Defective actin polymerization in EBV-transformed B-cell lines from patients with the Wiskott-Aldrich syndrome

, . 185: 99–107 (1998)

DEFECTIVE ACTIN POLYMERIZATION INEBV-TRANSFORMED B-CELL LINES FROM PATIENTS

WITH THE WISKOTT–ALDRICH SYNDROME

1*, 1, 1, . 2, 2, 2, 3, . 4 . 4

1Department of Pathology, University of Brescia, Italy2Department of Paediatrics, University of Brescia, Italy

3Department of Paediatrics, University of Göteborg, Sweden4Immunopathology Section, Metabolism Branch, National Cancer Institute, NIH, Bethesda, Maryland, U.S.A.

SUMMARY

The Wiskott–Aldrich syndrome (WAS) is a rare X-linked recessive disorder characterized by eczema, thrombocytopenia, andimmunodeficiency. An allelic variant of the disease is characterized by isolated thrombocytopenia (XLT). The gene responsible forWAS/XLT (WASP) encodes for a 502 amino acid protein (WASP) that is possibly involved in actin binding and cytoskeletonorganization. The expression of WASP and the distribution of F-actin and alpha-actinin (which binds to and stabilizes actin filaments)have been analysed in lymphoblastoid cell lines from six patients with WAS and one with XLT. Western blot and immunocytochemistrydid not reveal WASP expression in four WAS patients, whereas two WAS patients (with a moderate clinical course) expressedtrace amounts of mutant WASP. In contrast, the XLT patient expressed normal amounts of WASP. Furthermore, cell lines fromWAS and XLT patients also markedly differed in F-actin polymerization and alpha-actinin distribution. In particular, severe defectsof cytoplasmic F-actin expression and of F-actin-positive microvillus formation, and impaired capping of alpha-actinin, wereobserved in all patients who lacked WASP. As a whole, the degree of impairment of WASP protein expression in WAS/XLT seems tocorrelate with anomalies of cytoskeletal organization, strongly supporting a role for WASP in the regulation of F-actin polymerization.? 1998 John Wiley & Sons, Ltd.

J. Pathol. 185: 99–107, 1998.

KEY WORDS—Wiskott–Aldrich syndrome;

INTRODUCTION

The Wiskott–Aldrich syndrome (WAS) is a rareX-linked recessive disorder, with a heterogeneousclinical presentation.1 The classical triad represented byeczema, thrombocytopenia, and immunodeficiency2 isobserved in only one-third of patients;3 attenuatedforms of the disease have been also described, includ-ing isolated X-linked thrombocytopenia (XLT).4 Theimmunological abnormalities in WAS include progres-sive decline of T-cell number and function5 and B-celldefects, with inability to mount a response to poly-saccharide antigens6 and impaired antibody productionto T-dependent antigens (e.g., bacteriophage ö#174).5

In 1994, the gene responsible for WAS was cloned andnamed WASP (for Wiskott–Aldrich syndrome protein).7Mutations of the WASP gene8 have been shown toaccount for both full-blown WAS and XLT, indicatingthat in fact these are allelic disorders.9 Recently, it hasbeen shown that a correlation exists between theabsence/defects of WASP mRNA and protein in B-celllines and the severity of the disease.10

The pathophysiology of WAS/XLT has longremained obscure. The occurrence of small platelets11

and abnormal lymphocytes12 with short, rare, and bluntmicrovilli, typically observed in WAS, has suggestedimpaired organization of the cytoskeleton.13 Further-more, defective signalling in B-14 and T-lymphocytes15

following triggering of surface receptors has also beenreported. This array of cellular defects has been furtherevaluated following characterization of the WASP pro-tein, encoded by the WASP gene. The WASP protein iscomposed of 502 amino acids16 and contains multipledomains that enable interaction with different proteins.In particular, through its GTPase-binding domain(GBD), WASP can bind CDC42,17 a Rho-family proteininvolved in cytoskeletal organization.18 Moreover,WASP also contains a proline-rich region which binds tothe SH3 homology domain of the adaptor Nck,19 and ofseveral kinases, including Fyn, Tec, Itk, Btk,20–22 andc-src, p85á, phospholipase Cã and c-Fgr.23 Finally, at its3* end, WASP shares homology with the actin-regulatingproteins VASP (vasodilator stimulated phosphoprotein)and ENA17 (encoded by Drosophila enabled gene),which have been shown to play pivotal roles in cyto-skeletal organization and in the regulation of actinpolymerization.24 The structural and functionaldomains of the WASP protein are illustrated in Fig. 1,which also shows the locations of the mutations of thepatients reported here.

*Correspondence to: Fabio Facchetti, MD, PhD, Department ofPathology, University of Brescia, Spedali Civili Brescia, 25124 Brescia,Italy.

Contract grant sponsor: Telethon (grant to L.D.N.).

Contract grant sponsor: CNR; Contract grant number:97.04210.CT04.

CCC 0022–3417/98/050099–09 $17.50? 1998 John Wiley & Sons, Ltd.

Received 10 June 1997Accepted 7 November 1997

Actin is one of the main components of the cyto-skeleton and participates in many cellular processes,such as motility, control of shape, signal transduction,and cell division.25 In response to extracellular stimuli,actin goes through a spatially and temporally regulatedseries of polymerization and depolymerization reac-tions.26,27 Double-helix actin filaments (F-actin), thebiologically active form of actin, result from polymer-ization of the 42 kD actin monomer (G-actin).28 Thedynamics of cell assembly and disassembly of actin isfinely regulated by several actin-interacting proteins,involved in the capping, severing, and binding of actin.29

In immune responses, physical contact occurs betweenhelper T-lymphocytes and antigen-presenting cells. Thiscontact is fundamental and requires T-cell cytoskeletonpolarization,30 elicited following signalling through theTCR/CD3 complex. This process, which involves actinpolymerization and pseudopod formation, is CDC42-mediated and is also accompanied by polarization of theactin-binding proteins alpha-actinin, talin, and vinculin,with alpha-actinin playing a major role in pseudopodformation.31

The data suggesting a cytoskeletal abnormality inWAS was based on T-cells.13 On the other hand,although abnormalities of antibody production are alsotypically observed in WAS, little is known about thecytoskeleton organization in B-cells from WAS patients.Taking advantage of a monoclonal antibody raisedagainst human WASP, we have recently shown thatWASP is expressed in B-lymphocytes.32 Recent datahave shown that, at least in vitro, WASP interacts with

the Bruton’s tyrosine kinase Btk,22 a protein criticallyinvolved in B-cell differentiation and activation. Fur-thermore, abnormalities of signal transduction inresponse to B-cell receptor triggering have been demon-strated in B-cells from patients with WAS.14

In order to establish whether defective B-cell functionis associated with abnormalities of cytoskeleton organ-ization, in the present study we have analysed theexpression of WASP, as well as F-actin and alpha-actinin distribution, in EBV-transformed B-cell linesfrom a cohort of patients with WAS or XLT.

MATERIALS AND METHODS

Patients

Seven patients with WAS/XLT were studied. Theclinical and molecular features of the patients arereported in Table I. The severity of the disease wasscored according to Zhu et al.33 Briefly, a score of 1 wasgiven to patients with thrombocytopenia and smallplatelets, but without any other clinical findings. A scoreof 2 was attributed to patients with thrombocytopeniaand mild, transient eczema, with or without minorinfections. Thrombocytopenic patients with persistent,but treatable eczema and/or recurrent infections weregiven a score of 3. Persistent and difficult to treat eczemaassociated with life-threatening infections and thrombo-cytopenia was scored as 4. If autoimmune diseases werepresent, patients received a score of 5. According to thisscoring system, XLT patients fall into the group with

Fig. 1—Schematic representation of the WASP structure and function. The location of the mutation identified in the patientsdescribed is reported on top, with the following symbols: – – – –=large deletion; .=point mutation, missense; 0=deletion,frameshift. Bold numbers identify the various patients; patient 4 is not included as his mutation has not yet beencharacterized. In the diagram showing the WASP structure, the various domains are indicated as follows: WH1 and WH2:WASP homology domains 1 and 2; GBD: GTPase-binding domain; PRD: proline-rich domain. Below the diagram,molecular partners of WASP, other proteins sharing homologies with distinct domains of WASP, and putative functions ofthe various WASP domains are illustrated (CO: cytoskeleton organization; ST: signal transduction).

100 F. FACCHETTI ET AL.

? 1998 John Wiley & Sons, Ltd. , . 185: 99–107 (1998)

score 1, whereas patients with moderate and severetypical WAS have a score of 2–3 and 4–5, respectively.

Three age-matched individuals hospitalized for minorhead trauma were included as controls.

EBV-transformed B-cell lines

Peripheral blood mononuclear cells (PBMC) wereisolated from heparinized blood after Ficoll-Hypaquegradient centrifugation. For production of EBV-transformed B-cell lines, PBMC (5#105/ml) were incu-bated in RPMI1640 containing 10 per cent fetal calfserum (FCS), 2 m glutamine, and antibiotics in thepresence of supernatant from the B95-8 marmoset linecontaining EBV. Cell lines were cultured at 37)C in 5 percent CO2. The EBV-transformed B-cell lines frompatient 7 have been previously described (FS) in Stewartet al.32

Mutation analysis

For mutation analysis, DNA was extracted fromEDTA-containing blood samples, obtained frompatients and their mothers. Identification of the genedefect at the WASP locus was performed by polymerasechain reaction (PCR) amplification of exon–intronboundaries followed by single-strand conformationpolymorphism (SSCP) and heteroduplex analysis andgenomic sequencing, as previously described.34

The mutations in the WASP gene and the predictedeffect on WASP protein are reported in Table I. Forpatient 4, in whom mutation analysis has not yet beencarried out, non-random X-chromosome inactivationhas been demonstrated in maternal T- andB-lymphocytes and monocytes, using a PCR methyl-ation assay at the human androgen receptor locus asdescribed by Wengler et al.35

Western blotting

EBV-transformed B-cells from patients and fromhealthy controls were lysed at 108 cells/ml in buffercontaining 300 m NaCl, 50 m Tris–HCl, 2 m

EDTA, 0·5 per cent Triton X-100, 2·5 ì p-nitrophenylp*-guanidino-benzoate, 10 ìg/ml aprotinin, and 10 ìg/mlleupeptin. Lysates were centrifuged at 13 000 g for20 min and 100 ìg of protein was electrophoresed on a12 per cent polyacrylamide gel after dilution in anequal volume of SDS sample buffer. The proteins wereelectrotransferred onto nylon membrane (Immobilon-P,Millipore Corp., Bedford, MA, U.S.A.).

The membrane was blocked for 1 h in phosphate-buffered saline (PBS), 1 per cent fish gelatin, 0·1 per centbovine serum albumin (BSA), and 0·5 per cent Tween20, and then blotted for 1 h at room temperature withthe 11G8B7 monoclonal antibody (MAb) at a 1:2000dilution. After incubation, the filter was washed withPBS containing 0·1 per cent Tween 20 (PBS-T). The blotwas then incubated for 1 h with a biotinylated goatanti-mouse IgG and, after washing in PBS-T, withhorseradish peroxidase–streptavidin. Detection wasperformed using the enhanced chemiluminescencesystem ECL (Amersham Corp.).

Immunostaining

EBV-transformed B-cells were washed three times in0·9 per cent NaCl solution, resuspended at a concen-tration of 5#106 cells/ml, and utilized for cytospinpreparations (100 ìl/slide). Slides were air-dried for 24 h,then fixed in "20)C absolute ethanol for 30 min, dried,and used for immunocytochemical staining. WASPexpression was analysed using the MAb 11G8B7 (IgG2bkappa), which recognizes a peptide encompassingamino acid residues 202–302 of WASP (D. Stewart,unpublished results). Filamentous actin (F-actin) wasvisualized using both an anti-F-actin MAb (clone NH3,Serotec, Oxford, U.K.) and FITC-conjugated phalloidin(Sigma, St Louis, MO, U.S.A.). For alpha-actininexpression, the MAb JLN20 (BioGenex, San Ramon,CA, U.S.A.) was used. Anti-WASP was applied for60 min at a dilution of 1:50 in Tris buffer (pH 7·4),followed by the streptavidin–biotin complex immuno-peroxidase technique, with diaminobenzidine as chro-mogen. Anti-F-actin and anti-alpha-actinin were diluted1:5 and 1:1 in Tris buffer, respectively: applied for

Table I—Clinical and molecular features of the WAS/XLT patients

Patientno.

Clinicalscore*

WASP mutation

ExonNucleotide

changePredicted effect on

WASP protein

1 1 2 C201T A56V2 2 9 C907del Y291fs, stop at codon 2913 3 10 G1030del V332fs, stop at codon 4444 3 †5 4 10 C1184del P384fs, stop at codon 4446 4 Gross deletion at 5* Absence of protein7 5 1 C116del L27fs, stop at codon 44

*The clinical score was assessed according to Zhu et al.33

†Gene mutation analysis not yet carried out; evidence of non-random X-chromosome inactivation in the patient’smother.

del=deletion; fs=frameshift.

101ACTIN POLYMERIZATION IN WAS


60 min; and followed by FITC-conjugated rabbit anti-mouse immunoglobulin (Dako, Milan, Italy). FITC-conjugated phalloidin was applied at a dilution of 1:5 for30 min.

F-actin, alpha-actinin, and phalloidin staining wasanalysed with a Jenamed-2 fluorescence microscope.

From two WAS patients (Nos 3 and 7) and from threecontrols (unrelated to the EBV-transformed B-cell linedonors), fresh frozen tissue from spleen obtained aftersplenectomy was also available and used for immuno-histochemical evaluation of WASP and alpha-actininexpression; after incubation with the primary antibodyfor 60 min, the indirect streptavidin–biotin peroxidasetechnique was applied and the reaction product wasdeveloped using 3-amino-9-ethylcarbazole (AEC) aschromogen.

Controls for the immunostaining experiments (EBV-transformed B-cell lines and spleen), which were invari-ably negative, included both the omission of the primaryantibody and its replacement with an isotype-matchednon-relevant antibody.

RESULTS

Molecular analysis screening for WASP genemutations by SSCP resulted in abnormalities in exon 1(patient 7), exon 2 (patient 1), exon 9 (patient 2), andexon 10 (patients 3 and 5). In addition, a gross deletionin the 5* region of the gene was detected by PCR inpatient 6. The PCR products that resulted in abnormalSSCP patterns were further analysed by direct sequenc-ing. The specific mutations in the patients and theirpredicted effects at the protein level are shown in Fig. 1and Table I.

Analysis of WASP protein expression

In order to extend and validate at the protein level themolecular data obtained through mutation analysis,WASP protein expression was evaluated by Westernblotting and immunocytochemistry in EBV-transformedB-cell lines derived from six patients with WAS and onepatient with XLT; data are reported in Figs 2 and 3A.Three patients with severe WAS (Nos 5–7) and onepatient with moderate WAS (No. 4) were found not toexpress WASP as detected by the MAb 11G8B7 byimmunocytochemistry (Fig. 4B) and Western blotting.In patient 7, this finding was confirmed by staining ofspleen sections with the same antibody (data notshown). Trace amounts of WASP were detected byimmunocytochemistry and by Western blotting inpatients 2 and 3 with moderate WAS. Patient 2expressed a low molecular weight WASP protein, aspredicted by mutation analysis. In patient 3, the molecu-lar weight of the mutant WASP protein should also belower than normal, but the resolution property ofWestern-blot analysis did not allow the difference to beclearly detected. Furthermore, positivity for WASPexpression was detected in the spleen of patient 3 (datanot shown). Normal amounts of normal-sized WASPwere identified by immunocytochemistry (Fig. 4C) and

Western blotting in patient 1, who had a typical XLT.By immunohistochemistry, in all patients in whomWASP was expressed (although at variable levels),WASP had a cytoplasmic distribution, similar to thatobserved in lymphoblastoid cells lines from controls(Figs 4A and 4C).

F-actin distribution

To assess F-actin distribution, two different tech-niques were used: cells were stained with either FITC-conjugated anti-F-actin or FITC-conjugated phalloidin.The latter is known to bind and to stabilize the subunitsof actin filaments, and is widely used for quantitation ofpolymerized actinin cells.36

In EBV-transformed B-cell lines from normal con-trols, F-actin was mainly recognizable as diffuse cyto-plasmic staining; in addition, strong linear positivity wasfound to decorate microvilli, which were representedeither by single or numerous projections of the cyto-plasm, occasionally polarized at one cell edge (Fig. 5A).Cells with cytoplasmic reactivity were expressed as apercentage of all cells in the cytospin and evaluatedusing a #40 microscopic objective; cells with F-actin-filled microvilli were evaluated by counting 500 cellsusing a #100 microscopic objective. As reported in Figs3B and 5B, cells with diffuse cytoplasmic positivity forF-actin were markedly reduced or abolished in WASpatients (including patient 3, who expressed traceamounts of WASP), whereas no differences wereidentified between controls and the patient with XLT(Figs 3B and 5C). Furthermore, the number of cellsshowing actin-filled microvilli was preserved in the XLTpatient, but was largely, although variably, decreased inpatients with typical WAS (Fig. 3C).

Fig. 2—(A) Western-blot analysis of WASP protein expression (thickarrow) in EBV-transformed B-cell lines from a normal control (C) andfrom WAS/XLT patients 1 to 7. (B) Longer exposure of the same blotas in A, showing expression of trace amounts of WASP protein onWAS patient 2 (expressing a low molecular weight mutant WASP)(thin arrow) and on WAS patient 3



Alpha-actinin distribution

In EBV-transformed B-cell lines from controls, alpha-actinin showed diffuse staining in the cytoplasm of most

cells, with a wide variability of intensity; in addition,scattered cells with strong positivity at one pole of thecell were also recognizable (Fig. 5D). Similar resultswere found in XLT patient 1 (Fig. 5F). In contrast,no polar distribution of alpha-actinin was detected inEBV-transformed B-cell lines from patients with WAS(Fig. 5E), with the exception of patients 2 and 3. Thesedata were corroborated by immunostaining of alpha-actinin in the spleen. In controls, an obvious capping ofreactivity was recognized in lymphoid cells from thewhite pulp (Fig. 6A); in WAS patient 7, positivity wasdiffusely distributed through the cytoplasm withoutpolarization (Fig. 6B), and in patient 3, occasionalcapping was identified.

DISCUSSION

In the present study we have evaluated WASP proteinexpression and cytoskeleton organization in six patientswith WAS (three with severe and three with moderateform of the disease) and in one patient with XLT. TheMAb 11G8B7 failed to detect WASP protein in allpatients with WAS, with the exception of patient 3, whoexpressed trace amounts, and patient 2, who expressed alow molecular weight mutant form. In contrast, patient1 (with XLT) had normal amounts of normal-sizedWASP protein. These data are in keeping with thegenotypic analysis; in particular, patient 1 is the onlyone in our series who carries a missense mutation inexon 2 of the WASP gene. Single amino acid substitu-tions in this part of the protein have been commonlyfound in XLT patients.8,10

A single nucleotide deletion in exon 9 with prematuretermination has been identified in patient 2; this defect ispredicted to result in a truncated form of the protein, asin fact confirmed by Western blotting. Patient 3 (withmoderate WAS) and patient 5 (with severe WAS) hadsingle nucleotide deletions in exon 10, predicted to resultin frameshift and premature termination of codon 444 inboth cases. Trace amounts of WASP were detected inpatient 3, but not in patient 5. It is possible that themutant WASP proteins expressed in these patientsundergo intracellular degradation with some variabilityfrom patient to patient. The remaining two patientswith severe WAS (Nos 6 and 7) had gross deletion (No.6) or early frameshift and premature termination (No. 7)in the WASP gene, which inhibit WASP proteinexpression, as confirmed by immunocytochemistry andWestern blotting. Our study confirms and furtherexpands the data recently reported by O’Donnell et al.using Western and Northern blot, which show that theamount of WASP mRNA/protein expressed by B-celllines from patients with WAS correlates with the severityof the disease.10

Structural abnormalities in the form of markeddeficiency and shortening of microvilli have been pre-viously reported in T-cells and T-cell lines from WASpatients.12,37 In addition to differences in WASP proteinexpression, we have shown that EBV-transformedB-cell lines from the patients with WAS and the subjectwith XLT also differed markedly in terms of F-actin

Fig. 3—Immunocytochemical expression of WASP (A), cytoplasmicF-actin (B) and the number of cells showing microvilli filled withF-actin (C) in EBV-transformed B-cell lines from three controls (CNT)and in WAS/XLT patients 1 to 7. In A, +, +/", and " indicatepositivity on most cells, weak and variable positivity, and negativityfor WASP, respectively. In B, values are expressed as the percentage oftotal cells expressing cytoplasmic F-actin. In C, values are expressed asthe number of cells in a total of 500 showing F-actin-positive microvilli



polymerization. The number of cells expressing cyto-plasmic F-actin was significantly reduced or even absent

in WAS patients (with the exception of patient 2),compared with controls and the XLT subject; inaddition, cytoplasmic projections containing F-actinand recognizable as microvilli were also reduced inpatients with WAS. In particular, severe defects ofcytoplasmic F-actin and of F-actin-positive microvilliwere observed in all patients who lacked WASP proteinexpression (patients 4–7), strongly supporting a rolefor WASP in the organization of the cytoskeleton viaregulation of F-actin polymerization.

WASP shows partial structural and functionalidentity with other proteins, such as bovine neuralWASP (N-WASP),38 human VASP (vasodilator-stimulated phosphoprotein), and ENA encoded byDrosophila enabled gene,17 which have been demon-strated to play pivotal roles in the organization of thecytoskeleton and in the regulation of actin polymeriz-ation.24 WASP may participate in actin polymerizationvia different mechanisms.39 It has been shown thatWASP interacts with Rac and CDC42, members of Rhofamily proteins, which regulate the formation of poly-merized structures of actin.18 In NIH3T3 fibroblasts,Rac induces lamellipodia and membrane ruffles,whereas CDC42 leads to the formation of filopodiaprotrusions.40 Fyn,20 a protein-tyrosine kinase (PTK)belonging to the c-Src family, is another partner-proteinof WASP and has also been found to be involved in thepolymerization and pattern distribution of actin.41

Finally, Symons et al.17 have provided evidence that incells transfected with WASP, the expression of thecorresponding protein induces the formation of F-actin-rich particles and, in overexpressing cells, denseaggregates of WASP and polymerized actin occur.Interestingly, it has been recently demonstrated thatBee1, a yeast homologue of WASP, is critical for theorganization of the actin cytoskeleton, Bee1 mutationsresulting in aberrant assembly of actin structures.42

Taken together, these data support the hypothesis thatWASP and Bee1 represent phylogenetically conservedproteins involved in actin polymerization.43

We further investigated the biological role played byWASP in the cytoskeleton organization, looking at thealpha-actinin distribution in controls and WAS/XLTpatients. Alpha-actinin is an actin-binding protein and isknown to participate in F-actin polymerization.44 Invitro, alpha-actinin has been shown to interact withbeta-I-integrin,45 acting as a linker between actin andthe cell membrane. Furthermore, CD54/ICAM-1,46 thecytoplasmic domain of both CD1847 and L-selectin,48 isassociated with alpha-actinin, suggesting that this link-age may contribute to the attachment of actin filamentsto the cell membrane in certain locations.45 In thepresent study, capping of alpha-actinin was observed inEBV-transformed B-cell lines from controls and from

Fig. 4—Expression of WASP protein in EBV-transformed B-cell linesfrom a control (A), from patient 7 with full-blown WAS (B), and frompatient 1 with XLT (C). In A and C, similar immunoreactivity forWASP is observed in the cytoplasm, whereas no expression is detectedin B. (Monoclonal antibody 11G8B7 anti-WASP, streptavidin–biotincomplex immunoperoxidase technique, light counterstain with haema-toxylin; all #750)



Fig. 5—Examples of F-actin (A, B, C) and alpha-actinin (D, E, F) distribution in EBV-transformed B-cell lines from a control (A,D), from a patient with full-blown WAS (patient 5) (B, E), and from the XLT patient (patient 1) (C, F). In the control and in theXLT patient, F-actin is expressed in the cytoplasm of most cells; in addition, microvilli strongly positive for F-actin are alsorecognizable. In A, a higher magnification of a cell with F-actin-positive microvilli polarized at one cell edge is shown (inset). Strongpositivity for alpha-actinin with capping at one pole of the cell is recognizable in D and F, whereas in E only weak and diffusecytoplasmic reactivity is evident. [FITC-conjugated phalloidin (A, B, and C) and indirect immunofluorescence for alpha-actinin (D,E, and F). A, B, and C: #300; inset, D, E, and F: #750]


one XLT patient; it was reduced but still detectable inpatients 2 and 3 with moderate WAS who expressed atrace amount of mutant WASP protein, but it wascompletely abrogated in all WAS patients who lackeddetectable WASP protein. Furthermore, lack of alpha-actinin capping was also documented in the spleen fromone patient with severe WAS.

As a whole, the severity of the complex array ofanomalies of the cytoskeleton organization in WASseems to correlate with the degree of impairment ofWASP protein expression. It is likely that theseabnormalities affect B-cell function, since it has beenshown that after B-cell activation, structural changesin the distribution of surface immunoglobulins andactin polymerization occur,49–51 and surface immuno-globulins form macromolecular complexes with alpha-actinin.52 In WAS patients, a primary defect in thehumoral immune response is indicated by the lack ofB-cell activation to thymus-independent type 2 (TI-2)antigens, such as polysaccharides. Since it has beenshown that B-cell activation by T-independent antigensrequires cross-linking of surface immunoglobulins inorder to achieve the threshold level of antigen that isnecessary to induce B-cell stimulation,53 we can specu-late that in WAS patients this process is defective due toabnormal cytoskeletal organization and cell surfaceremodelling.

As a whole, our data on EBV-transformed B-cell linesderived from WAS patients support the notion thatmutations of WASP gene interfere with the cytoskeletonarchitecture in B cells and may thus contribute to thedefective humoral immunity observed in WAS. Furtherstudies on freshly isolated B-lymphocytes under variousactivation conditions may help in the further assessmentof the role of WASP in cytoskeletal organization andperipheral B-cell functions.

ACKNOWLEDGEMENTS

This work was supported in part by Telethon (Grantto L. D. Notarangelo) and the CNR (Grant97.04210.CT04 to L. D. Notarangelo). We thank Mrs L.

Salvi and M. L. Breda for technical assistance, Mrs C.Facchini for typing the manuscript and Mr F. Alpi forpreparing the microphotographs.

REFERENCES

1. Aldrich RA, Steinberg AG, Campbell DC. Pedigree demonstrating asex-linked recessive condition characterized by draining ears, eczematoiddermatitis and bloody diarrhea. Pediatrics 1954; 13: 133–139.

2. Wiskott A. Familiarer, angeborener Morbus Werlhofii? Monatschr Kinder-heilk 1937; 68: 212–216.

3. Sullivan KE, Mullen CA, Blaese RM, Winkelstein JA. A multiinstitutionalsurvey of the Wiskott–Aldrich syndrome. J Pediatr 1994; 125: 876–885.

4. Donner M, Schwartz M, Carlsson KU, Holmberg L. Hereditary X-linkedthrombocytopenia maps to the same chromosomal region as the Wiskott–Aldrich syndrome. Blood 1988; 72: 1849–1853.

5. Ochs HD, Slichter SJ, Harker LA, Von Behrens WE, Clark RA, WedgwoodRJ. The Wiskott–Aldrich syndrome: studies of lymphocytes, granulocytesand platelets. Blood 1980; 55: 243–252.

6. Cooper MD, Chase HP, Lowman JT, Krivit W, Good RA. Wiskott–Aldrich syndrome: immunologic deficiency disease involving the afferentlimb of immunity. Am J Med 1968; 44: 499–513.

7. Derry JMJ, Ochs HD, Francke U. Isolation of a novel gene mutated inWiskott–Aldrich syndrome. Cell 1994; 78: 635–644.

8. Schwarz K. WASPbase: a database of WAS and XLT-causing mutations.Immunol Today 1996; 17: 496–502.

9. Villa A, Notarangelo LD, Macchi P, et al. X-linked thrombocytopenia andWiskott–Aldrich syndrome are allelic diseases with mutation in the WASPgene. Nature Genet 1995; 9: 414–417.

10. Remold-O’Donnel E, Colley J, Shcherbina A, et al. Variable expression ofWASP in B cell lines of Wiskott–Aldrich syndrome patients. J Immunol1997; 158: 4021–4025.

11. Kenney DM. Wiskott–Aldrich syndrome and related X-linked thrombo-cytopenia. Curr Opin Pediatr 1990; 2: 931–934.

12. Kenney DM, Cairns L, Remold-O’Donnel E, Peterson J, Rosen FS,Parkman R. Morphological abnormalities in the lymphocytes of patientswith the Wiskott–Aldrich syndrome. Blood 1986; 68: 1329–1332.

13. Remold-O’Donnel E, Rosen FS, Kenney DM. Defects in Wiskott–Aldrichsyndrome blood cells. Blood 1996; 87: 2621–2631.

14. Simon HU, Mills GB, Hashimoto S, Siminovitch KA. Evidence fordefective transmembrane signalling in B cells from patients with Wiskott–Aldrich syndrome. J Clin Invest 1992; 90: 1396–1405.

15. Molina IJ, Sancho J, Terhorst C, Rosen FS, Remold-O’Donnel E. T cells ofpatients with the Wiskott–Aldrich syndrome have a restricted defect inproliferative responses. J Immunol 1993; 151: 4383–4390.

16. Kwan SP, Hagemann TL, Blaese RM, Knutsen A, Rosen FS. Scanning ofthe Wiskott–Aldrich syndrome (WAS) gene: identification of 18 novelalterations including a possible mutation hotspot at Arg86 resulting inthrombocytopenia, a mild WAS phenotype. Hum Mol Genet 1995; 4:1995–1998.

17. Symons M, Derry JMJ, Karlak B, et al. Wiskott–Aldrich syndrome protein,a novel effector for the GTPase CDC42Hs, is implicated in actin poly-merization. Cell 1996; 84: 723–734.

18. Tapon N, Hall A. Rho, Rac and CDC42 GTP-ases regulate the organiz-ation of the actin cytoskeleton. Curr Opin Cell Biol 1997; 9: 86–92.

Fig. 6—Expression of alpha-actinin in the white pulp of the spleen from a control (A) and from patient 7 was WAS (B). In thecontrol lymphoid cells show strong and polar reactivity, whereas in the WAS patient alpha-actinin is diffusely distributed inthe cytoplasm without capping. (Monoclonal antibody anti-alpha-actinin, streptavidin–biotin complex immunoperoxidasetechnique, light counterstain with haematoxylin; all #350)



19. Rivero-Lezcano OM, Marcilla A, Sameshima JH, Robbins KC. Wiskott–Aldrich syndrome protein physically associates with Nck through Srchomology 3 domains. Mol Cell Biol 1995; 15: 5725–5731.

20. Banin S, Truong O, Katz DR, Waterfield MD, Brickell PM, Gout I.Wiskott–Aldrich syndrome protein (WASP) is a binding partner for c-Srcfamily protein-tyrosine kinases. Curr Biol 1996; 6: 981–988.

21. Bunnel SC, Henry P, Kollury R, Kirchhausen T, Rickles RJ, Berg LJ.Identification of Itk/Tsk Src homology 3 domain ligands. J Biol Chem 1996;271: 25 646–25 656.

22. Cory GOC, Mac-Carty-Morrogh L, Banin S, et al. Evidence that theWiskott–Aldrich syndrome protein may be involved in lymphoid cellsignalling pathway. J Immunol 1996; 157: 3791–3795.

23. Finnan PM, Soames CJ, Wilson L, et al. Identification of regions of theWiskott–Aldrich syndrome protein responsible for association with selectedSrc homology 3 domains. J Biol Chem 1996; 271: 26 291–26 295.

24. Haffner C, Jarchau T, Reinhard M, Hoppe J, Lohmann SM, Walter U.Molecular cloning, structural analysis and functional expression of theproline-rich focal adhesion and microfilament-associated protein VASP.EMBO J 1995; 14: 19–27.

25. Bremer A, Aebi U. The structure of the F-actin filament and the actinmolecule. Curr Opin Cell Biol 1992; 4: 20–26.

26. Stossel TP. On the crawling of animal cells. Science 1993; 260: 1086–1094.27. Welch MD, Mallavarapu A, Rosenblatt J, Mitchison TJ. Actin dynamics in

vivo. Curr Opin Cell Biol 1997; 9: 54–61.28. Pollard TD. Actin. Curr Opin Cell Biol 1990; 2: 33–40.29. Vandekerckhove J. Actin-binding proteins. Curr Opin Cell Biol 1990; 2:

41–50.30. Kupfer A, Swain SL, Janewaj CA, Singer SJ. The specific direct interaction

of helper T cells and antigen-presenting B cells. Proc Natl Acad Sci USA1986; 83: 6080–6083.

31. Selliah N, Brooks WH, Roszman TL. Proteolytic cleavage of alpha-actininby calpain in T-cells stimulated with anti-CD3 monoclonal antibody.J Immunol 1996; 156: 3215–3221.

32. Stewart DM, Treiber-Held S, Kurman CC, Facchetti F, Notarangelo LD,Nelson DL. Studies of the expression of the Wiskott–Aldrich syndromeprotein. J Clin Invest 1996; 97: 2627–2634.

33. Zhu Q, Zhang M, Blaese RM, et al. The Wiskott–Aldrich syndrome andX-linked congenital thrombocytopenia are caused by mutation of the samegene. Blood 1995; 86: 3797–3804.

34. Wengler GS, Notarangelo LD, Berardelli S, et al. High prevalence ofnonsense, frame shift, and splice-site mutations in 16 patients with full-blown Wiskott–Aldrich syndrome. Blood 1995; 86: 3648–3654.

35. Wengler GS, Parolini O, Fiorini M, et al. A PCR-based non-radioactiveX-chromosome inactivation assay for genetic counselling in X linkedprimary immunodeficiencies. Life Sci 1997; 61: 1405–1411.

36. Cooper JA. Effects of cytochalasin and phalloidin on actin. J Cell Biol 1987;105: 1473–1478.

37. Molina IJ, Kenny DM, Rosen FS, Remold-O’Donnel E. T cell linescharacterize events in the pathogenesis of the Wiskott–Aldrich syndrome.J Exp Med 1992; 176: 867–874.

38. Miki H, Miura K, Takenawa T. N-WASP. A novel actin-depolymerizingprotein, regulates the cortical cytoskeletal rearrangement in a PIP2-dependent manner downstream of tyrosine kinases. EMBO J 1996; 15:5326–5335.

39. Featherstone C. The many faces of WAS protein. Science 1997; 275: 27–28.

40. Kozma R, Ahmed S, Best A, Lim L. The Ras-related protein Cdc42Hs andbradykinin promote formation of peripheral actin microspikes and filopodiain Swiss 3T3 fibroblasts. Mol Cell Biol 1995; 15: 1942–1952.

41. Thomas SM, Soriano P, Imamoto A. Specific and redundant roles of Srcand Fyn in organizing cytoskeleton. Nature 1995; 376: 267–271.

42. Li R. Bee1, a yeast protein with homology to Wiskott–Aldrich syndromeprotein, is critical for the assembly of cortical actin cytoskeleton. J Cell Biol1997; 136: 649–658.

43. Lechler T, Li R. In vitro reconstitution of cortical actin assembly sites inbudding yeast. J Cell Biol 1997; 138: 95–103.

44. Blanchard A, Ohanion V, Critchley D. The structure and function ofalpha-actinin. J Muscle Res Cell Motil 1989; 10: 280–289.

45. Otey CA, Pavalko FM, Burridge K. An interaction between alpha-actininand the beta-1 integrin subunit in vitro. J Cell Biol 1990; 111: 721–729.

46. Carpen O, Pallai P, Staunton DE, Springer TA. Association of intercellularadhesion molecule-1 (ICAM-1) with actin-containing cytoskeleton andalpha actinin. J Cell Biol 1992; 118: 1223–1234.

47. Pavalko FM, Walker DM, Graham L, Goheen M, Doerschuk CM, KansasGS. The cytoplasmic domain of -selectin interacts with cytoskeletalproteins via alpha-actinin: receptor positioning in microvilli does notrequire interaction with alpha-actinin. J Cell Biol 1995; 129: 1155–1564.

48. Pavalko FM, LaRoche SM. Activation of human neutrophils induces aninteraction between the integrin beta-subunit (CD18) and the actin bindingprotein alpha-actinin. J Immunol 1993; 151: 3795–3807.

49. Albrecht DL, Noelle RJ. Membrane Ig–cytoskeletal interactions. I. Flowcytofluorometric and biochemical analysis of membrane IgM–cytoskeletalinteractions. J Immunol 1988; 141: 3915–3922.

50. Braun J, Hochman PS, Unanue ER. Ligand-induced association of surfaceimmunoglobulin with the detergent-insoluble cytoskeletal matrix of the Blymphocyte. J Immunol 1982; 128: 1198–1204.

51. Melamed I, Downey GP, Roifman CM. Microfilament assembly is requiredfor antigen receptor mediated activation of human B-lymphocytes. JImmunol 1991; 147: 1139–1146.

52. Gupta SK, Woda BA. Ligand-induced association of surface immunoglobu-lin with the detergent insoluble cytoskeleton may involve alpha-actinin.J Immunol 1988; 140: 176–182.

53. Mond JJ, Vos Q, Lees A, Snapper CM. T cell independent antigens. CurrOpin Immunol 1995; 7: 349–354.

NOTE ADDED IN PROOF

Very recently, results similar to those presented in thispaper were obtained by Gallego et al., who showed thatT-cells from WAS patients immortalized with Herpes-virus Saimiri display defective actin polymerizationand reorganization in response to CD3-mediatedstimulation.11. Gallego MD, Santamaria M, Pena J, Molina IJ. Defective actin reorgan-

ization and polymerization of Wiskott-Aldrich T cells in response toCD3-mediated stimulation. Blood 1997; 90: 3089–3097.



Documents

Defective actin polymerization in EBV-transformed B-cell lines from patients with the Wiskott-Aldrich syndrome