Competitive Interactions of Collagen and a Jararhagin-derivedDisintegrin Peptide with the Integrin �2-I Domain*

Received for publication, December 24, 2007, and in revised form, April 15, 2008 Published, JBC Papers in Press, April 16, 2008, DOI 10.1074/jbc. M710483200

Lester J. Lambert, Andrey A. Bobkov, Jeffrey W. Smith, and Francesca M. Marassi1

From the Burnham Institute for Medical Research, La Jolla, California 92037

Integrin �2�1 is a major receptor required for activation andadhesion of platelets, through the specific recognition of colla-gen by the �2-I domain (�2-I), which binds fibrillar collagen viaMg2�-bridged interactions. The crystal structure of a truncatedform of the �2-I domain, bound to a triple helical collagen pep-tide, revealed conformational changes suggestive of a mecha-nism where the ligand-bound I domain can initiate and propa-gate conformational change to the full integrin complex.Collagen binding by �2-I and fibrinogen-dependent plateletactivity can be inhibited by snake venom polypeptides. Here wedescribe the inhibitory effect of a short cyclic peptide derivedfrom the snake toxin metalloprotease jararhagin, with specificamino acid sequence RKKH, on the ability of �2-I to bind triplehelical collagen. Isothermal titration calorimetry measure-ments showed that the interactions of �2-I with collagen orRKKH peptide have similar affinities, and NMR chemical shiftmapping experiments with 15N-labeled �2-I, and unlabeledRKKHpeptide, indicate that the peptide competes for the colla-gen-binding site of �2-I but does not induce a large scale con-formational rearrangement of the I domain.

The integrins constitute a functionally versatile family ofintegral membrane receptors that mediate cell-cell and cell-extracellular matrix interactions through their regulation ofcell adhesion, differentiation, migration, and the immuneresponse (1–5). Signal transduction is bi-directional throughboth outside-in and inside-out mechanisms. All integrins areheterodimers composed of subunits � and �. Different combi-nations of subunits are expressed ondifferent cell typeswith theinterplay of 19 � and 8 � subunits, generating a family of 25different heterodimers (3, 5).The integrin receptors share common structural features.

The extracellular portions of the � and � subunits combine toform a globular “head” domain that is attached to a pair ofmembrane-spanning helical “stalks.” Signal transduction isbelieved to involve an allosteric rearrangement characterizedby the separation and reorientation of the stalk segments. Thebidirectional nature of signal transduction is complex. Extra-cellular ligands induce outside-in signals by binding to fixed

motifs in the head domain, whereas inside-out signaling ensuesfrom intracellular interactions between relatively short struc-turally plastic control elements and a large repertoire of cellularproteins.In nine of the human� subunits, ligand recognition is carried

out by a 200-residue structurally conserved inserted (I) domainor a von Willebrand factor A domain (3, 5). The I and Adomains adopt a Rossmann dinucleotide-binding fold, with a6-stranded �-sheet surrounded by seven �-helices, and ligandrecognition requires the binding of a singleMg2� ion to ametalion-dependent adhesion (MIDAS)2 motif (6, 7). The impor-tance of the � I domain for understanding conformational reg-ulation and ligand binding for all integrins has been reviewedrecently (5).Integrin �2�1 is a member of the collagen/laminin receptor

family and is amajor receptor required for activation and adhe-sion of platelets, through the specific recognition of collagen bythe �2-I domain (�2-I) (8), which binds fibrillar collagen viaMg2�-bridged interactions, supported by the MIDAS motifresidues Asp-151, Ser-153, Thr-221, and Asp-254. The crystalstructure of the �2-I domain, bound to a triple helical collagenpeptide, revealed conformational changes from an unbound“closed” form to a bound “open” form, suggestive of a mecha-nism where the ligand-bound I domain can initiate and propa-gate conformational change to the full integrin complex (9).Collagen binding and fibrinogen-dependent platelet activity

can be inhibited by snake venompolypeptide toxins, enhancingthe effects of hemorrhagic venom metalloproteases. These so-called disintegrins are functional homologues of the Arg-Gly-Asp (RGD) motif found in extracellular matrix proteins. Inte-grin�2�1 associates with Jararhagin, a 52-kDametalloproteaseisolated from the venomof theBrazilian pit viperBothrops jara-raca, that targets multiple components in hemostasis, includ-ing vonWillebrand factor, fibrinogen, and platelet aggregation.Anti-platelet activity is thought to stem from its specificity forthe �2�1 integrin (10–14).

Notably, a short cyclic peptide derived from the jararhaginmetalloprotease domain, containing the specific amino acidsequence RKKH, is sufficient to prevent binding of type I colla-gen to�2-I in a competitivemanner and is capable of disruptingcell adhesion to type I collagen (15). The cyclic RKKH peptide-binding site coincides with the collagen-binding site, near the Idomain MIDAS motif (16). The inhibitory effects of theRKKH peptides on the homologous �1�1 integrin have been

* This work was supported, in whole or in part, by National Institutes of HealthGrants HL080718 and HL072862. The costs of publication of this articlewere defrayed in part by the payment of page charges. This article musttherefore be hereby marked “advertisement” in accordance with 18 U.S.C.Section 1734 solely to indicate this fact.

1 To whom correspondence should be addressed: Burnham Institute for Med-ical Research, 10901 North Torrey Pines Rd., La Jolla, CA 92037. Tel.: 858-795-5282; Fax: 858-713-6268; E-mail: fmarassi@burnham.org.

2 The abbreviations used are: MIDAS, metal ion-dependent adhesion; HSQC,heteronuclear single quantum spectroscopy; ITC, isothermal titration cal-orimetry; HPLC, high pressure liquid chromatography.

THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 283, NO. 24, pp. 16665–16672, June 13, 2008© 2008 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in the U.S.A.

JUNE 13, 2008 • VOLUME 283 • NUMBER 24 JOURNAL OF BIOLOGICAL CHEMISTRY 16665

by guest on April 2, 2018

http://ww

w.jbc.org/

Dow

nloaded from

suggested to reflect the ability of the peptide to mimic thenatural type I collagen ligand by inducing or stabilizing aconformational transition from the closed to the open formof the I domain (17).In this study we characterize the conformational dynamics

of the �2-I domain and its interactions with collagen andRKKH peptides. Our studies show that the RKKH peptidebinds the �2-I domain and inhibits its association with typeI collagen, without inducing a conformational change in the�2-I domain.

EXPERIMENTAL PROCEDURES

Materials—The pET-28b expression plasmid was fromInvitrogen. (15NH4)2SO4, [13C]glucose, and D2O were pur-chased from Cambridge Isotopes Laboratories. Nickel-nitrilo-triacetic acid-agarose was from Qiagen. The reverse phaseDelta Pak C18 HPLC column (300 � 7.8 mm; 300 Å) wasobtained from Waters. The 24-residue type I collagen peptidewith the GFOGER recognition sequence (O � hydroxyproline,Hyp) was purchased in HPLC-purified powder form fromBiomer Technology (Hayward, CA). The partially purifiedRKKHpeptide was purchased fromGenscript (Piscataway, NJ).The peptide sequences are shown in Fig. 1.Expression and Purification of Integrin �2-I Domain—The

recombinant �2-I domain used in these studies corresponds toresidues 144–334 of the human sequence (NCBI accessionnumber NP_002194). The DNA coding for the sequence wasinserted into the NdeI/XhoI site of the pET-28b expressionplasmid (Invitrogen). For eventual chemical derivatization, Ser-334, at the C terminus, was mutated to Cys. Site-directedmutagenesis was used to convert Cys-150 to Leu. The resultingplasmid, pNHis-�2-I(144–334)C150L, has a His6 tag at the Nterminus for purification. The sequence of the expressed pro-tein is shown in Fig. 1.Recombinant �2-I domain was expressed in Escherichia coli

strain BL21(DE3). To prepare 15N- and 13C-labeled protein forNMR experiments, cells bearing pNHis-�2-I(144–334)C150Lwere grown in minimal M9 media containing (15NH4)2SO4and/or [13C]glucose. Induction atA600 � 1with 1mM isopropyl1-thio-�-D-galactopyranoside for 2 h at 37 °C gave high levelexpression of the protein in the soluble fraction. Cells werelysed using a French press in buffer D (50 mM Tris, pH 8.0, 1 MNaCl, 30 mM imidazole), and the �2-I domain was purified bynickel-affinity chromatography on nickel-nitrilotriacetic acid-agarose, in buffer C (50 mM Tris, pH 7, 1 M NaCl, 300 mMimidazole). The purified protein solution was dialyzed againsttwo changes of buffer A (50 mM Tris, pH 8.0, 150 mM NaCl)with 1 mM EDTA, followed by two changes of buffer B (50 mMPO4, pH 7.0, 150 mM NaCl, 5 mM MgSO4) with 1 mM dithio-threitol, using dialysis membranes with a molecular weightcutoff of 10,000. The final yield of purified protein was 20mg/liter of cell culture. For NMR experiments, the proteinsolution was concentrated by ultrafiltration to 0.7 mM (� �17330 cm�1 M�1).For experiments with the RKKH peptide, the surface-ex-

posed C-terminal Cys residue of the �2-I domain was alkylatedwith iodoacetamide as follows. Purified protein was dialyzedagainst four changes of bufferA.The proteinwas removed from

the dialysis bag, reacted with 10 mM iodoacetamide for 10 h at4 °C, dialyzed against two changes of buffer B, and concentratedby ultrafiltration.Formation of Collagen Triple Helix—To obtain a collagen

triple helix, the 24-residue collagen peptide powder was sus-pended in buffer B at 1 mM concentration, heated to 45 °C, andthen allowed to equilibrate at 4 °C for at least 12 h, as describedpreviously (18).Formation of Cyclic RKKHPeptide—Formation of cyclic pep-

tide was obtained by forming a disulfide link between the ter-minal Cys residues as described (15). The peptide was dissolvedat 1 mg/ml concentration in 0.1 M NH4CO3 and incubated at4 °C for 24 h. The reaction was flash-frozen and the waterremoved by lyophilization. The cyclized peptide was resus-pended in HPLC grade water and purified by reverse phaseHPLC. Peak fractions were combined, frozen, and lyophilizedto powder. A colorimetric assay using 5,5�-dithiobis(2-nitro-benzoic acid), to test for the presence of free thiol, showed com-plete conversion to the cyclized product within the limits ofdetection. The cyclized peptide was suspended in buffer Bimmediately before experiments.NMR Spectroscopy—NMR experiments were performed on

Bruker AVANCE 600- and 800-MHz spectrometers. Thestandard 1H/15N fastHSQCpulse sequencewas used for exper-iments with peptides (19). Backbone resonance assignmentswere made using a standard CBCA(CO)NH experiment (20)and by comparison with the assignments reported previouslyby Elshorst et al. (21) for the same polypeptide. The chemicalshifts are referenced to the 1H2O resonance, set to its expectedposition of 4.87 ppm at 20 °C (22). The NMR data were pro-cessed using NMRPipe (23), and the spectra were assigned andanalyzed using Sparky (24). All experiments were performed atmillimolar concentrations of collagen or the RKKH to obtainsaturation of the �2-I domain, for a single site bindingmodel ofinteraction and the measured affinity of the peptides for �2-I(see below).Isothermal Titration Calorimetry (ITC)—The �2-I domain,

the triple helix collagen peptide, and the cyclized RKKH pep-tide were all dissolved in buffer B. The pH of each solutionwas measured to ensure that no changes were produced bythe polypeptide components. For the collagen bindingexperiments the concentration of �2-I domain in the cell was100 �M and that of the collagen peptide solution 1 mM. Forthe RKKH binding experiments, the �2-I domain was Cys-alkylated with iodoacetamide, and its concentration in thesample cell was 100 �M. The concentration of cyclizedRKKH was 1 mM.

ITC experiments were performed with a Microcal VP-ITCcalorimeter. Measurements were made by titration of collagenor RKKH peptide into the �2-I domain at a temperature of10 °C. For titration experiments, the �2-I domain was degassedand placed in the 1.4-ml reaction cell. The collagen or RKKHpeptides were loaded in the 250-�l injection syringe, and aseries of 8-�l injections over 16 s were made, with a spacing of500 s between injections over 300 min. The reference powerwas set to 20�cal/s, and the stirring speedwas 300 rpm. Parallelcontrol experiments, to correct for the heat of mixing, were

Integrin I Domain Collagen and RKKH Peptide Binding

16666 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 283 • NUMBER 24 • JUNE 13, 2008

by guest on April 2, 2018

http://ww

w.jbc.org/

Dow

nloaded from

performed by adding the peptide to a sample cell containingonly buffer without �2-I domain.The thermodynamic data were processed with the ORIGIN

program (Microcal) to extract the enthalpic, entropic, andequilibrium constants. Nonlinear least squares fitting was doneusing a single site binding model.Size Exclusion Chromatography—Size exclusion chromatog-

raphy was performed at 4 or 22 °C, using an Acta Prime flowsystem with a Superdex 75 10/300 column (GE Healthcare),running in buffer B plus 0.25 mM dithiothreitol. Samples of the�2-I domain (0.3 mM in buffer B) were combined with 0.15 mMof equilibrated collagen peptide at 4 °C in buffer B. Injections ofthe collagen and �2-I domain alone were done using the samebuffer. The flow rate was 0.4 ml/min for the 22 °C experimentsand 0.55ml/min for the 4 °C experiments. The protein-collagencomplexes were detected at UV wavelengths of 254 or 215 nm.For size comparison we utilized bovine serum albumin (66.3kDa) and lysozyme (14.4 kDa).

RESULTS

Association of �2-I with Type I Collagen Peptide and RKKHDisintegrin Peptide—In activated platelets �2�1 integrin hasa high affinity for soluble collagen with an associated Kd of35–90 nM (25), whereas the affinity of the �2-I domain fortype I collagen measured by surface plasmon resonance isweaker, with a Kd in the low micromolar range (26, 27). Rec-ognition by �2�1 resides totally within the collagensequence GFOGER, where Glu cannot be replaced by Asp,and sequence recognition is entirely dependent upon thepresence of a triple helical conformation (28). To determinethe affinity of �2-I for the triple helical type I collagen pep-tide, we performed ITC experiments where the 26-residuecollagen peptide, containing the GFOGER recognitionsequence (Fig. 1B), was titrated into the �2-I domain. Thispeptide sequence is similar to that which was co-crystallizedwith �2-I (9).To maintain the collagen peptide in its active triple helical

form, the ITC cell temperature was kept at 10 °C, below thepredicted melting temperature of about 20 °C (9). The resultsare shown in Fig. 2A, and the free energy parameters arereported in Table 1. Fitting the ITC data to a single-site bindingcurve (Fig. 2C) yields a Kd of 7.8 �M.

ITC was also performed to determine the affinity of �2-I fora cyclic RKKH peptide whose sequence had been previouslyidentified to have themost potent inhibitory effect on�2-I (Fig.1C) (15). To enable direct comparison with the affinity deter-mined for collagen, this study was also performed at 10 °C. Fig.2B shows the titration profile, and the thermodynamic param-eters are reported inTable 1. The data fit to a single-site bindingmodel (Fig. 2D) with a Kd of 8.0 �M, consistent with the IC50 of1.2 �M, estimated in competition assays for the inhibition ofcollagen binding (15).Characterization of the �2-I-Collagen Complex—To further

characterize the formation of the �2-I-collagen complex, weperformed size exclusion chromatography at temperaturesbelow or above the melting transition of the collagen triplehelix (Fig. 3, A and B). Isolated �2-I elutes with an apparentmolecular mass near 30 kDa at both temperatures (peak a),whereas the collagen peptide elutes near 20 kDa (peak b). Weattribute the differences between these observed values andthose expected from the calculated molecular weights of theproteins (23 kDa for �2-I; 6.8 kDa for triple helical collagenpeptide; 2.3 kDa for monomeric collagen peptide) to thehydrodynamic radii of the molecules, which govern the elutionprofiles. In particular, the elution of collagen is likely to be domi-nated by the rod-like shape of the triple helix. However, it is inter-esting to note that the peptide elutes at a slightly higher apparentmolecular weight at 4 than 22 °C, reflecting triple helix formationbelow themelting temperature.This is further corroborated by the elution profiles of pre-

mixed �2-I and collagen. At 22 °C, the elution profile is identi-

FIGURE 1. Amino acid sequences of �2-I domain (A), collagen peptide (B),and RKKH peptide (C). A, underlined sequence corresponds to the His tagand thrombin cleavage site which were not removed.

FIGURE 2. Thermodynamic ITC characterization of �2-I domain interac-tions with triple helical collagen (A and C) and cyclic RKKH peptides (Band D). The ITC titration profiles are shown at top (A and B) for the incrementaladdition of either peptide into 100 �M �2-I at 10 °C. The fits of heat absorbedper mol of titrant are shown at the bottom (C and D).

TABLE 1Thermodynamic binding constants for the �2-I domain measuredby ITC

Peptide Kd �G �H T�S�M kcal/mol kcal/mol kcal/mol

(GPO)3GFOGER(GPO)3-NH2 7.8 � 0.24 �6.6 �1.171 5.44

CTRKKHDNAQC-NH2 8.0 � 0.33 �6.6 �0.254 6.35

Integrin I Domain Collagen and RKKH Peptide Binding

JUNE 13, 2008 • VOLUME 283 • NUMBER 24 JOURNAL OF BIOLOGICAL CHEMISTRY 16667

by guest on April 2, 2018

http://ww

w.jbc.org/

Dow

nloaded from

cal to that of the individual components, with two resolvedpeaks corresponding to either �2-I (Fig. 3A, peak a) or collagen(Fig. 3A, peak b). However, when the molecules are combinedand eluted at 4 °C, a new peak appears at a higher apparentmolecular mass of about 45 kDa (Fig. 3B, peak c), indicatingcomplex formation with triple helical collagen.These results are consistent with specific recognition by�2-I

of the GFOGER sequence properly displayed in triple helicalcollagen (28). The size exclusion results in Fig. 3 further showthat the�2-I domain and collagen form a stable long lived com-plex and help explain the NMR results, where the extremebroadening observed in the presence of collagen reflects theformation of a large slowly tumbling biomolecular species (seebelow).Effects ofMg2� and Collagen on �2-I—The �2-I domain pos-

sesses high affinity for type I collagen in the presence of thedivalent metal cation Mg2�. The metal-binding site consists ofMIDAS motif residues Asp-151, Ser-153, Ser-155, Thr-221,Asp-254, and Glu-256 which form an octahedral coordinationsphere composed of direct andwater-bridged interactions. Thestructural role ofMg2�has been examinedusing both x-ray andNMR methods in the CD11a/LFA-1 I domain (29–32), whereMg2�was found to play a role in ligand binding, and its removaldid not cause large scale structural change in the CD11a/LFA-1I domain.The crystal structure of �2-I bound to triple helical collagen

suggests that collagen binding is accompanied by a large con-formational rearrangement of the C-terminal helix, coupledwith changes in the coordination the Mg2� metal (9). Collagenbinding causes three concerted changes in the I domain; theloops of theMIDASmotif are perturbed because of a rearrange-ment upon insertion of a collagen Glu side chain into the metalcoordination sphere; helices h6 and h7 rearrange to open up thetop surface; and helix h7 moves downward to the opposite poleof the MIDAS motif. The rearrangement of helix h7 is thoughtto produce the large scale conformational changes experiencedby the integrin heterodimer during signaling. To see if the pro-tein dynamics and conformation associatedwithMg2� and col-lagen binding could be characterized in solution, we examined

the 1H/15N HSQC NMR spectrum of �2-I in the presence orabsence of metal and collagen.The 1H and 15N chemical shifts from protein backbone

amide groups are very sensitive to changes in protein confor-mation or chemical environment and can be used to monitorthe equilibrium exchange between states arising from free andligand-boundprotein (22). If the exchange rate is faster than thedifference between the chemical shifts measured for the twostates, then the system is in fast exchange, and one peak isobserved at the population-weighted average chemical shift ofthe two states. NMR can be used to detect weak binding orminor conformational rearrangements, and chemical shiftchanges as small as 0.02 ppmhave been reported forminor localeffects on protein structure resulting from binding of smallmolecules or modifications, whereas much larger changes (�1ppm) can reflect major conformational rearrangements(33–36).A sample of metal-free �2-I domain (isotopically enriched

with 15N) was prepared by exhaustive dialysis against EDTA.The 1H-15N HSQC spectra of the metal-free and Mg2�-bound forms of the �2-I domain are shown in Fig. 4A. Severalpeaks undergo measurable frequency changes reflectingmetal binding. A plot of the total change in 1H and 15Nchemical shifts against amino acid number shows that thepeaks with the largest frequency changes are localized toresidues involved in direct coordination of Mg2� in theMIDAS motif (Fig. 5).This is further highlighted by mapping the frequency

changes on the previously determined crystal structure of the�2-I domain (Fig. 6C). However, most of the peaks from otherresidues throughout the protein structure do not change, indi-cating that Mg2� binding does not induce a large scale confor-mational change of the �2-I domain in solution. This is similarto the results reported for LFA-1 (31).A potential indicator of close association between the metal-

binding site and conformational change in helix h7 is the chem-ical shift change observed for Glu-318 in the spectra frommet-al-bound and unbound �2-I. Glu-318 is located in the looppreceding the C-terminal helix and is characterized by a dis-tinctive downfield chemical shift (1H � 11.5 ppm). In the com-parison of the structures of the collagen-bound and unboundforms of the �2-I domain, it was noted that Glu-318 in theunbound form of �2-I domain is engaged in a salt bridge inter-action with Arg-288, which is not present in the collagen-bound form (9). In addition, Glu-318 andAsp-317 are observedto undergo a pronounced torsion angle change associated withthe displacement of helix h7. The observed change in Glu-318chemical shift upon addition of Mg2� indicates close associa-tion of this residue with the MIDAS-binding site. However,other residues in helix h7, including Asp-317, do not changeupon addition of metal, indicating that metal binding alonedoes not induce the conformational transition from the closedto the open state of the I domain.To probe conformational changes of the �2-I domain upon

binding to type I collagen, a series of NMR titration experi-ments were performed with a 26-residue collagen peptide con-taining the GFOGER recognition sequence, similar to that pre-viously co-crystallized with the �2-I domain (Fig. 1B). The

FIGURE 3. Characterization of the �2-I-collagen interaction by size exclu-sion chromatography. Samples of �2-I (0.3 mM; solid line), collagen peptide(0.15 mM; broken line), or combined �2-I/collagen (0.3 mM/0.15 mM; dottedline) were chromatographed at either 22 °C (A) or 4 °C (B). For each tem-perature, the apparent molecular weights were estimated using bovineserum albumin (66.3 kDa) and lysozyme (14.4 kDa) molecular weightstandards. The arrows indicate elution of �2-I (a), collagen (b), or the �2-I-collagen complex (c).

Integrin I Domain Collagen and RKKH Peptide Binding

16668 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 283 • NUMBER 24 • JUNE 13, 2008

by guest on April 2, 2018

http://ww

w.jbc.org/

Dow

nloaded from

melting temperature for triple helix formation of this peptide ispredicted to be around 20 °C (9). The collagen triple helix wasallowed to form for 12 h at 6 °C and then combined with �2-Isolution at 10 °C for NMR studies, which were also performedat 10 °C to maintain the triple helical conformation.Additions of increasing concentrations of triple helical colla-

gen lead to a dramatic progressive loss of signal intensity for allpeaks in the NMR spectrum of �2-I (Fig. 4B), reflecting theformation of a large slowly tumbling complex and/or slow con-formational exchange on theNMR time scale. The formation ofa large complex is consistent with the size exclusion data at lowtemperature. The reduced peak intensity is also exacerbated byslower molecular tumbling rates due the need to performexperiments at 10 °C. Further addition of collagen to �2-I,above 1:1 molar ratio, causes the NMR signals to disappearcompletely (data not shown). The apparent lack of concomitantline broadening for the remaining peaks (see Fig. 8, A–D) sug-gests that complex formation leads to the complete disappear-ance of the signals from �2-I/collagen, rather than a reductionin local backbone dynamics.

A few peaks are also seen to undergo a change in chemicalshift frequencies (Fig. 7), and for some of these, the presence oftwo distinct resonances, one shifted and the other at the samefrequencies as free�2-I, may reflect slow exchange between thebound and unbound forms of �2-I. Interestingly, these peaksmap to residues situated in (Gly-329) or near (Asp-145, Val-194, Thr-202, Ser-236, Ala-245, Glu-309, and Glu-318) theC-terminal helix (Fig. 6C) and may reflect the conformationalchange indicated by the crystal structure of collagen-bound�2-I. We note that in the crystal structure the last 13 aminoacids in the C terminus, including seven residues in helix h7,were not visible.Effect of Cyclic RKKH Peptide on �2-I—The ability of recom-

binant �2-I domain to recapitulate the interaction of the inte-grin�2�1 heterodimerwith RKKHhas been demonstrated (15)and provides the basis for the experiments described below.Mutational studies with �2-I indicate that cyclic RKKH pep-tides bind near the MIDAS motif �2-I (15, 16), suggesting amechanism of direct competition for the collagen-binding site,and the interaction of cyclic RKKHwith�1I has been suggestedto induce or stabilize a conformational transition from theclosed to the open form of integrin (17). To characterize the�2-I-RKKH interaction, we examined the influence of anRKKH sequence, previously identified to inhibit the associationof �2-I with type I collagen (15), on the solution NMR spectraof �2-I.The effect of RKKHon the 1H/15NHSQC spectrumof�2-I is

shown in Fig. 4C. Favorable linewidths and chemical shift dis-persion allowed 172 of the expected 188 �2-I resonances to beanalyzed. The addition of RKKH (1.5:1 RKKH:�2-I, molarratio) causes almost no changes in the HSQC spectrum of �2-I,with the exception of four peaks that undergo very small butmeasurable frequency changes (N 0.01 ppm; H 0.01ppm). Three of these peaks correspond to residues (Asn-189,Thr-221, and Asn-222) that map to the MIDAS motif of �2-I,near the predicted collagen-binding site (Fig. 6C), whereas afourth peak corresponds to His-272, situated on the opposite

FIGURE 4. Effects of Mg2� (A), triple helical collagen (B), and cyclic RKKH peptide (C) on the 1H/15N HSQC spectrum of �2-I domain. A, spectra ofMg2�-bound (black) or Mg2�-free (red) �2-I (0.5 mM) were obtained at 20 °C and 600 MHz, with either 1.75 or 0 mM MgCl2, respectively. Selected peaks thatundergo significant frequency changes are labeled. B, spectra of collagen-free (black) and collagen-bound (red) �2-I (0.3 mM) were obtained at 10 °C and 600MHz, with 0.3 mM triple helical collagen peptide (0.9 mM monomeric collagen peptide; 1:1 collagen:�2-I molar ratio). C, spectra of RKKH-free (black) andRKKH-bound (red) �2-I (0.5 mM) were obtained at 10 °C and 800 MHz, with 0.75 mM peptide (1.5:1 RKKH:�2-I molar ratio). B and C, spectra of free and bound �2-Iwere obtained in the presence of saturating Mg2� concentration.

FIGURE 5. Total change in �2-I domain backbone amide chemical shiftsinduced by the addition of Mg2� from 0 to 1. 7 mM (96% saturation). Thetotal combined change in chemical shift () for each residue was calculatedby adding the changes in 1H (H) and 15N (N) chemical shifts, according tothe equation � ((H)2 � (N/5)2)1/2, where the 15N chemical shift is scaledby 1/5 to account for the 5-fold difference between the chemical shift disper-sions of 15N and 1H (35, 36). Residue numbers are indicated for peaks with �0.1.

Integrin I Domain Collagen and RKKH Peptide Binding

JUNE 13, 2008 • VOLUME 283 • NUMBER 24 JOURNAL OF BIOLOGICAL CHEMISTRY 16669

by guest on April 2, 2018

http://ww

w.jbc.org/

Dow

nloaded from

pole far from the other perturbed residues, which is susceptibleto slight changes in buffer conditions.Notably, Thr-221 in the MIDAS motif is absolutely critical

for collagen binding to the recombinant �2-I domain, as well asfor ligand binding in the�M-I and�L-I domains (9, 37, 38). Thefinding that this is the residue most affected by cyclic RKKHpeptide indicates that it also plays a role in mediating the inter-action of �2-I with RKKH. Thr-221 is near Asp-219, which isalso critical for collagen binding (9). Asp-219 was predicted toplay a role in RKKH peptide binding (16); however, mutatingthis residue to Arg caused RKKH to bind with higher affinity(16), and theNMRdata showedno effect onAsp-219 because ofRKKH binding.Aside for the small changes observed for these few residues,

the near lack of changes in theNMR spectrumdoes not supportthe notion of a structural transition induced in the�2-I domainby the RKKH peptide. Instead, the high content of charged res-idues in the RKKH sequence suggests that charge-charge sidechain interactions are responsible formediating the associationbetween �2-I and RKKH.

To characterize the inhibitory effect of cyclic RKKHon�2-I/collagen binding, we examined the effect of RKKH peptide onthe intensity of NMR signals from the �2-I/collagen complex.Addition of RKKH to the�2-I/collagen sample restores someofthe �2-I signal intensity which had been lost upon collagenbinding. This is illustrated for four isolatedNMRpeaks in Fig. 8(A–D), however, the effect is uniform across the spectrum of�2-I, reflecting the effective inhibition of the �2-I/collageninteraction by RKKH. The effect of RKKH on peak intensityincreases with increasing peptide concentration up to a 2:1molar ratio of RKKH to �2-I, beyond which no further changesare observed (Fig. 8E). Taken together with the small frequencychanges observed upon RKKH binding, these results suggestthat RKKH inhibits by competing for the collagen-binding siteon �2-I.

To further test whether RKKH and collagen compete for thesame binding site on �2-I, we examined the effects of RKKH on�2-I peak intensity in sampleswhere collagenwas added to�2-Iafter RKKH (Fig. 8F). Although the addition of collagen to �2-I(1:1, collagen:�2-I, molar ratio) dramatically reduces the�2-I signal intensity by as much as 84%, the addition of RKKHto free �2-I (1.5:1, RKKH:�2-I, molar ratio) causes no changesin intensity. When �2-I is combined first with RKKH and thencollagen is added, the peak intensity decreases only by 22%, in amanner distinctly different from the addition of collagen alone.Further addition of collagen peptide up to 2:1 (collagen:�2-I,molar ratio) shows that as the RKKH concentration isexceeded, the peak heights are reduced to a level similar to thatof the addition of collagen alone.Thus we conclude that the mechanism of RKKH inhibition

involves direct binding of RKKH to �2-I. Both RKKH and col-lagen compete for the same binding site on the �2-I domain,with RKKH exchanging more rapidly than triple helical colla-

A B C

N189 T221N222

H272

E309

T202

S236 G329

N222

D151I156

G260

S257

E318

G284

V252

A226

A188T221

N190

A245 D145

V162

V194

CN

FIGURE 6. Molecular backbone representations of �2-I domain. The coordinates of the crystal structure (7) were obtained from the Protein Data Bank code1AOX. The Mg2� atom bound to the MIDAS motif is shown in yellow, and helix h7 is shown in teal. Residues shown in red have 1H/15N HSQC peaks that undergomeasurable frequency changes because of�2-I association with Mg2� at saturating concentration (A), triple helical collagen at 1:1, collagen:�2-I, molar ratio (B),and RKKH peptide at 1.5:1, RKKH:�2-I, molar ratio (C). Their C-� atoms are shown as spheres. A, residues in red have peaks with � 0.1 ppm.B, residues in red correspond to peaks that move in Fig. 7. C, residues in red have peaks that undergo minor but measurable frequency changes.

FIGURE 7. Expanded regions of the superimposed 1H/15N HSQC spectra of�2-I domain, obtained in the absence (black) or presence (red) of triplehelical collagen (1:1, collagen:�2-I, molar ratio) at 10 °C. The spectra showexamples of the frequency changes observed upon collagen binding.

Integrin I Domain Collagen and RKKH Peptide Binding

16670 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 283 • NUMBER 24 • JUNE 13, 2008

by guest on April 2, 2018

http://ww

w.jbc.org/

Dow

nloaded from

gen; however, RKKHbinding does not induce a conformationalchange in the �2-I domain.Pentikainen and co-workers (16) tested three integrin I

domain-specific antibodies (12F1, 5E8, andGi9) for their abilityto inhibit biotinylated RKKH peptide binding. Both 12F1,believed to target �2-I residues 173–199 (39) and/or residues212–216 (40), and Gi9, believed to target residues 199–216(41), were able to inhibit RKKH binding. Because the epitopesfor 12F1 reside near theMIDASmotif, these results are consist-ent with an RKKH/collagen competition for the �2-I-bindingsite, as demonstrated in this study.

DISCUSSION

The collagen binding activity of the �2�1 integrin residesentirely in the I domain and provides the basis for the develop-ment of integrin signaling inhibitors. Such inhibitors includethe jararhagin RKKHpeptide that is the focus of this study. TheRKKH inhibitors were originally identified by screening shortcyclic peptides for their ability to bind the recombinant �2-Idomain and disrupt collagen binding (15).Our findings confirm the inhibitory properties of the cyclic

RKKH peptide. However, they point to a mechanismwhere theRKKH cyclic peptide inhibits collagen binding without causinga conformational change of the �2-I domain. The jararhagintoxin possesses multiple domains that can inhibit �2�1 inte-grin, and these activities are required to enhance metallopro-tease-mediated cleavage of the �1 subunit (13, 14) with the neteffect of inhibition of collagen-induced activation of platelets.Jararhagin disintegrin possesses multiple motifs, and its activity iscomplex. Evidence supports an ability to induce �2�1-mediatedintegrin signaling inplatelets and fibroblasts.However, neither theRKKH motif nor the parent jararhagin metalloprotease domainhas beendemonstrated to induce activation of�2�1. Rather, if theRKKH motif anchors the jararhagin metalloprotease to the �2subunit, it may serve to facilitate other jararhagin motifs to playmore direct roles in integrin activation.Our results also indicate a mechanism of direct competition of

the RKKH ligand for the collagen-binding site. The �2-I residueMIDAS Thr-221 is involved in Mg2� metal coordination and isslightly perturbed by RKKH, suggesting that the RKKH-bindingsite coincideswith theMIDASmotif and thecollagen-binding site.The NMR perturbation data do not offer structural restraints fordocking the RKKH ligand on �2-I. However, the small chemicalshift change for a residue that directly coordinates Mg2� metalsuggests that the interaction of RKKH with �2-I is mediated bycharge-chargeorwater-bridgedcontacts to themetal-bindingsite.In contrast, the NMR results do not support the conclusions

of a previous study that suggested that RKKH binding causes aconformational change in the �1-I domain (17). A model ofRKKH docked on the crystal structure of recombinant �1-Ipredicted that the cyclic peptide could make extensive interac-tions with �1-I residues Arg-218, Glu-255, Ser-256, His-257,Asn-259, Ser-291, Glu-297, Glu-298, and Ser-301, near theMIDAS motif and in helix h6 (17). However, we did not detectany NMR peak perturbations for the corresponding residues inthe homologous �2-I domain.

The inhibition of collagen binding byRKKH ismetal depend-ent, and clearly the RKKH ligand does not displace Mg2�,because the removal of Mg2� has comparatively dramaticeffects on the NMR spectrum of �2-I (Fig. 4).Water-bridged orcharge-charge contacts with the binding site would alsoaccount for the ability of RKKH to inhibit collagen binding tothe free �2-I domain, and its inability to disrupt the �2-I-colla-gen complex once formed.Although the thermodynamic data show that the affinities of

cyclic RKKH and collagen are comparable, the types of molecularinteractions and possible conformational changes experienced bythe �2-I domain may contribute to a disparity in the apparent offrate. The ability to mimic triple helical collagen with a simplerligand would offer a powerful tool to study integrin signaling insolution and pave the way to drug development (42, 43).

Acknowledgment—The NMR studies utilized the Burnham InstituteNMR Facility, supported by National Institutes of Health Grant P30CA030199.

REFERENCES1. Hynes, R. O. (1992) Cell 69, 11–252. Hynes, R. O. (2002) Cell 110, 673–687

FIGURE 8. Isolated 1H NMR peaks and peak intensities for four selectedresidues (Thr-221, Met-249, Gly-260, and Glu-318) showing the inhibi-tion of �2-I collagen binding by cyclic RKKH. A–D, peaks were extractedfrom the 1H/15N HSQC spectra of �2-I (0.3 mM), obtained at 10 °C withoutcollagen or RKKH (solid line), after addition of 0.3 mM triple helical collagen(dotted line), and after the subsequent addition of 0.78 mM RKKH (broken line).E, plot of the peak intensity relative to that of free �2-I (a), measured afteraddition of 0.3 mM triple helical collagen (b), followed by addition of 0.3 mM

(c), 0.45 mM (d), or 0.6 mM (e) RKKH. Peak intensities are corrected for sampledilution. F, plot of the peak intensity relative to that of free �2-I (a), measuredafter addition of 0.45 mM RKKH (b), followed by 0.3 mM (c), 0.45 mM (d), or 0.6mM (e) collagen. Peak intensities are corrected for sample dilution.

Integrin I Domain Collagen and RKKH Peptide Binding

JUNE 13, 2008 • VOLUME 283 • NUMBER 24 JOURNAL OF BIOLOGICAL CHEMISTRY 16671

by guest on April 2, 2018

http://ww

w.jbc.org/

Dow

nloaded from

3. Shimaoka, M., Takagi, J., and Springer, T. A. (2002) Annu. Rev. Biophys.Biomol. Struct. 31, 485–516

4. Arnaout, M. A., Mahalingam, B., and Xiong, J. P. (2005) Annu. Rev. CellDev. Biol. 21, 381–410

5. Luo, B. H., Carman, C. V., and Springer, T. A. (2007)Annu. Rev. Immunol.25, 619–647

6. Lee, J. O., Rieu, P., Arnaout, M. A., and Liddington, R. (1995) Cell 80,631–638

7. Emsley, J., King, S. L., Bergelson, J. M., and Liddington, R. C. (1997) J. Biol.Chem. 272, 28512–28517

8. Tuckwell, D., Calderwood, D. A., Green, L. J., andHumphries,M. J. (1995)J. Cell Sci. 108, 1629–1637

9. Emsley, J., Knight, C. G., Farndale, R. W., Barnes, M. J., and Liddington,R. C. (2000) Cell 101, 47–56

10. Paine,M. J., Desmond, H. P., Theakston, R. D., andCrampton, J.M. (1992)J. Biol. Chem. 267, 22869–22876

11. De Luca, M., Ward, C. M., Ohmori, K., Andrews, R. K., and Berndt, M. C.(1995) Biochem. Biophys. Res. Commun. 206, 570–576

12. Kamiguti, A. S., Hay, C. R., Theakston, R. D., and Zuzel,M. (1996)Toxicon34, 627–642

13. Kamiguti, A. S., Hay, C. R., and Zuzel, M. (1996) Biochem. J. 320, 635–64114. Laing, G. D., and Moura-da-Silva, A. M. (2005) Toxicon 45, 987–99615. Ivaska, J., Kapyla, J., Pentikainen, O., Hoffren, A. M., Hermonen, J., Hut-

tunen, P., Johnson, M. S., and Heino, J. (1999) J. Biol. Chem. 274,3513–3521

16. Pentikainen,O.,Hoffren, A.M., Ivaska, J., Kapyla, J., Nyronen, T.,Heino, J.,and Johnson, M. S. (1999) J. Biol. Chem. 274, 31493–31505

17. Nymalm, Y., Puranen, J. S., Nyholm, T. K., Kapyla, J., Kidron, H., Penti-kainen, O. T., Airenne, T. T., Heino, J., Slotte, J. P., Johnson, M. S., andSalminen, T. A. (2004) J. Biol. Chem. 279, 7962–7970

18. Feng, Y., Melacini, G., Taulane, J. P., and Goodman, M. (1996) J. Am.Chem. Soc. 118, 10351–10358

19. Mori, S., Abeygunawardana, C., Johnson, M. O., and Vanzijl, P. C. M.(1995) J. Magn. Reson. 108, 94–98

20. Grzesiek, S., and Bax, A. (1992) J. Am. Chem. Soc. 114, 6291–629321. Elshorst, B., Jacobs, D. M., Schwalbe, H., and Langer, T. (2003) J. Biomol.

NMR 27, 191–19222. Cavanagh, J., Fairbrother, W. J., Palmer, A. G., and Skelton, N. J. (1996)

Protein NMR Spectroscopy: Principles and Practice, Academic Press, SanDiego

23. Delaglio, F., Grzesiek, S., Vuister, G. W., Zhu, G., Pfeifer, J., and Bax, A.(1995) J. Biomol. NMR 6, 277–293

24. Goddard, T. D., and Kneller, D. G. (2004) SPARKY 3, University of Cali-fornia, San Francisco

25. Jung, S. M., and Moroi, M. (1998) J. Biol. Chem. 273, 14827–1483726. Calderwood,D.A., Tuckwell, D. S., Eble, J., Kuhn, K., andHumphries,M. J.

(1997) J. Biol. Chem. 272, 12311–1231727. Cruz, M. A., Chen, J., Whitelock, J. L., Morales, L. D., and Lopez, J. A.

(2005) Blood 105, 1986–199128. Knight, C. G., Morton, L. F., Peachey, A. R., Tuckwell, D. S., Farndale,

R. W., and Barnes, M. J. (2000) J. Biol. Chem. 275, 35–4029. Qu, A., and Leahy, D. J. (1995) Proc. Natl. Acad. Sci. U. S. A. 92,

10277–1028130. Qu, A., and Leahy, D. J. (1996) Structure (Lond.) 4, 931–94231. Legge, G. B., Kriwacki, R. W., Chung, J., Hommel, U., Ramage, P., Case,

D. A., Dyson, H. J., and Wright, P. E. (2000) J. Mol. Biol. 295, 1251–126432. Shimaoka, M., Xiao, T., Liu, J. H., Yang, Y., Dong, Y., Jun, C. D., McCor-

mack, A., Zhang, R., Joachimiak, A., Takagi, J., Wang, J. H., and Springer,T. A. (2003) Cell 112, 99–111

33. Rajagopal, P., Waygood, E. B., and Klevit, R. E. (1994) Biochemistry 33,15271–15282

34. Shuker, S. B., Hajduk, P. J., Meadows, R. P., and Fesik, S.W. (1996) Science274, 1531–1534

35. Grzesiek, S., Bax, A., Clore, G. M., Gronenborn, A. M., Hu, J. S., Kaufman,J., Palmer, I., Stahl, S. J., and Wingfield, P. T. (1996) Nat. Struct. Biol. 3,340–345

36. Hoffman, R. M., Li, M. X., and Sykes, B. D. (2005) Biochemistry 44,15750–15759

37. Kamata, T., and Takada, Y. (1994) J. Biol. Chem. 269, 26006–2601038. Kamata, T., Wright, R., and Takada, Y. (1995) J. Biol. Chem. 270,

12531–1253539. Kamata, T., Puzon, W., and Takada, Y. (1994) J. Biol. Chem. 269,

9659–966340. King, S. L., Kamata, T., Cunningham, J. A., Emsley, J., Liddington, R. C.,

Takada, Y., and Bergelson, J. M. (1997) J. Biol. Chem. 272, 28518–2852241. Tuckwell, D. S., Smith, L., Korda, M., Askari, J. A., Santoso, S., Barnes,

M. J., Farndale, R. W., and Humphries, M. J. (2000) Biochem. J. 350,485–493

42. Yin, H., Gerlach, L. O., Miller, M. W., Moore, D. T., Liu, D., Vilaire, G.,Bennett, J. S., and DeGrado, W. F. (2006) Bioorg. Med. Chem. Lett. 16,3380–3382

43. Kapyla, J., Pentikainen, O. T., Nyronen, T., Nissinen, L., Lassander, S.,Jokinen, J., Lahti, M., Marjamaki, A., Johnson, M. S., and Heino, J. (2007)J. Med. Chem. 50, 2742–2746

Integrin I Domain Collagen and RKKH Peptide Binding

16672 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 283 • NUMBER 24 • JUNE 13, 2008

by guest on April 2, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Lester J. Lambert, Andrey A. Bobkov, Jeffrey W. Smith and Francesca M. Marassi2-I Domainαwith the Integrin

Competitive Interactions of Collagen and a Jararhagin-derived Disintegrin Peptide

doi: 10.1074/jbc.M710483200 originally published online April 16, 20082008, 283:16665-16672.J. Biol. Chem. 

  10.1074/jbc.M710483200Access the most updated version of this article at doi:

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/283/24/16665.full.html#ref-list-1

This article cites 41 references, 18 of which can be accessed free at

by guest on April 2, 2018

http://ww

w.jbc.org/

Dow

nloaded from