses using an activation-spe- cific mAb (mAb 24), which has been used previously by us [16,17] and others [8,18] to probe the activation status of aMb2. Fig. 4D shows that expression of radixin N-ERMAD significantly increased mAb 24 binding to aMb2 (P = 0.002), suggesting that in the presence of radixin N-ERMAD aMb2 ex- ists in an active conformation. Thus, radixin N-ERMAD most likely enhanced aMb2-mediated cell adhesion by activating the receptor. 4. Discussion Talin is the only known FERM-containing protein capable of integrin activation [1]. In this study, we identified another FERM domain-containing protein radixin that exhibits similar integrin activation capability. Compared to talin, radixin exhibits two distinct features: first, unlike talin, radixin recog- nizes the membrane-proximal region but not the NPxY/F se- quence within the integrin b cytoplasmic tail; and second, talin but not radixin, is sensitive to calpain proteolysis and β2 CT: WKALIHLSDLREYRRFEKEKLKSQWNNDNPLFKSATTTVMNPKFAES ICAM-2 CT: WHRRRTGTYGVLAAWRRL A B Fig. 5. The integrin b2 cytoplasmic domain binding site within the radixin FERM domain. (A) Sequence alignment between the cytoplasmic domains of ICAM-2 and integrin b2. Functional residues that are conserved between these two proteins are shown in boxes. (B) The b2 binding site within the radixin FERM domain. A schematic drawing of the b2 cytoplasmic tail/radixin FERM complex was prepared from the crystal structure of the radixin/ICAM-2 complex [22]. The b2 peptide is represented in red. The radixin FERM domain is composed of three subdomains: Fl (in cyan), F2 (in orange) and F3 (in green). The conserved residues are shown in balls and sticks. P. Tang et al. / FEBS Letters 581 (2007) 1103–1108 1107
  • 6. activation [19]. Thus, radixin may function as a talin-indepen- dent integrin activation pathway under specific settings. Radixin interacts with the cytoplasmic domains of a number of cell surface proteins, including ICAM-1, ICAM-2, ICAM-3, CD44, and CD43 [20,21]. Though no consensus sequences among these different cytoplasmic domains have been identi- fied for radixin binding, most of the known radixin recognition sequences reside within the membrane-proximal regions and contain clusters of hydrophobic and positively charged resi- dues [21]. Indeed, results from this study demonstrate that radixin binds to the membrane-proximal region of integrin b2, which shares a significant homology with the radixin-recog- nition sequence within ICAM-2, especially surrounding those resides (shown in boxes, Fig. 5A) that are in direct contacts with the PTB/F3 domain of radixin (Fig. 5B) [22]. Accord- ingly, we hypothesize that the LxxLxxYRRF sequence within the b2 cytoplasmic domain, a sequence that is homologous to the radixin-binding site (YxxLxxWRRL) within ICAM-2, rep- resents the major recognition site for radixin (Fig. 5B, con- served residues are shown in balls and sticks). Experiments to test this hypothesis are under way. In summary, using a number of different approaches, we have established radixin as an important player in integrin activation. Our data demonstrate that radixin is capable of binding and activating the b2 integrin receptor, thus identifying radixin as the second FERM-containing protein, in addition to talin, that supports integrin ‘‘inside-out’’ signaling. As radixin and talin recognize different regions of the integrin cytoplasmic tail, and are activated by independent intracellular signaling pathways, for example, radixin is activated by Rho-dependent kinase (ROCK) [23] and phosphatidylinositol 4-phosphate 5-kinase [24], whereas talin is primarily activated by calpain-mediated cleavage [2], radixin may represent a novel talin-independent pathway for integrin activation under specific settings. Acknowledgments: We thank Drs. Hawley, Medved, Hogg and Plow for providing valuable reagents. This work was supported in part by grants (R01 HL61589 and P01 HL54710) from the National Institutes of Health. References [1] Tadokoro, S., Shattil, S.J., Eto, K., Tai, V., Liddington, R.C., de Pereda, J.M., Ginsberg, M.H. and Calderwood, D.A. (2003) Talin binding to integrin beta tails: a final common step in integrin activation. Science 302, 103–106. [2] Yan, B., Calderwood, D.A., Yaspan, B. and Ginsberg, M.H. (2001) Calpain cleavage promotes talin binding to the beta 3 integrin cytoplasmic domain. J. Biol. Chem. 276, 28164–28170. [3] Martel, V., Racaud-Sultan, C., Dupe, S., Marie, C., Paulhe, F., Galmiche, A., Block, M.R. and Albiges-Rizo, C. (2001) Confor- mation, localization, and integrin binding of talin depend on its interaction with phosphoinositides. J. Biol. Chem. 276, 21217– 21227. [4] Rees, D.J., Ades, S.E., Singer, S.J. and Hynes, R.O. (1990) Sequence and domain structure of talin. Nature 347, 685–689. [5] Calderwood, D.A., Yan, B., de Pereda, J.M., Alvarez, B.G., Fujioka, Y., Liddington, R.C. and Ginsberg, M.H. (2002) The phosphotyrosine binding-like domain of talin activates integrins. J. Biol. Chem. 277, 21749–21758. [6] Hughes, P.E., Diaz-Gonzalez, F., Leong, L., Wu, C., McDonald, J.A., Shattil, S.J. and Ginsberg, M.H. (1996) Breaking the integrin hinge. A defined structural constraint regulates integrin signaling. J. Biol. Chem. 271, 6571–6574. [7] Cheng, L., Du, C., Murray, D., Tong, X., Zhang, Y.A., Chen, B.P. and Hawley, R.G. (1997) A GFP reporter system to assess gene transfer and expression in human hematopoietic progenitor cells. Gene Ther. 4, 1013–1022. [8] Dransfield, I. and Hogg, N. (1989) Regulated expression of Mg2+ binding epitope on leukocyte integrin alpha subunits. EMBO J. 8, 3759–3765. [9] Muchowski, P.J., Zhang, L., Chang, E.R., Soule, H.R., Plow, E.F. and Moyle, M. (1994) Functional interaction between the integrin antagonist neutrophil inhibitory factor and the I domain of CD1 1b/CD18. J. Biol. Chem. 269, 26419–26423. [10] Zhang, L. and Plow, E.F. (1996) Overlapping, but not identical, sites are involved in the recognition of C3bi, neutrophil inhibitory factor, and adhesive ligands by the alphaMbeta2 integrin. J. Biol. Chem. 271, 18211–18216. [11] Cao, C., Lawrence, D.A., Li, Y., von Arnim, C.A., Herz, J., Su, E.J., Makarova, A., Hyman, B.T., Strickland, D.K. and Zhang, L. (2006) Endocytic receptor LRP together with tPA and PAI-1 coordinates Mac-1-dependent macrophage migration. EMBO J. 25, 1860–1870. [12] Ugarova, T.P., Solovjov, D.A., Zhang, L., Loukinov, D.I., Yee, V.C., Medved, L.V. and Plow, E.F. (1998) Identification of a novel recognition sequence for integrin alphaM beta2 within the gamma-chain of fibrinogen. J. Biol. Chem. 273, 22519–22527. [13] Li, Y., Lawrence, D.A. and Zhang, L. (2003) Sequences within domain II of the urokinase receptor critical for differential ligand recognition. J. Biol. Chem. 278, 29925–29932. [14] Magendantz, M., Henry, M.D., Lander, A. and Solomon, F. (1995) Interdomain interactions of radixin in vitro. J. Biol. Chem. 270, 25324–25327. [15] Geiger, C., Nagel, W., Boehm, T., van Kooyk, Y., Figdor, C.G., Kremmer, E., Hogg, N., Zeitlmann, L., Dierks, H., Weber, K.S. and Kolanus, W. (2000) Cytohesin-1 regulates beta-2 integrin- mediated adhesion through both ARF-GEF function and inter- action with LFA-1. EMBO J. 19, 2525–2536, JID-8208664. [16] Ehirchiou, D., Xiong, Y.M., Li, Y., Brew, S. and Zhang, L. (2005) Dual function for a unique site within the beta2I domain of integrin alphaMbeta2. J. Biol. Chem. 280, 8324–8331. [17] Xiong, Y.M., Chen, J. and Zhang, L. (2003) Modulation of CD11b/CD18 adhesive activity by its extracellular, membrane- proximal regions. J. Immunol. 171, 1042–1050. [18] Lu, C., Shimaoka, M., Zang, Q., Takagi, J. and Springer, T.A. (2001) Locking in alternate conformations of the integrin alpha Lbeta 2 I domain with disulfide bonds reveals functional relationships among integrin domains. Proc. Natl. Acad. Sci. USA 98, 2393–2398. [19] Shcherbina, A., Bretscher, A., Kenney, D.M. and Remold- O’Donnell, E. (1999) Moesin, the major ERM protein of lymphocytes and platelets, differs from ezrin in its insensitivity to calpain. FEBS Lett. 443, 31–36. [20] Tsukita, S., Oishi, K., Sato, N., Sagara, J., Kawai, A. and Tsukita, S. (1994) ERM family members as molecular linkers between the cell surface glycoprotein CD44 and actin-based cytoskeletons. J. Cell Biol. 126, 391–401. [21] Yonemura, S., Hirao, M., Doi, Y., Takahashi, N., Kondo, T., Tsukita, S. and Tsukita, S. (1998) Ezrin/radixin/moesin (ERM) proteins bind to a positively charged amino acid cluster in the juxta-membrane cytoplasmic domain of CD44, CD43, and ICAM-2. J. Cell Biol. 140, 885–895. [22] Hamada, K., Shimizu, T., Yonemura, S., Tsukita, S., Tsukita, S. and Hakoshima, T. (2003) Structural basis of adhesion-molecule recognition by ERM proteins revealed by the crystal structure of the radixin-ICAM-2 complex. EMBO J. 22, 502–514. [23] Matsui, T., Maeda, M., Doi, Y., Yonemura, S., Amano, M., Kaibuchi, K. and Tsukita, S. (1998) Rho-kinase phosphorylates COOH-terminal threonines of ezrin/radixin/moesin (ERM) pro- teins and regulates their head-to-tail association. J. Cell. Biol. 140, 647–657. [24] Matsui, T., Yonemura, S., Tsukita, S. and Tsukita, S. (1999) Activation of ERM proteins in vivo by Rho involves phospha- tidyl-inositol 4-phosphate 5-kinase and not ROCK kinases. Curr. Biol. 9, 1259–1262. 1108 P. Tang et al. / FEBS Letters 581 (2007) 1103–1108