a6 peptide activates cd44 adhesive activity, induces fak ...a6 shares sequence homology with cd44,...

Therapeutic Discovery

A6 Peptide Activates CD44 Adhesive Activity, Induces FAKand MEK Phosphorylation, and Inhibits the Migration andMetastasis of CD44-Expressing Cells

Randolph S. Piotrowicz1, Bassam B. Damaj2, Mohamed Hachicha1, Francesca Incardona3,Stephen B. Howell4, and Malcolm Finlayson5

AbstractThe A6 peptide (acetyl-KPSSPPEE-amino) has antitumor activity in the absence of significant adverse

events in murine tumor models and clinical trials. A6 shares sequence homology with CD44, an adhesion

receptor involved in metastasis that is also a marker of cancer stem cells and drug-resistant phenotypes. We

investigated the mechanism of action of A6 by examining its effects on CD44 activity, cell migration, and

metastasis. A6 inhibited themigration of a subset of ovarian and breast cancer cell lines, exhibiting IC50 values

of 5 to 110 nmol/L. The ability of A6 to inhibit migration in vitro correlated with CD44 expression.

Immunopreciptation studies showed that CD44 binds A6 and that biotin-tagged A6 can be cross-linked

to CD44. The binding of A6 altered the structure of CD44 such that it was no longer recognized by a

monoclonal antibody to a specific epitope. Importantly, A6 potentiated the CD44-dependent adhesion of

cancer cells to hyaluronic acid and activated CD44-mediated signaling, as evidenced by focal adhesion kinase

and MAP/ERK kinase phosphorylation. In vivo, A6 (100 mg/kg delivered s.c. twice daily) reduced the

number of lung foci generated by the i.v. injection of B16-F10melanoma cells by 50% (P¼ 0.029 in an unpaired

t test). We conclude that A6 potently blocks themigration of CD44-positive cells in vitro through an interaction

with CD44 that alters its structure and activates CD44 to enhance ligand binding and downstream signaling.

The concurrent ability of A6 to agonize the CD44 receptor suggests that CD44 activationmay represent a novel

strategy for inhibiting metastatic disease. Mol Cancer Ther; 10(11); 2072–82. �2011 AACR.

Introduction

Short peptides with unique biological activities areemerging as potential modifiers of the biology of manytypes of cancer. A6 is an 8-amino acid peptide (acetyl-KPSSPPEE-amino) that has been shown to have anti-invasive, antimigratory, and antiangiogenic activities ina variety of in vitro and in vivo model systems (1–3). Onthe basis of these activities, A6 was advanced into clinicaltrials in which it was found to have an excellent safetyprofile with no systemic adverse events (4–6). In onestudy, conducted in ovarian cancer patients with earlybiochemical relapse, A6 treatment was associated withprolongation of progression-free survival (6).

CD44 is a heavily glycosylated membrane protein thatfunctions in cell–cell and cell–matrix adhesion (reviewedin Goodison and colleagues; ref. 7). The major ligand for

CD44 is hyaluronic acid (HA), which is an integral part ofthe extracellular matrix formed by tumor and stromalcells. The amino acid sequence of A6 exhibits markedhomology with a linear sequence within CD44. Thehomologous sequence (120-NASAPPEE-127) resideswithin the HA-binding domain of CD44 in an exposedlinker between 2 b strands (8). The A6-like sequence ispresent in all isoforms of CD44 and straddles the splicejunction between standard exons 3 and 4 (9). The aspar-agine residue, a potentialN-glycosylation site, is located 2residues carboxyl from a cysteine residue that is part of adisulfide bond that is critical for conformational integrityof CD44 and its ability to bind HA (8). Thus, the A6-likesequence in CD44 seems to be located within a region ofthe ligand-binding domain that is structurally importantand potentially conformationally dynamic.

Human CD44 gene is composed of 10 exons that areexpressed in the ubiquitous standard isoform of CD44(CD44s) and 10 variant exons that are alternativelyspliced within CD44s at an insertion site after the fifthstandard exon (9). In addition to HA, CD44 can bindseveral other extracellular matrix ligands includingosteopontin (10, 11), fibronectin (12), and collagen (13),and it interacts with other membrane proteins, includingHER2 (14). Three discrete regions of CD44 interact withHA (15–18). Whereas CD44 is constitutively active on

Authors' Affiliations: 1Bio-Quant, Inc., 2Apricus Bio, Inc., and 3NexMed,Inc., San Diego; 4Moores UCSD Cancer Center, La Jolla; and 5A

�ngstrom

Pharmaceuticals, Inc., Solana Beach, California

Corresponding Author: Randolph S. Piotrowicz, Angstrom Pharmaceu-ticals, 990 Highland Drive, Suite 314, Solana Beach, CA 92075. Phone:858-722-1939, Fax: 858-314-2356; E-mail: [email protected]

doi: 10.1158/1535-7163.MCT-11-0351

�2011 American Association for Cancer Research.

MolecularCancer

Therapeutics

Mol Cancer Ther; 10(11) November 20112072

on January 25, 2021. © 2011 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst September 1, 2011; DOI: 10.1158/1535-7163.MCT-11-0351

http://mct.aacrjournals.org/

some cells, ligand binding seems to be influenced in acell-specific manner byN-glycosylation (19) and, in someinstances, requires activation induced by the binding ofanti-CD44 monoclonal antibodies or treatment withphorbol esters and interleukins (20–22). Several mechan-isms have been implicated in CD44 activation includingvariant exon usage, receptor oligomerization, glycosyla-tion, and sulfation (for a review, see Ponta and colleagues;ref. 23). The key conformational determinants of theactivated form of CD44 remain unidentified; however,the interaction of HA with CD44 induces conformationalchanges that alter the orientation of a crucial HA-bindingarginine residue (24).This study identifies CD44 as a target for the anti-

migratory and antitumor effects of the A6 peptide. Itwas found that A6 inhibited chemotaxis, enhancedadhesion to HA, and induced focal adhesion kinase(FAK) and MAP/ERK kinase (MEK) phosphorylationin a manner dependent on CD44 expression. That A6directly interacts with CD44 was shown by the obser-vation that CD44 is affinity labeled by biotinylated A6and that A6 perturbs the binding of a monoclonal anti-CD44 antibody (DF-1485). A6 significantly inhibited thegeneration of lung foci in an experimental metastasismodel, suggesting that activation of CD44 by agentssuch as A6 may represent a novel strategy for prevent-ing the formation of metastases.

Materials and Methods

A6 and modified A6 peptidesThe capped A6 peptide (acetyl-Lys-Pro-Ser-Ser-Pro-

Pro-Glu-Glu-NH2) was synthesized, purified, and pre-pared as a 100 mg/mL solution in sterile PBS by AltheasTechnologies, Incorporated. A6C (acetyl-Lys-Pro-Ser-Ser-Pro-Pro-Glu-Glu-Cys-NH2) was generated (Bio-Quant,Inc.) to introduce a free sulfhydryl moiety for the purposeof tagging A6 with maleimide-(polyethylene glycol)11-biotin (Pierce Chemical Co.). The product of this coupling(A6C-biotin) was purified via reverse-phase high-perfor-mance liquid chromatography and subjected to massspectrometry to confirm placement of the biotin moietyon the peptide. A6C and A6C-biotin bioactivity was con-firmed in the chemotaxis assay.

Cell cultureAll cell lines were obtained from American Type Cul-

ture Collection with the exception of OVCAR4, OVCAR5,and OVCAR8, which were obtained from the Develop-mental Therapeutics Program of the National CancerInstitute, and A2780, which was obtained from the Eu-ropean Collection of Cell Cultures. Cell lines were cul-tured for less than 6 months before use in this study withthe exception of A2780, whichwas cultured formore than50 passages. All culture reagents were obtained fromCellgro Mediatech, Inc. All media were supplementedwith 10% FBS, 1,000 U/mL penicillin, 100 mg/mL strep-tomycin, and 4 mmol/L L-glutamine. SKOV3 cells were

cultured in McCoy’s 5A medium. OVCAR3, OVCAR4,OVCAR5, OVCAR8, and A2780 cells were cultured inRPMI 1640. The murine melanoma B16-F10 line and allother cells used were maintained in Dulbecco’s ModifiedEagle’s Medium (DMEM), to which 1 mmol/L sodiumpyruvate was included with the above supplements.

In vitro chemotaxis assayChemotaxis of cancer cell lines were done in Boyden

chambers essentially as described (25). Cell migrationwas done through collagen I–coated filters containing8-mm pores toward the conditioned media of growingNIH3T3 fibroblasts (serum-free DMEM), supplementedwith 10 ng/mL vascular endothelial growth factor(CM/V) for 20 hours at 37�C. When assessing the effectof A6, cells were preincubated for 30minutes at 37�Cwith0 to 100 mmol/L A6, before addition to the Boydenchambers. In these instances, peptide was also includedin the lower chamber as well. After migration, the mem-branes were stained with 30% Giemsa and the number ofmigrating cells determined in 4 microscopic fields(�1 mm2/field) at a total magnification of �200. Experi-ments were conducted in quadruplicate and data arepresented as the mean � SD.

Experimental lung metastasis modelThe antimetastatic activity of A6 was tested in the

murine B16-F10 lung metastasis model (26). Treatmentwith A6 (100 mg/kg, s.c. twice daily, n ¼ 15) or DPBS(vehicle, n ¼ 15) was initiated 2 hours postinoculation of106 cells per mouse and continued for 11 days. The micewere then sacrificed and the lungs harvested and stainedwith Bouin’s fixative and staining solution (Sigma Chem-ical Co.). For each mouse, the number of black B16-F10tumor nodules in themedial lobe of the right lung and theleft superior lobe was determined and the data averagedwithin the groups.

Binding and cross-linking of A6C-biotin to SKOV3cells

A6C-biotin (100 mmol/L in cold PBS) was bound toSKOV3 cells and then cross-linked by the addition of 5mmol/L bis(sulfosuccinimidyl) suberate (Pierce Chemi-cal Co.), a homo-, bifunctional cross-linker reagent thatcontains an amine-reactive NHS ester at each end of an11.4 angstrom (8-atom) spacer arm (27). Triton X-100 celllysates were generated, cleared by centrifugation, andthen subjected to immunoblot analyses after transfer topolyvinylidene difluoride (PVDF) membranes. Proteinscross-linked to A6C-biotin were detected by incubatingthe membranes with 10 ng/mL of streptavidin conjuga-ted to horseradish peroxidase (HRP-SA; EMD Biochemi-cals). CD44 was detected on blots by staining with1 mg/mL DF1485 and peroxidase-donkey anti-mouseIgG (Jakson ImmunoResearch). Detection was done withfreshly prepared chemiluminescence reagent (ECL Plus;GE Healthcare) and exposure to Hypermax ECL film (GEHealthcare).

A6 Peptide Activates CD44, Inhibits Migration and Metastasis

www.aacrjournals.org Mol Cancer Ther; 10(11) November 2011 2073




Immunoprecipitation of A6C-biotin–taggedproteins

Triton X-100 lysates, generated from SKOV3 cells thathad been bound and cross-linked to A6C-biotin, wereprecleared by incubating with Protein A–agarose (SigmaChemical Co.) for 2 hours at room temperature. Aliquotsrepresenting 100 mg of the cleared lysates were thenincubated overnight at room temperature with 20 mL ofProtein A–agarose plus 2 mg of anti-CD44 (DF1485; SantaCruz Biotech) or murine IgG1 (Sigma Chemical Co.).After washing 5 times with lysis buffer, the immunopre-cipitated proteins were eluted from the Protein A–aga-rose by boiling in SDS-PAGE sample buffer. The elutedmaterial and 5 mg aliquots of the starting material werethen subjected to SDS-PAGE, transfer to PVDF mem-branes, and blotting with HRP-SA.

Fluorescence-activated cytometryWashed cultured cells were suspended in ice cold,

serum-free McCoy’s 5A media containing 0.5% bovineserum albumin [fluorescence-activated cytometry(FAC) buffer] at 106/mL. Aliquots (0.5 mL) were incu-bated with 1 mg/mL anti-CD44 antibodies DF1485, B-F24, F-4 (Santa Cruz Biotechnology) or IM7 (Biolegend)for 1 hour at 4�C. When assessing the effect of A6 onantibody-binding, cell aliquots were preincubated withA6 for 30 minutes at 4�C before the addition of primaryantibody. After 2 washes, the cells were suspended incold FAC buffer containing 10 mg/mL fluorescein iso-thiocyanate–secondary antibody (Jackson ImmunoRe-search) and incubated for 30 minutes at 4�C. After asingle wash, cell staining was assessed with a BeckmanCoulter Epics XL-MCL cytofluorimeter, collecting10,000 nongated events for each sample using CoulterEpic software for data analysis. The mean fluorescenceintensity of the negative control sample (murine IgG1)for each cell type was adjusted during data collectionto a value lesser than 0.5 before analysis of the othersamples.

Adhesion to hyaluronic acidSKOV3 cells were cultured overnight with titrating

concentrations (0–100 mmol/L). The cells were then de-tached, washed, and fluorescently labeled by incubatingwith 20 mg/mL CFDA-SE (Invitrogen, Inc.) in Hank’sbuffered saline (Cellgro, Inc.) for 30 minutes at 37�C.Afterwashing, the cells were suspended inDPBS contain-ing 10% FBS at 106/mL and incubated for 30 minutes atroom temperature with no additions, 0.001 to 100 mmol/LA6 or with 1 mg/mL IM7, a blocking anti-CD44 ratmonoclonal antibody (BioLegend). The cells were thenplaced into quadruplicate microtiter wells that had beencoated with 50 mL of 1 mg/mLHA (Calbiochem #385902)overnight and blocked for 30 minutes with 10% FBS inDPBS. The cells were allowed to adhere at room temper-ature for 30 minutes, after which the wells were decantedand washed 3 times. The fluorescence in each well wasassessed (excitation/emission: 460/536 nm) with an Fmax

plate fluorimeter (Molecular Device Corp.) and the mean� SD fluorescence values of the quadruplicate determi-nations calculated.

Phospho-FAK and phopsho-MEK immunoblotanalysis

Cells were detached from their culture vessels bybrief trypsinization, washed 3 times in DPBS, and resus-pended at 3.33E5/mL in CM/V. The cells were thentreated with or without 100 mg/mL soluble HA (Cal-biochem #385902) for 30 minutes at 37�C under 5% CO2.Aliquots were then plated into 10-cm culture dishes andA6 added to yield final concentrations of 0, 10, 100, and1000 nmol/L. The cells were cultured for 18 hours at37�C under 5% CO2.

After 18 hours of treatment, Triton X-100 lysates weregenerated in cold radioimmunoprecipitation assay(RIPA) buffer (Thermo Scientific #89901) containing cock-tails of broad spectrum phosphatase inhibitors (Calbio-chem #524625) and protease inhibitors (Sigma #P8340).For each test condition, the lysates of the nonadherent andadherent populations were generated and combined.Samples representing 20 mg of total protein of each com-bined lysate were subjected to immunoblot analyses,with rabbit polyclonal antibodies specific for FAK,FAK-pY397, FAK-pY576/577, FAK-pY925, MEK-pS217/221, or a monoclonal anti-MEK (41GP; Cell SignalingTechnology).

DensitometryThe integrated densities of the bands of interest were

measured using ImageJ software (NIH). The densities ofthe stained bands of each RIPA lysate were normalizedto the integrated density obtained for the mouse–anti-glyceraldehyde-3-phosphate dehydrogenase signal of thelysate. For each antigen and lysate, the density obtainedfor the anti-phospho–antigen stained band was normal-ized to that of the anti-panantigen stained band.

Results

A6 inhibits in vitro chemotaxis and experimentalmetastasis

A6 perturbed the chemotaxis of SKOV3 ovarian cancercells towardNIH3T3 conditionedmedium supplementedwith VEGF (CM/V). A6 blocked the migration of SKOV3cells in a concentration-dependent manner. In the repre-sentative experiment shown in Fig. 1A, the IC50 was 92.4nmol/L. The mean IC50 of 3 experiments was 40.0 � 26.7nmol/L (SEM). To assess the breadth of its antimigratoryactivity, the ability of A6 to inhibit chemotaxis was testedagainst apanelof additional cell lines.As showninTable1,A6waseffectiveat blocking themigrationof theOVCAR8,OVCAR3, ES2, IGROV-1, MDA-MB-468, and MDA-MB-361 cells with IC50 values in the range of 10 to 100 nmol/L.The MDA-MB-231 breast cancer line exhibited an inter-mediate IC50 value of 288 nmol/L. A6was less effective atinhibiting migration of MDA-MB-435 and CaOV3 cells,

Piotrowicz et al.

Mol Cancer Ther; 10(11) November 2011 Molecular Cancer Therapeutics2074




exhibiting IC50 values in themmol/L range. For 3 cell lines,OVCAR4, OVCAR5, and a late passage population ofA2780, migration was not inhibited by A6 at any concen-tration tested. Thus, although A6 was very potent against

some cell lines, it lacked any activity against others,suggesting a mechanism dependent upon expression ofa specific receptor or signal transductionpathway.A6wasnot cytotoxic at any concentration used in this study (datanot shown).

To determine whether A6 inhibits the colonization ofsecondary tissues by circulating cancer cells, its activitywas tested in the well-characterized B16-F10 lung met-astatic model. In vitro, A6 was as effective in blockingthe chemotactic migration of B16-F10 cells as it was forSKOV3 cells. In the representative experiment whoseresults are shown in Fig. 1B, the IC50 was 28.8 nmol/L;the mean IC50 of 3 experiments was 21.3 � 7.6 (SEM)nmol/L. In the in vivo model C57Bl/6 mice wereinjected intravenously with 106 cells and then treatedwith either vehicle alone or A6 at a dose of 100 mg/kgs.c. twice daily for 11 days before sacrifice and enumer-ation of the number of nodules in the lungs. As shown inFigs. 1C and D, treatment with A6 reduced the numberof lung nodules. In the representative experimentshown in Fig. 1C, A6 reduced the number of lungmetastases to 50% of control. In 3 independent experi-ments, the number was decreased to just 49.4 � 16.0(SEM)% of those in the control mice (P ¼ 0.0029 in anunpaired t test). Body weights of both groups werestable throughout the short study, and no gross changesother than those associated with metastatic lung diseasewere noted at postmortem. Thus, A6 exhibited substan-tial antimetastatic activity in this aggressive in vivomodel.

Figure 1. A6 inhibits the migrationof SKOV3 and B16-F10 cells invitro and the colonization of lungsby B16-F10 cells in vivo. SKOV3(A) and B16F10 (B) cells werepreincubated with A6 and thenallowed to migrate in the presenceof the indicated concentrations ofA6 in response toward CM/V.C57Bl/6 mice were inoculatedwith B16-F10 cells and thentreated with vehicle or A6(100 mg/kg, twice a day, s.c.) for11 days. The lungs of 4representative mice of each group(D) and the mean number ofmetastases/mouse determinedfrom the total tally of all theanimals in the study (C) arepresented.

1405055

IC50 = 92.4 nmol/L IC50 = 28.8 nmol/L

BA

60

80

100

120

202530354045

0

20

40

Log A6 (mol/L)Log A6 (mol/L)

No.

mig

rating c

ells

–10 –9 –8 –7 –6 –5–4 –3 –10 –9 –8 –7 –6 –5–4 –305

1015

Vehicle A6-Treated

No.

mig

rating c

ells

125

150

175DC

50

75

100

B1

6-F

10

me

tasta

se

s

Vehicle A6 Peptide0

25

Table 1. Inhibition of the chemotaxis of themurine melanoma B16-F10 line and of selectedhuman ovarian and breast cancer lines by A6

Cell line IC50 (nmol/L) Classification

B16-F10 21.3 � 7.57 (SEM) ResponsiveSKOV3 40.0 � 26.7 (SEM) ResponsiveOVCAR8 15.7 ResponsiveOVCAR3 33.9 ResponsiveES2 11.0 ResponsiveIGROV-1 21.5 ResponsiveMDA-MB-468 62.3 ResponsiveMDA-MB-361 111 ResponsiveMDA-MB-231 288 Moderately

responsiveMDA-MB-435 1,480 Weakly

responsiveCaOV3 12,700 Weakly

responsiveOVCAR5 >100,000 NonresponsiveOVCAR4 >100,000 NonresponsiveA2780 (late

passage, CD44-)>100,000 Nonresponsive






The inhibition of ovarian cancer cell migration byA6 correlates with CD44 expression

The panel of ovarian cancer cell lines was screened forassociation of the expression of cell surface moleculesknown to be required for migration and sensitivity to theantimigratory effect of A6. As shown in Fig. 2, a corre-lation was found between the expression of CD44 andsensitivity to A6, as tested by flow cytometric analysis

with 4 different anti-CD44 antibodies. All 4 antibodieswere made against intact cells or partially purified pro-tein. The epitopes towhich these antibodies bind have notbeen mapped, with the exception of IM7. A6 greatlyinhibited the migration of SKOV3 and OVCAR3 cellsbut failed to inhibit OVCAR4 and late passage A2780cells. A6 slightly inhibited the migration of OVCAR5cells; however, the effect was not clearly concentration

120140160180A

200

406080

100

No.

mig

ratin

g ce

lls

0 µmol/L1 µmol/L100 µmol/L

4

6

8

10

12

MnI

MnI

MnI

MnI

SKOV3OVCAR3OVCAR4OVCAR5A2780

SKOV3 OVCAR3 OVCAR4 OVCAR5 A2780

0

2

DF1485

A2780

SKOV3OVCAR3OVCAR4OVCAR520

253035404550

SKOV3OVCAR3

OVCAR5A2780

05

101520

B-F24

2.53

3.54

4.5

OVCAR4OVCAR5A2780

708090

100

SKOV3

00.5

11.5

2

F-4

010203040506070

IM7

OVCAR3OVCAR4OVCAR5A2780

B

00

00

1836

5474

1428

4256

1938

5776

–1 1 10 100 1,000

–1 1 10 100 1,000

–1 1 10 100 1,000

–1 1 10 100 1,000

1632

4864

Single parameter

Single parameter

Single parameter

Single parameter

DF1485

B-F24

F-4

IM7

Figure 2. A6 inhibition of ovariancancer cell migration correlateswith CD44 expression. A, theeffect of 1 and 100 mmol/L A6 onthe migration of SKOV3, OVCAR3,OVCAR4, OVCAR5, and A2780cells. B, FAC analysis of thebinding of anti-CD44 antibodiesDF1485, B-F24, F-4, and IM7 tocell lines tested in A. For eachantibody, overlays of the individualhistograms are presented on theleft and graphical representationsof the mean fluorescence intensity(MnI) values are presented on theright.

Piotrowicz et al.





dependent over the range tested. Flow cytometric analy-ses (Fig. 2B) showed the same relative pattern in CD44expression levels for each of the anti-CD44 antibodiestested: SKOV3>OVCAR3>OVCAR4, OVCAR5>A2780.Thus, the 2 cell lines most affected by A6 in the migrationassay (SKOV3andOVCAR3) expressed the greatest levelsof CD44. Conversely, the lines expressing the lowest levelof CD44, A2780, and OVCAR4 were least affected.

A6C-biotin affinity labels CD44The correlation between sensitivity to the antimigra-

tory activity of A6 and expression of CD44 suggested thatA6 may directly interact with CD44. To explore thispossibility, SKOV3 cells were bound and cross-linkedto biotinylated A6 peptide (A6C-biotin). Lysates pre-pared from the cross-linked cells were subjected to im-munoprecipitation using an anti-CD44 monoclonalantibody (DF1485; Santa Cruz Biotech.) or its isotypecontrol. The eluted proteins, as well as starting lysates,were subjected to PAGE/blotting and then stained for thepresence of the biotin tag and CD44withHRP-SA and theDF1485 antibody, respectively.

As shown in Fig. 3A, A6C-biotin labeled a number ofcellular proteins in the starting material (first lane ofthe left panel). HRP-SA stained only 3 bands in thelysates of cells that had been cross-linked in the ab-sence of A6C-biotin, indicating that, with these excep-tions, HRP-SA staining of the A6C-biotin-labeledproteins in the samples cross-linked in the presenceof A6C-biotin was specific. DF1485 specifically precip-itated a large band representing protein of approxi-mately 85 kDa that had been tagged with A6C-biotin.This band was stained to a significantly greater extentin the precipitate compared with the starting material,indicating that this protein was greatly enriched by theanti-CD44 immunoprecipitation. DF1485 stained aprotein with this molecular weight in the startingmaterial and precipitates generated from both theA6C-biotin-labeled and control cells, confirming theidentity of CD44 on these blots and the presence ofCD44 in all samples. These results indicate that A6interacts with CD44 in a configuration that is intimateand stable enough to undergo cross-linking with bis(sulfosuccinimidyl) suberate.

Figure 3. A6C-biotin affinity labelsSKOV3 CD44 and A6 inhibits thebinding of DF1485 anti-CD44 toSKOV3 cells. A, SKOV3 cells wereincubated in the presence (þ) orabsence (�) of A6C-biotin andthen with the cross-linking agent.Lysates were generated and thensubjected to immunoprecipitationwith anti-CD44 antibody (DF1485)or its isotype control. Theprecipitated proteins, as well asthe starting material (St. Mat.)were then subjected to Westernblot analysis, staining with HRP-SA to detect A6C-biotin–taggedproteins, and DF1485 to detectCD44. B, FAC analysis of thebinding of DF1485 1 mg/mL toSKOV3 cells in the presence orabsence of 10 mmol/L A6. IP,immunoprecipitation.

A

B

250

148

98

64

5095

6923

046

A6C-biotin:IP: St. Mat.

Blotting: HRP-SA

Single parameter

+A6

–1 1 10 100 1,000

–A6

DF1485

+ –St. Mat.+ –

DF1485 Isotype Con.+ – + –

DF1485 Isotype Con.+ – + –






A6 perturbs the binding of DF148 anti-CD44monoclonal antibody

Further evidence for an interaction between A6 andCD44 was provided by flow cytometric analysis of theability of A6 to interfere with the binding of the anti-CD44DF1485 antibody to SKOV3 cells. As shown in Fig. 3B,treatment of the SKOV3 cells with 10 mmol/L A6 for 30minutes before incubationwith 1 mg/mLDF1485 reducedthe mean fluorescence intensity obtained. Maximal inhi-bition was observed with 10 mmol/L A6 which reducedthe MFI by 38.4� 15% (SEM, n¼ 4). A6 treatment did notsignificantly reduce the binding of any of 4 other anti-CD44 antibodies tested, including the ligand blockingIM7 rat monoclonal antibody (data not shown). Thisresult indicates that A6 does not nonspecifically changethe entire extracellular structure of CD44 but ratherproduces a localized change in a single epitope ofCD44. Unfortunately, the epitope to which DF1485 bindshas not been mapped; nevertheless, the result is consis-tent with the idea that A6 induces a relatively subtlechange in the structure of one part of the CD44 receptor.

A6 potentiates SKOV3 adhesion to a hyaluronicacid–coated surface

HA is a principal ligand for CD44. The ability of A6 toaffect this interaction was investigated with an assaymeasuring the adhesion of fluorescently labeled cells toHA-coated microtiter wells. In this assay, approximately10% of the 50,000 SKOV3 cells added to each HA-coatedwell adhered after 30 minutes. Inclusion of a blockingmonoclonal antibody (IM7) at a concentration of 1 mg/mLinhibited adhesion of SKOV cells to HA by 61.3 � 10.2%(SEM, n ¼ 4), showing the degree of specificity of thisassay for CD44 activity.

Figure 4A shows that A6 produced a concentration-dependent increase in the number of SKOV3 cells adher-ing to HA-coated plates. The EC50 was determined bynonlinear regression analysis to be 1.08 mmol/L. EC50

values determined from a series of 5 experiments inwhich adhesion times (15 or 30 minutes), the numberof cells added to each well (50,000 to 100,000), andwashing techniques were altered, ranged from 0.4 to5.03 mmol/L.

The ability of A6 to augment adhesion to HA of 3 celllines expressing different levels of CD44 was assessed(Fig. 4B). Fewer than 10% of the A2780 cells, which do notexpress CD44, adhered to HA-coated plates. This lowlevel adhesion was not blocked by IM7, indicating that itwas not CD44 dependent. As expected, A6 did notenhance the adhesion of A2780 cells. In contrast, theadhesion of SKOV3 and OVCAR3 cells, which expressedrelatively greater and lesser amounts of CD44, respec-tively, was blocked by IM7 indicating CD44 dependence.Importantly, treatment with 10 mmol/L A6 augmentedthe adhesion of the CD44-positive SKOV3 and OVCAR3cells but did not enhance the CD44-independent adhe-sion of the A2780 cells. These results indicate that A6 notonly binds to CD44, it also alters at least one of thefunctions of CD44.

A6-treatment induces FAK and MEKphosphorylation in CD44-positive lines and isantagonized by soluble HA

To identify signaling events induced or modulated byA6, a survey was made of the expression and phosphor-ylation of selected signaling proteins in SKOV3, B16-F10,and A2780 cells cultured in the presence of chemotacticstimuli and the presence or absence of A6 and soluble

90EC

50 = 1.08 µmol/LA B

200

50

60

70

80

100

50

150

–12 –10 –8 –6 –4 –220

30

40

log [A6]

Ad

h. ce

lls (f

l u

nits)

0

10 µmol/L 10 µmol/LNo add.

A6

SKOV3 OVCAR3 A2780

10 µmol/LNo add.IM7

A6

No add.IM7

A6

IM7

HA

ad

h. %

co

n.)

Figure 4. A6 enhances the adhesion to HA of CD44-expressing SKOV3 and OVCAR3 cells but not CD44-nonexpressing A2780 cells. The effect oftitrating concentrations of A6 on the adhesion of fluorescently loaded SKOV3 cells to HA is presented in A. Adhesion is presented as the mean fluorescenceof quadruplicate wells. The effects of preincubating and including 10 mmol/L A6 on the adhesion of 2 CD44-positive cell lines (SKOV3 and OVCAR3)and a CD44-negative line (A2780) are presented in B. Inclusion of the blocking IM7 antibody shows CD44-dependence of the adhesion. Data wasnormalized to the adhesion of the control samples receiving no additions.

Piotrowicz et al.





HA. The proteins surveyed included FAK, Rho, thep85 subunit of phosphoinositide-3-kinase (PI3K), proteinkinase Ca (PKCa), Akt, phospholipase D (PLD), Erk1/2,and MEK. The most consistent and striking resultsobtained from this survey were that A6 enhanced phos-phorylation of tyrosine residues Y397, 576/577, and Y925of FAK and serine residues S217/S221 of MEK in the A6-responsive SKOV3 and B16F10 lines, but not in thenonresponsive A2780 line (Fig. 5). Densitometric analyses

of the blots presented in Fig. 5A showed 30-, 6.7-, and 7-fold increases in the integrated densities of the FAKpY397, pY576/577, and pY925 signals, respectively, whenSKOV3 cells were cultured in the presence of 100 nmol/LA6. The signals obtained with the anti-pY576/577 andpY925 antibodies for the B16-F10 cells cultured in thepresence of A6 were approximately 2.2-fold greater thanthat of those obtained when the cells were cultured in itsabsence. Interestingly, the pY397 signal obtained with the

Figure 5. A6 enhances pFAK andpMEK levels in CD44-expressingSKOV3 and B16-F10 cells but notCD44-nonexpressing A2780 cellsand is antagonized by solubleHA. The effects of including100 nmol/L A6 on pFAK (pY397,pY576/577, and pY925) or pMEK(pS217/221) in the culture ofSKOV3, B16-F10, and A2780 cellsin chemotactic media (CM/V) arepresented in A and B, respectively.C, pFAK (pY397, pY576/577, andpY925) levels after culture ofSKOV3 cells with 0, 10, 100, and1,000 nmol/L A6 in CM/V in thepresence or absence of 100 mg/mL soluble HA. D, densitometricanalysis of the blots presentedin C.

A

SKOV3

pFAK

FAK

–HA

FAK

pFAK (Y397)

20

15

10

5

0

–HA

+HA

0 10 100 1,000 0 10 100 1,000 0 10 100 1,000 nmol/L A6

pFA

K/F

AK

pFAK (Y579/577)

pFAK (Y525)

pFAK Y397 pFAK Y576/577 pFAK Y925

GAPDH

nmol/L A6: 0 10 1001,000 0 10 1001,000

+HA

– + – + – + – +

– + – +

pY397 pY576/577 pY925 MEKpMEK

pS217/221

B16-F10

A2780

100 nmol/L A6:

SKOV3

B16-F10

A2780

100 nmol/L A6:

B

C

D






B16-F10 cells did not respond to A6. The percent of totalFAK that became phosphorylated could not be deter-mined as the pan anti-FAK antibody failed to detect thephosphorylated forms. Treatment with A6 did not in-crease the FAK pY397, pY576/577, and pY925 signals ofthe non-CD44–expressing A2780 cells. In fact, a decreasein the pY925 signal was observed in the experimentpresented in Fig. 5A. A6 treatment of SKOV3 and B16-F10 cells, but not A2780 cells, enhanced MEK phosphor-ylation (Fig. 5B). Densitometric analyses showed a 2.1- to2.5-fold increase in the integrated densities of the pS217/221 signal generated for the CD44-positive cell lines,when cultured in the presence of 100 nmol/L A6. Theseresults indicate that in the cells that expressed CD44, butnot in those that did not, A6 activated elements of a signaltransduction cascade known to be linked to CD44.

The effect of soluble HA on the ability of A6 to enhanceFAK phosphorylation in SKOV3 cells was examined.SKOV3 cells were incubated with or without 100 mg/mLsoluble HA for 30 minutes before being cultured inCM/V in the presence of 0, 10, 100, or 1,000 nmol/L A6.Lysates were prepared after 18 hours of culture and sub-jected to Western blot analysis for the levels of FAK, FAKpY397, pY576/577, and pY925. As shown in Fig. 5C andD,pretreatmentwith solubleHAand its inclusion in the assayantagonized the A6-induced phosphorylation of FAK. Aclear dose-dependent induction of the pY397, pY56/577,andpY925 signalswas observed in the absence ofHA,withthe response to 100 nmol/LA6 being similar inmagnitudeto those generated in the experiment presented in Fig. 5A.The response was biphasic; the signals induced by culturewith 1 mmol/L A6 were reduced compared with thoseinduced with 100 nmol/L A6. For all pY antigens and allconcentrations of A6 tested, the inclusion of soluble HAreduced the anti-pY signal intensities. Although the natureof this antagonism is unclear, the demonstration that aCD44 ligand modulates the signaling response to A6 pro-vides additional evidence for a functional relationshipbetween A6 and CD44.

Discussion

In this study, the A6 peptide was found to inhibit thechemotaxis of a subset of the ovarian and breast cancercell lines (Table 1). The antimigratory activity of A6correlated with CD44 expression (Fig. 2), suggesting thatA6 may modulate CD44-dependent activities leading tothe inhibition of cancer cell motility. That A6 directlyinteracts with CD44 is indicated by the demonstrationthat A6C-biotin affinity labeled SKOV3 CD44 and per-turbed the binding of DF1485, a murine anti-CD44 mono-clonal antibody, to intact SKOV3 cells (Fig. 3). This partialinhibition does not seem to be the result of a competitionof A6 and CD44 for the antibody; the antibody failed torecognize the peptide or inhibit the binding of an anti-A6antibody in an ELISA assay when the peptide wasabsorbed onto microtiter plates (data not shown). There-fore, A6 may induce conformational or other changes in

the receptor that either lower the affinity of the epitopefor the antibody or exclude the antibody from binding.Notably, A6 did not affect the binding of other anti-CD44antibodies (i.e., B-F24, F-4, and IM7 antibodies; data notshown), showing that A6 treatment did not reduce CD44levels over the time course of the staining required for theassay and suggesting the A6-induced changes in CD44are localized. Importantly, the changes induced in CD44by A6 represent an activation of the receptor: A6 poten-tiated cell adhesion to HA, a principal CD44 ligand(Fig. 4). This ability of A6, like its antichemotactic activity,was restricted to CD44-positive cells.

The most striking results obtained from the initialsurvey of the effects of A6 on the expression and phos-phorylation of cancer cell signaling proteins was thedemonstration that A6 elevates pFAK (pY397, pY576/577, and pY925) levels in the A6-responsive SKOV3 andB16F10 lines but failed to do so in the nonresponsiveA2780 line. FAK has been implicated in diverse cellularroles including cell locomotion, mitogen response, andcell survival, primarily through its role in modulating thecytoskeleton (reviewed in Schaller and colleagues; ref.28). FAK associates with the cytoplasmic domains ofintegrin adhesion receptors in focal adhesions and isactivated by autophosphorylation of Y397 in responseto integrin clustering (29, 30) This site serves a bindingsite for members of the Src family of tyrosine kinaseswhich phosphorylate additional FAK tyrosine residues,including Y576/77 and Y925 (29, 30). Src binds with highaffinity to a single site on the cytoplasmic domain ofCD44 and Src activity is stimulated as a result of the HA-CD44 interaction (31, 32). Phospho-Y925 serves as a dock-ing site for Grb and triggers Ras-dependent activation ofthe MAP kinase cascade, including MEK phosphoryla-tion (33), as was observed in this study (Fig. 5B). Thus,A6-induced activation of CD44 seems to initiate signalingthrough CD44-mediated Src activation of FAK. Impor-tantly, inclusion of soluble HA in the culture of SKOV3cells with A6 abrogated the A6-induced FAK phosphor-ylation (Fig. 5, panels C and D). The precise mechanismfor this antagonism remains to be determined.

The stimulatory effect of A6 on the adhesion of SKOV3cells exhibited an IC50 of 1 mmol/L, 2 orders of magnitudegreater than that observed for its inhibitory effect on cellmigration. The reason for this discrepancy is unclear,given that both phenomena seem to correlate withCD44 expression. It is possible that the discrepancy isdue to A6 absorbing onto the HA-coated plates, thuslowering the effective concentration of soluble A6 in theadhesion assay. Alternatively, the 2 phenomena mayhave different thresholds as to the number of activatedCD44molecules required to initiate ameasurable effect. Itshould also be noted that it is likely that the cellularevents taking place in each assay may render CD44conformationally different, thus presenting different af-finities for A6. Lastly, it can not be excluded that A6 mayexert an effect on cell adhesion to HA through a CD44-independent mechanism and that the positively effected

Piotrowicz et al.





cell lines tested to date surreptitiously express this re-ceptor along with CD44 and that the negative A2780 linelacks both receptors.It is likely that the sequence within the ligand-binding

domain of CD44 sharing homology with A6 (CD44: 120-NASAPPEE-127) will be central to the mechanism bywhich A6 activates the adhesive activity of CD44. A6maymimic theCD44 sequence and engage in a homotypicinteraction that induces conformational changes in CD44and/or CD44 dimerization leading to greater affinityand/or avidity for ligand.Asmentioned, the perturbationof DF1485 binding by A6 does indeed suggest that A6 isinducing conformational changes in the receptor. Alter-natively, the ability of A6 to mimic the CD44 sequencemay permit it to interact with a CD44-binding partnerresulting in release of repression of CD44 activity. Lastly,the possibility that A6 interacts with another membraneprotein independent of CD44 and induces signals thatsecondarily activate CD44 cannot be excluded at this time.From a clinical point of view, the most important

observation to emerge from these studies is that A6

inhibits the formation of metastases in vivo. BecauseCD44 is a marker of cancer stem cells and has beenlinked to the expression of drug-resistant phenotypes(reviewed in Toole and colleagues; ref. 34), an intrigu-ing issue currently being addressed is whether A6modulates drug resistance. The ability of A6 to activateCD44 suggests that the peptide may modulate thiscritical determinant of therapeutic efficacy as well.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Grant Support

The work was support by A�ngstrom Pharmaceuticals, Inc., Solana Beach, CA.

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received May 20, 2011; revised August 25, 2011; accepted August 25,2011; published OnlineFirst September 1, 2011.

References1. Guo Y, Mazar AP, Lebrun JJ, Rabbani SA. An antiangiogenic uroki-

nase-derived peptide combined with tamoxifen decreases tumorgrowth and metastasis in a syngeneic model of breast cancer. CancerRes 2002;62:4678–84.

2. Boyd DD, Kim SJ, Wang H, Jones TR, Gallick GE. A urokinase-derivedpeptide (A6) increases survival of mice bearing orthotopically grownprostate cancer and reduces lymph node metastasis. Am J Pathol2003;162:619–26.

3. Mishima K, Mazar AP, Gown A, Skelly M, Ji XD, Wang XD, et al. Apeptide derived from the non-receptor-binding region of urokinaseplasminogen activator inhibits glioblastoma growth and angiogenesisin vivo in combination with cisplatin. Proc Natl Acad Sci USA2000;97:8484–9.

4. van Troostenburg AR, Lee D, Jones TR, Dyck-Jones JA, SilvermanMH, Lam GN, et al. Safety, tolerability and pharmacokinetics ofsubcutaneous A6, an 8-amino acid peptide with anti-angiogenicproperties, in healthy men. Int J Clin Pharmacol Ther 2004;42:253–9.

5. Berkenblit A, Matulonis UA, Kroener JF, Dezube BJ, Lam GN, CuasayLC, et al. A6, a urokinase plasminogen activator (uPA)-derived peptidein patients with advanced gynecologic cancer: a phase I trial. GynecolOncol 2005;99:50–7.

6. Ghamande SA, Silverman MH, Huh W, Behbakht K, Ball G, CuasayLC, et al. A phase 2, randomized, double-blind, placebo-controlledtrial of clinical activity and safety of subcutaneous A6 in women withasymptomatic CA125 progression after first-line chemotherapy ofepithelial ovarian cancer. Gynecol Oncol 2008;111:89–94.

7. Goodison S, Urquidi V, Tarin D. CD44 cell adhesion molecules. J ClinPathol 1999;52:189–96.

8. Teriete P, Banerji S, Noble M, Blundell CD, Wright AJ, Pickford AR,et al. Structure of the regulatory hyaluronan binding domain in theinflammatory leukocyte homing receptor CD44. Structure of theregulatory hyaluronan binding domain in the inflammatory leukocytehoming receptor CD44. 2004;13:483–96.

9. Goodfellow PN, BantingG,WilesMV, Tunacliffe A, Parker M, SolomonE, et al. The gene, MIC4, which controls expression of the antigendefined by monoclonal antibody F10.44.2, is on human chromosome11. Eur J Immunol 1982;12:659–63.

10. Ue T, Yokozaki H, Kiadai Y, Yamamoto S, Yasui W, Ishikawa T, et al.Co-expression of osteopontin and CD44v9 in gastric cancer. Int JCancer 1998;79:127–32.

11. Weber GF, Ashkar S, Cantor H. Interaction between CD44 andosteopontin as a potential basis for metastasis formation. Proc AssocAm Physicians 1997;109:1–9.

12. Jalkanen S, Jalkanen M. Lymphocyte CD44 binds the COOH-terminalheparin-binding domain of fibronectin. J Cell Biol 1992;116:817–25.

13. Faassen AE, Schrager JA, Klein DJ, Oegema TR, Couchman JR,McCarthy JB. A cell surface chondroitin sulfate proteoglycan,immunologically related to CD44, is involved in type I collagen-mediated melanoma cell motility and invasion. J Cell Biol 1992;116:521–31.

14. Bourguignon LY, Zhu H, Chu A, Iida N, Zhang L, Hung MC. Interactionbetween the adhesion receptor, CD44, and the oncogene product,p185HER2, promotes human ovarian tumor cell activation. J BiolChem 1997;272:27913–8.

15. Yang B, Yang BL, Savani RC, Turley EA. Identification of a commonhyaluronan binding motif in the hyaluronan binding proteins RHAMM,CD44 and link protein. EMBO J 1994;15:286–96.

16. Liao HX, Lee DM, Levesque MC, Haynes BF. N-terminal and centralregions of the human CD44 extracellular domain participate in cellsurface hyaluronan binding. J Immunol 1995;155:3938–45.

17. Bajorath J, Greenfield B, Munro SB, Day AJ, Aruffo A. Identification ofCD44 residues important for hyaluronan binding and delineation of thebinding site. J Biol Chem 1998;273:338–43.

18. Peach RJ, Hollenbaugh D, Stamenkovic I, Aruffo A. Identification ofhyaluronic acid binding sites in the extracellular domain of CD44. JCell Biol 1993;122:257–64.

19. Lesley J, English N, Perschi A, Gregoroff J, Hyman R. Variant cell linesselected for alterations in the function of the hyaluronan receptorCD44 show differences in glycosylation. J Exp Med 1995;182:431–7.

20. Lesley J, Kincade PW, Hyman R. Antibody-induced activation of thehyaluronan receptor function of CD44 requires multivalent binding byantibody. Eur J Immunol 1993;23:1902–9.

21. Lesley J, Hyman R, Kincade PW. CD44 and its interaction withextracellular matrix. Adv Immunol 1993;54:271–335.

22. Liu D, Liu T, Sy MS. Identification of two regions in the cytoplasmicdomain of CD44 through which PMA, calcium, and foskolin differen-tially regulate the binding of CD44 to hyaluronic acid. Cell Immunol1998;190:132–40.

23. Ponta H, Wainwright D, Herrlich P. The CD44 protein family. Int JBiochem Cell Biol 1998;30:299–305.






24. Banerji S, Wright AJ, Noble M, Mahoney DJ, Campbell ID, Day AJ,et al. Structures of the Cd44-hyaluronan complex provide insight intoa fundamental carbohydrate-protein interaction. Nat Struct Mol Biol2007;14:234–9.

25. Doerr ME, Jones JI. The roles of integrins and extracellular matrixproteins in the insulin-like growth factor I-stimulated chemotaxis ofhuman breast cancer cells. J Biol Chem 1996;271:2443–7.

26. Parhar RS, Lala PK. Amelioration of B16F10 melanoma lung metas-tasis in mice by a combination therapy with indomethacin and inter-leukin 2. J Exp Med 1987;165:14–28.

27. Mattson G, Conklin E, Desai S, Nielander G, Savage MD, MorgensenS. A practical approach to crosslinking. Mol Biol Rep 1993;17:167–83.

28. Schaller MD. Cellular functions of FAK kinases: insight into molecularmechanisms and novel functions. J Cell Sci 2010;123:1007–13.

29. SchallerMD,HildebrandJD,ShannonJD,FoxJW,VinesRR,ParsonsJT.Autophosphorylation of the focal adhesion kinase, pp125FAK, directsSH2-dependent binding of pp60src. Mol Cell Biol 1994;14:1680–8.

30. Cobb BS, Schaller MD, Leu TH, Parsons JT. Stable association ofpp60src and pp59fyn with the focal adhesion-associated proteintyrosine kinase, pp125FAK. Mol Cell Biol 1994;14:147–55.

31. Bourguignon LY, Zhu H, Shao L, Chen Y-W. CD44 interaction with c-Src kinase promotes cortactin-mediated cytoskeleton function andhyaluronic acid-dependent ovarian tumor cell migration. J Biol Chem2001;276:7327–36.

32. Bourguignon LY, Zhu H, Shao L, Zhu D, Chen Y-W. Rho-kinase (ROK)promotes CD44v(3,8-10)-ankyrin interaction and tumor cell migrationin metastatic breast cancer cells. Cell Motil Cytoskeleton 1999;43:269–87.

33. Schlaepferr DD, Hanks SK, Hunter T, van der Gear P. Integrin-medi-ated signal transduction linked to Ras pathway by GRB2 binding tofocal adhesion kinase. Nature 1994;372:786–91.

34. Toole BP, Siomiany MG. Hyaluronan: a constitutive regulator ofchemoresistance and malignancy in cancer cells. Semin Cancer Biol2008;18:244–50.

Piotrowicz et al.





2011;10:2072-2082. Published OnlineFirst September 1, 2011.Mol Cancer Ther Randolph S. Piotrowicz, Bassam B. Damaj, Mohamed Hachicha, et al. CD44-Expressing CellsMEK Phosphorylation, and Inhibits the Migration and Metastasis of A6 Peptide Activates CD44 Adhesive Activity, Induces FAK and

Updated version

10.1158/1535-7163.MCT-11-0351doi:

Access the most recent version of this article at:

Cited articles

http://mct.aacrjournals.org/content/10/11/2072.full#ref-list-1

This article cites 33 articles, 15 of which you can access for free at:

Citing articles

http://mct.aacrjournals.org/content/10/11/2072.full#related-urls

This article has been cited by 2 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://mct.aacrjournals.org/content/10/11/2072To request permission to re-use all or part of this article, use this link



http://mct.aacrjournals.org/lookup/doi/10.1158/1535-7163.MCT-11-0351

http://mct.aacrjournals.org/content/10/11/2072.full#ref-list-1

http://mct.aacrjournals.org/content/10/11/2072.full#related-urls

http://mct.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://mct.aacrjournals.org/content/10/11/2072


a6 peptide activates cd44 adhesive activity, induces fak ...a6 shares sequence homology with cd44,...

Documents