SEARCH:   Advanced

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Rights and Permissions Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Ward, C. M. Articles by Newman, P. J. Search for Related Content PubMed PubMed Citation Articles by Ward, C. M. Articles by Newman, P. J. Related Collections Hemostasis, Thrombosis, and Vascular Biology Social Bookmarking                   What's this?  Previous Article  |  Table of Contents  |  Next Article  Blood, Vol. 96 No. 1 (July 1), 2000: pp. 161-169 HEMOSTASIS, THROMBOSIS, AND VASCULAR BIOLOGY A Leu262Pro mutation in the integrin 3 subunit results in an IIb-3 complex that binds fibrin but not fibrinogen Christopher M. Ward, Anita S. Kestin, and Peter J. Newman From the Blood Research Institute, The Blood Center of Southeastern Wisconsin, Milwaukee, WI; the Department of Medicine, Brown University School of Medicine, Providence, RI; and the Departments of Cellular Biology and Pharmacology, Medical College of Wisconsin, Milwaukee, WI.     Abstract Top Abstract Introduction Patients and methods Results Discussion Acknowledgments References Platelet retraction of a fibrin clot is mediated by the platelet fibrinogen receptor, IIb3. In certain forms of the inherited platelet disorder, Glanzmann thrombasthenia (GT), mutant IIb3 may interact normally with fibrin yet fail to support fibrinogen-dependent aggregation. We describe a patient (LD) with such a form of GT. Platelets from LD supported normal clot retraction but failed to bind fibrinogen. Platelet analysis using flow cytometry and immunoblotting showed reduced but clearly detectable IIb3, findings consistent with type II GT. Genotyping of LD revealed 2 novel 3 mutations: a deletion of nucleotides 867 to 868, resulting in a premature stop codon at amino acid residue 267, and a T883C missense mutation, resulting in a leucine (Leu) 262-to-proline (Pro) substitution. Leu262 is highly conserved among  integrin subunits and lies within an intrachain loop implicated in subunit association. Leu262Pro3 cotransfected with wild-type IIb into COS-7 cells showed delayed intracellular maturation and reduced surface expression of easily dissociable complexes. In human embryonic kidney 293 cells, Leu262Pro3 formed a complex with endogenous av and retracted fibrin clots similarly to wild-type 3. The same cells, however, were unable to bind immobilized fibrinogen. The molecular requirements for IIb3 to interact with fibrin compared with fibrinogen, therefore, appear to differ. The region surrounding 3 Leu262 may maintain 3 in a fibrinogen-binding, competent form, but it appears not to be required for receptor interactions with fibrin. (Blood. 2000;96:161-169) © 2000 by The American Society of Hematology.     Introduction Top Abstract Introduction Patients and methods Results Discussion Acknowledgments References The aggregation of platelets and subsequent thrombus formation is mediated by the platelet-specific integrin IIb3 (glycoprotein IIb-IIIa). Platelet IIb3 is a heterodimeric surface receptor consisting of an alpha (IIb) and a beta (3) subunit in a noncovalent, cation-dependent complex.1 Platelet aggregation depends on IIb3 binding of fibrinogen,2 but the receptor can recognize other ligands, including fibronectin, von Willebrand factor (vWf), and vitronectin.3 IIb3 binding of soluble fibrinogen requires activation of the receptor by extracellular or intracellular events, a process that is poorly understood. A valuable tool for studying IIb3 structure and function is the rare, autosomal recessive bleeding disorder Glanzmann thrombasthenia (GT), in which IIb3 is dysfunctional or absent.4 GT mutations characterized at the molecular level are heterogenous and affect the IIb and 3 genes equally.4,5 In many patients, gene deletions, frameshifts, and premature terminations result in IIb3 expression at less than 5% of normal levels (type I GT). Of more interest are cases in which a dysfunctional IIb3 complex is present on the platelet surface at reduced (type II GT) or nearly normal (variant GT) levels. Type II and variant GT are predominantly due to point mutations of the 3 gene in 2 critical areas. First, mutations of the charged residues arginine (Arg) 119, Arg214, and Arg2164 in the 3 extracellular domain lead to the identification of a cation binding motif (DXSXS), which is critical for ligand binding and homologous to the metal ion-dependent adhesion site (MIDAS) of certain integrin  subunits.6,7 Second, mutations in the cytoplasmic domain of 3, such as Arg724Stop,8 result in a receptor that cannot be activated by intracellular signals (inside-out signaling) but still binds ligand in response to extracellular events. Therefore, thrombasthenic mutations can identify critical residues in IIb3 that participate in ligand binding and receptor activation. The molecular events involved in IIb3 ligand binding have been partly described; like many integrins, IIb3 recognizes an Arg-glycine (Gly)-aspartic acid (Asp), or RGD, motif present in some adhesive ligands.3 One of these, fibrinogen, is an oligomeric complex of 3 paired subunits, the A, B, and  chains, and contains 4 RGD motifs, 2 in each A chain, at residues 95 to 97 and 572 to 574.9 In addition, a 12-amino acid motif, HHLGGAKQAGDV, comprising residues 400 to 411 of the  chain, binds IIb3 and competes with RGD for binding.10 Studies with fibrinogen mutants have shown that this C-terminal dodecapeptide is essential for fibrinogen-dependent platelet aggregation, whereas the A-chain RGD sequences are not required.11-13 Several research groups have attempted to localize the fibrinogen-binding site at the receptor level. Recombinant truncated extracellular domains of IIb (residues 1-233) and 3 (residues 111-318) formed an RGD-binding complex.14 Cross-linking and peptide studies identified 3 residues 109 to 171 and 211 to 222 as potential fibrinogen-binding sites.15-17 These 2 regions include residues identified in GT or by mutagenesis6 as part of a MIDAS-like cation binding site in 3. In IIb, the -chain peptide of fibrinogen was cross-linked to residues 294 to 314 of IIb, suggesting that this is also a ligand-recognition site,18 and mutation of IIb residues Gly184 to Gly193 was reported to abolish fibrinogen binding.19 The molecular interactions of IIb3 with insoluble fibrin are less well understood. During clot retraction, activated platelets cause a marked reduction in the volume of a fibrin clot.20 Clot retraction can also be induced by nucleated cells21 and is integrin dependent.22 In particular, the 3 subunit is important in binding fibrin, triggering retraction through links between the cytoplasmic domain of 3 and cytoskeletal proteins.23 In cultured mammalian cell lines, 3, as part of the vitronectin receptor, v3, induced spontaneous fibrin-clot retraction, whereas IIb3 required prior activation of the integrin for retraction.24,25 In contrast to the situation with platelet aggregation, the C-terminal of the fibrinogen  chain does not appear to be required for fibrin-clot retraction mediated by IIb3,12,13 suggesting that fibrinogen and fibrin may bind to different sites on the receptor. In the current study, we identified a novel 3 mutation, leucine (Leu) 262 to proline (Pro) (Leu262Pro), in a patient with type II GT whose platelets showed normal clot retraction. This mutation is predicted to alter the conformation of a region of 3 located between 2 putative ligand-binding sites and involved in subunit interactions. By expressing Leu262Pro3 in complex with v in mammalian cells, we demonstrated that the mutant receptor binds fibrin in a similar manner to wild-type 3 but does not recognize fibrinogen.     Patients and methods Top Abstract Introduction Patients and methods Results Discussion Acknowledgments References Patient studies Patient LD was a 3-year-old girl who presented postnatally with skin petechiae and ecchymoses. She had frequent bruising with minimal trauma and gum bleeding after tooth eruption. The patient's mother had a history of menorrhagia but was found to have von Willebrand disease; her father had no history of abnormal bleeding. Laboratory testing of LD showed that she had a normal platelet count (360 × 109/L) and a prolonged bleeding time. Prothrombin time, partial thromboplastin time, factor VIII, vWF antigen, and ristocetin cofactor levels were also normal. Clot retraction by platelets Whole blood from patient LD, her parents, and healthy controls was anticoagulated with acid citrate dextrose A, and prostaglandin E1 (50 ng/mL) was added before centrifugation at 220g to separate platelet-rich plasma (PRP). PRP samples were diluted in homologous platelet-poor plasma to final platelet counts of 2.5, 1.25, and 0.6 × 108 platelets/mL and incubated in glass tubes for 5 minutes at 37°C. Clot formation was induced by calcium chloride (8 mmol/L final concentration), and samples were photographed after 1 hour of incubation at 37°C. Antibodies The IIb-specific monoclonal antibody (mAb) Tab was a gift from Dr Rodger McEver (University of Oklahoma).26 The 3-specific mAb AP3 was described previously.27 The mAb AP1, which recognizes GPIb, and the mAb AP2, which binds to a complex-dependent epitope on IIb3,28 were provided by Dr Robert R. Montgomery (Blood Research Institute, Blood Center of Southeastern Wisconsin, Milwaukee, WI). The mAb LM609, which recognizes the v3 complex,29 was a gift from Dr David A. Cheresh (Scripps Institute, La Jolla, CA). The mAb P4C10 against human 1 integrin was obtained from GIBCO BRL (Gaithersburg, MD). Rabbit polyclonal anti-IIb and anti-3 antibodies were prepared at the Blood Research Institute by using standard procedures. Flow-cytometric analysis of platelet membrane glycoproteins PRP from patient and control samples was centrifuged at 500g, and the platelet pellet was washed and resuspended at 4 × 108 platelets/mL in RCD-EDTA buffer (108 mmol/L sodium chloride [NaCl], 38 mmol/L potassium chloride, 1.7 mmol/L sodium bicarbonate, 21.2 mmol/L sodium citrate, 27.8 mmol/L glucose, 1.1 mmol/L magnesium chloride-6H20 , and 2 mmol/L EDTA [pH 6.5]). For each analysis, 1 × 107 platelets in RCD-EDTA buffer plus 0.2% (wt/vol) bovine serum albumin (BSA) were incubated with a mouse mAb at a final concentration of 20 µg/mL for 1 hour at room temperature. Samples were then washed in the same buffer and incubated with a 1:20 dilution of fluorescein isothiocyanate-conjugated-labeled goat-antimouse IgG (Jackson Immunoresearch Laboratories, West Grove, PA) for 30 minutes in the dark. The samples were washed again, diluted in RCD-EDTA buffer and analyzed on a fluorescence-activated cell-sorter scan flow cytometer (Becton Dickinson, Mountain View, CA). Semiquantitative Western blot analysis of platelet glycoproteins Platelet lysates were prepared by lysing washed platelets in 50 mmol/L Tris (pH 6.8), 1% (vol/vol) Triton X-100, 10 mmol/L N-ethylmaleimide, 2 mmol/L phenylmethylsulfonyl fluoride (PMSF), and 20 µmol/L leupeptin. After centrifugation at 1500g, the total protein concentration of the supernatants was measured by a bicinchoninic acid assay (Pierce, Rockford, IL). Comparable amounts of thrombasthenic and control total platelet protein were analyzed on an sodium dodecyl sulfate (SDS)-9% polyacrylamide gel under reducing conditions and transferred to a polyvinylidene fluoride (PVDF) membrane (Immobilon; Millipore, Bedford, MA). The membrane was incubated with a mixture of rabbit polyclonal antibodies directed against IIb and 3 (both at a final concentration of 5 µg/mL), then detected with a 1:2000 dilution of alkaline phosphatase-conjugated goat-antirabbit IgG (Jackson Immunoresearch) and color development with the nitro blue tetrazolium-5-bromo-4-chloro-3-indolyl phosphate substrate pair (Sigma, St Louis, MO). Polymerase chain reaction amplification of IIb and 3 fragments from genomic DNA Genomic DNA was prepared from peripheral blood leukocytes and amplified by polymerase chain reaction (PCR) with Taq polymerase by using oligonucleotide primers from intronic flanking sequences of IIb and 3. All exons of both genes were sequenced at least twice to exclude Taq errors. Exon 5 of 3 was amplified by using the sense primer 5'-CTCTACCAGTGACATGGCTGAA-3' and the antisense primer 5'-CAAGCTGAAACGAGCCCTGCC-3'. The PCR protocol entailed 5 minutes of denaturation at 100°C and 3 minutes of annealing at 56°C before the addition of Taq polymerase, then 30 cycles each of a 1-minute extension at 72°C, 45 seconds of denaturation at 96°C, and 45 seconds of annealing at 56°C. PCR products were extracted from agarose gels and analyzed by direct cycle sequencing. To detect multiple alleles from the patient LD and her parents, PCR products were subcloned into a TA overhang vector, pCRII (Invitrogen, Carlsbad, CA), and individual clones were sequenced. PCR-based cartridge mutagenesis of 3 complementary DNA Full-length complementary DNA (cDNA) encoding 3, mutated to delete an internal EcoRI cleavage site (a gift of Dr Gilbert C. White, University of North Carolina School of Medicine), was cloned into plasmid vector pBluescript SK (Stratagene, San Diego, CA). Overlap PCR was used to generate a fragment of 3 containing the T883C mutation found in patient LD. Briefly, primers with a single nucleotide mismatch at nucleotide 883 (sense primer, 5'-GGACGGAAGGCCGGCAGGCATTGTC-3', and antisense primer, 5'-GACAATGCCTGCCGGCCTTCCGTCC-3') were paired with wild-type 3 cDNA primers (sense primer, nucleotides 332-349: 5'-CCAGGTCACTCAAGTCAG-3'; and antisense primer, nucleotides 1799-1771:5'-GCAGGTGTCAGTAC-GCGTGGTACAGTTGC-3') to generate a 1467-base-pair (bp) cDNA fragment containing the T883C mutation. The mutated 332-1799 fragment was subcloned into pCRII and digested with the unique restriction enzymes NsiI and MluI, which produced a 966-bp mutated cartridge that was then ligated into NsiI- and MluI-digested wild-type 3-pBluescript SK. Finally, the reconstituted T883C 3 was excised from pBluescript SK with EcoRI and cloned into the mammalian expression vector pCDNA3 (Invitrogen). Sequence analysis of the T883C 3 construct was done to confirm the point mutation and proper insertion of the cartridge into the wild-type 3 cDNA. Mammalian cell transfection COS-7 cells were transfected with IIb- and 3-containing expression vectors in the presence of diethylaminoethyl-dextran as described previously.30 After 48 to 60 hours of culture, transfected COS-7 cells were surface labeled with biotin or metabolically labeled with sulfur 35 (S35)-methionine, as described previously.30 The labeled cells were then solubilized in lysis buffer (50 mmol/L Tris-hydrochloric acid [HCl] at pH 7.4, 150 mmol/L NaCl, 1% (vol/vol) Triton X-100, and 2 mmol/L PMSF) and mixed for 30 minutes at 4°C. The lysates were centrifuged at 1600g for 30 minutes at 4°C and the supernatants stored at 80°C. For stable cell transfections, wild-type- and Leu262Pro3-pcDNA3 plasmids were introduced into human embryonal kidney 293 cells in the presence of cationic lipid (Lipofectamine; GIBCO BRL). Transfected cells were selected in media containing 0.7 mg/mL neomycin. Pulse-chase metabolic labeling of COS-7 cells Forty-eight hours after transfection, COS-7 cells were incubated for 30 minutes in Dulbecco modified Eagle medium (DMEM; Sigma) without methionine, then pulsed with 2.22 × 107 Bq/10-cm plate of 35S-methionine (DuPont-NEN, Boston, MA) for 30 minutes at 37°C. After the pulse, the cells were washed 3 times with Dulbecco phosphate-buffered saline (D-PBS) containing 1 mg/mL cold methionine (Sigma). The chase was begun by incubating the cells in DMEM containing 1 mg/mL cold methionine for 1, 4, 8, and 22 hours at 37°C. Labeled cells were washed 5 times in D-PBS containing 1 mg/mL cold methionine and then solubilized as described above. Immunoprecipitation analysis Lysates were precleared by incubating with 1% (wt/vol) BSA, 10 µg of preimmune mouse IgG (Sigma), and 50 µL of a 50% slurry of protein A-Sepharose CL-4B (Pharmacia, Uppsala, Sweden) for 1 hour at room temperature with constant mixing. The beads were then separated by centrifugation at 1600g for 5 minutes, and the supernatant was incubated with 10 µg of mAb for 18 hours at 4°C, with mixing. Ten micrograms of rabbit antimouse IgG (DAKO, Carpinteria, CA) was added, and mixing was done for 1 hour at room temperature. Protein A-Sepharose (50 µL) was then added, and mixing continued for another hour at room temperature. The samples were then centrifuged at 1600g for 5 minutes and the supernatants discarded. The beads were washed 5 times in immunoprecipitation buffer (50 mmol/L Tris-HCl [pH 7.4], 150 mmol/L NaCl, and 1% (vol/vol) Triton X-100), added to 50 µL of 2 × reducing buffer (4% SDS, 10% -mercaptoethanol, 100 mmol/L Tris-HCl [pH 6.8], 10% glycerol, and 0.001% bromophenol blue), and boiled for 10 minutes. Samples were centrifuged at 1600g for 5 minutes and the supernatants analyzed by SDS-polyacrylamide gel electrophoresis. Biotin-labeled samples were transferred to PVDF membrane and detected with horseradish peroxidase-streptavidin and chemiluminescence reagents as described previously.30 Gels containing 35S-methionine-labeled samples were fixed in 45% methanol and 12% acetic acid for 30 minutes, incubated in Enlightening solution (DuPont-NEN) for 30 minutes, dried, and analyzed by autoradiography. Clot retraction by 293 cells expressing av3 Chromatographically purified fibrinogen I (AA) was a gift of Dr Michael W. Mosesson (Sinai-Samaritan Medical Center, Milwaukee, WI). Plasma from healthy subjects was depleted of fibronectin by passage through a gelatin-Sepharose column (Sigma) as described previously.24 Preliminary experiments (data not shown) demonstrated that untransfected 293 cells showed moderate retraction of clotted plasma but were unable to retract clots formed from fibronectin-depleted plasma. Therefore, fibronectin-depleted plasma was used in retraction assays to minimize the contribution of endogenous v1 and other fibronectin receptors. Cultured untransfected and transfected 293 cells were washed twice in D-PBS (Sigma), lifted with 0.01% (wt/vol) trypsin and 3.5 mmol/L EDTA, and washed twice again in D-PBS. The cells (10 × 106/mL) were then resuspended in modified Eagle medium containing Earle salts (MEM-ES; Sigma) and 0.1 mmol/L L-glutamine. For clot-retraction assays, 300 µL of the cell solution was added to 100 µL of medium containing 5 µg aprotinin (Sigma) and either 100 µL of fibronectin-depleted plasma or 100 µL of medium containing 100 µg of purified fibrinogen, in a 10- × 75-mm borosilicate glass tube (Fisher Scientific, Pittsburgh, PA). To initiate clot formation, 1 U of human  thrombin was added and the tubes were incubated at 37°C. The external dimensions of each clot were measured at 30-minute intervals, and a cylindrical model (volume = radius2 × height) was used to calculate the approximate clot volume. In some experiments, plasma or 293 cells were preincubated with peptides or mAbs for 15 minutes at 22°C. The peptides RGDW and RGEW and the fibrinogen peptide HHLGGAKQAGDV (H12), corresponding to -chain residues 400 to 411, were synthesized on a Pepsynthesizer (Model 9050; Millipore) with 9-fluoronylmethoxycarbonyl chemistry using standard manufacturers' procedures. Cell adhesion to immobilized substrates Adhesion of fluorescently labeled cells was preformed as described previously.8 Protein substrates or antibodies (2-10 µg/mL in D-PBS) were added to an Immulon 2 96-well plate (Dynatech, Chantilly, VA) and incubated overnight at 4°C. Before use, the coated plates were washed with D-PBS and blocked with 1% (wt/vol) BSA in D-PBS for 1 hour at 22°C. Transfected or control 293 cells were washed and trypsinized as described above, resuspended at 2 × 106 cells/mL in MEM-ES medium, and incubated with 2 µg/mL calcein AM (Molecular Probes, Eugene, OR) for 30 minutes at 37°C. After labeling, the cells were washed twice in D-PBS and resuspended in MEM-ES at 3 × 106/mL. One hundred microliters of labeled cell suspension was added to each well and allowed to adhere for 1 hour at 37°C. The total fluorescence per well was measured in a multiwell plate reader (Cytofluor; Perseptive Biosystems, Framingham, MA) immediately after incubation and 2 washes with MEM-ES to remove nonadherent cells.     Results Top Abstract Introduction Patients and methods Results Discussion Acknowledgments References The clinical history of patient LD indicated a marked hemostatic defect. Platelets from the patient did not aggregate in response to adenosine diphosphate, epinephrine, or arachidonic acid and showed only shape change in response to collagen, but they agglutinated with ristocetin (Figure 1). The failure of the platelets to support fibrinogen-dependent aggregation suggests a defect in platelet IIb3. View larger version (24K): [in this window] [in a new window]   Fig 1. Platelets from the proband (LD) do not support fibrinogen-dependent platelet aggregation. Platelet-rich plasma (PRP) from patient LD was stirred in an aggregometer at 37°C before the addition of agonists (time point indicated by arrow). Platelets from LD showed no aggregation in response to adenosine diphosphate (ADP), epinephrine (Epi), or arachidonic acid (AA) and showed shape change only in response to collagen (Coll). Platelets from LD agglutinated reversibly with high-dose ristocetin. On flow cytometry (Figure 2A), platelets from LD showed reduced (30% of normal) binding of subunit-specific mAbs against IIb or 3 but no binding of a complex-specific antibody (AP2). Total platelet levels of IIb3 were assessed by semiquantitative immunoblotting of platelet lysates (Figure 2B). In lysates from LD, faint but detectable bands of the expected apparent molecular weight were seen, corresponding to approximately 10% of control levels. These findings are consistent with a diagnosis of Type II GT. In both assays, platelets from the proband's father showed reduced (50% of normal) levels of platelet IIb3, findings consistent with a heterozygous defect that affects IIb3 expression. No platelet samples from the patient's mother were available. The failure of platelets from LD to aggregate implies a defect in binding of soluble fibrinogen. View larger version (40K): [in this window] [in a new window]   Fig 2. Platelets from LD express IIb and 3 subunits but little IIb3 complex. (A) Flow-cytometric analysis of surface IIb3. Monoclonal antibodies directed against a control glycoprotein, GPIb (AP1), the IIb3 complex (AP2), the 3 subunit (AP3), or the IIb subunit (TAB) were incubated with washed platelets from a healthy donor (control), the patient (LD), or the patient's father (F), and binding was assessed. Control platelet binding of nonimmune mouse IgG yielded a relative mean fluorescence of 2. (B) Western blot analysis of total platelet IIb3. The indicated amounts of Triton X-100-treated platelet lysates from a healthy control, the patient (LD), and the patient's father (F) were separated by sodium dodecyl sulfate (SDS)-7% polyacrylamide gel electrophoresis, transferred to a polyvinylidene fluoride membrane, and detected with rabbit polyclonal antibodies against IIb and 3. The positions of IIb (upper arrow) and 3 (lower arrow) are indicated. The ability of platelets from LD to interact with fibrin was assessed in a clot-retraction assay. Clotted PRP from LD (Figure 3) and both parents (data not shown) retracted as strongly as control plasma, indicating that the defective platelet IIb3 complex was still capable of interacting with fibrin. View larger version (48K): [in this window] [in a new window]   Fig 3. Platelets from LD retract a fibrin clot. PRP from a healthy control or patient LD was diluted with autologous plasma to the indicated platelet concentration and clotted by adding 8 mmol/L calcium chloride. After 1 hour of incubation at 37°C, clot retraction of platelets from LD was equivalent to that of control platelets. Genetic analysis of LD and her family (Figure 4) demonstrated that LD was a compound heterozygote for 2 novel mutations of the 3 gene in exon 5. First, in both LD and her father, a dinucleotide deletion of bases GC867 to GC868 was found. This deletion is predicted to result in a frameshift and substitution of a premature stop codon for amino acid 267. Second, LD and her mother had a point mutation of T883C, which is predicted to cause the substitution of a Pro for Leu at position 262. Sequencing of both alleles by subcloning of PCR-amplified genomic DNA confirmed that each of LD's parents carried 1 wild-type and 1 mutant 3 allele, whereas LD had inherited both exon 5 mutations. No mutations were found in the IIb gene of LD. The half-normal levels of IIb3 observed on platelets from LD's father (Figure 2) are consistent with failure to express the truncated D867-868 form of 3. This suggests that all the 3 present on platelets from LD carries the Leu262Pro mutation. View larger version (23K): [in this window] [in a new window]   Fig 4. LD is a compound heterozygote for 2 mutations of the 3 gene. Analysis of genomic DNA amplified by polymerase chain reaction identified a deletion of GC867-868 in LD's father (F) and a T883C substitution in her mother (M). Both mutant alleles were found in LD. The patient's siblings were not analyzed but are asymptomatic. Comparison of the human 3 amino acid sequence with 3 subunits from other species and other human  integrin subunits showed that the position corresponding to 3 262 was always occupied by a Leu residue (Figure 5A). The effect of a Leu262-to-Pro substitution on the predicted structure of 3 residues 200 to 300 was modeled by using a selection of protein structure algorithms (GeneWorks; Intelligenetics, Mountain View, CA). Compared with the wild type, the presence of a Pro at position 262 led to a reduction in hydrophobicity (by the Kyle-Doolittle hydrophobicity-scoring system), a reduced probability of an  helix but an increased probability of a  turn (by Garnier protein-structure analysis), and an increased probability of surface exposure (Figure 5B). Therefore, it is likely that the Pro2623 mutant adopts a local conformation different from that of Leu2623. View larger version (36K): [in this window] [in a new window]   Fig 5. Structural analysis of the Leu262Pro3 mutation. (A) Leucine (Leu) 262 is completely conserved among  subunits. The amino acid sequence of human 3 residues cystine (Cys) 232 to Cys273 is shown aligned with frog (Xenopus) and mouse 3 and with the sequences of 7 other human  subunits. Notably, the Leu residue at position 262 (in boldface) is invariant. The proposed intramolecular disulfide loop Cys232-27330 is shown. (B) Predicted hydrophobicity and secondary structure are shown for 3 amino acids 200 to 300 containing either wild-type Leu262 (black line) or mutant Pro262 (stippled line with arrow). The position of amino acid 262 is indicated by the vertical line. Mammalian cell-expression systems were used to confirm that the Leu262Pro form of IIb3 showed altered function. Leu262Pro3 cDNA was constructed and cotransfected with wild-type IIb into COS-7 cells. Biotin surface labeling of these transiently transfected cells showed that the mutant complex was poorly expressed and less stable than the wild type: from a detergent lysate, the complex-specific antibody AP2 immunoprecipitated only trace amounts of IIbLeu262Pro3 (data not shown). The 3 subunit-specific antibody, AP3, precipitated both subunits from lysates of wild-type IIb3 transfected cells but only the 3 subunit from cells transfected with IIbLeu262Pro3 (data not shown), resultsconsistent with the idea that the IIbLeu262Pro3 complex is easily dissociable. The addition of 2 mmol/L calcium to IIbLeu262Pro3 lysates did not preserve subunit association. Similarly, the IIb-specific antibody, Tab, captured only the IIb subunit from lysates of cells transfected with the mutant receptor (data not shown). Thus, the surface expression of IIbLeu262Pro3 in COS-7 cells recapitulated the platelet phenotype, with both subunits detectable but little or no binding of a complex-specific antibody. Metabolic labeling of transfected COS-7 cells with 35S-methionine was used to assess the biosynthesis of IIbLeu262Pro3. Normal synthesis and expression of IIb3 follows an ordered series of events31,32 in which cleavage of preIIb to mature IIb indicates that the nascent integrin complex has reached the Golgi and undergone posttranslational processing. Pulse-chase experiments were used to follow the intracellular maturation of IIb and either wild-type 3 or Leu262Pro3. As shown in Figure 6, immunoprecipitation of Leu262Pro3 by an anti-3 antibody captured preIIb, indicating that the mutant subunit had formed a heterodimer in the endoplasmic reticulum (ER). There was no apparent defect in the extent or rate of Leu262Pro3 synthesis. However, the maturation of the mutant complex was markedly delayed. In cells expressing IIb3, mature IIb was observed within 4 hours of the methionine pulse and cleavage was virtually complete by 22 hours. In contrast, preIIb associated with Leu262Pro3 was not cleaved before 22 hours, a finding consistent with a defect in heterodimer export from the ER. These results provide a likely molecular explanation for the reduced expression of IIbLeu262Pro3 observed in platelets from LD. View larger version (42K): [in this window] [in a new window]   Fig 6. Maturation of IIbLeu262Pro3 is delayed in COS-7 cells. COS-7 cells transfected with wild-type (WT) IIb and either WT 3 or Leu262Pro3 (LD) were pulsed with sulfur 35-methionine for 30 minutes and chased for up to 22 hours with medium containing 1 mg/mL cold methionine. Cell lysates prepared at the indicated times were immunoprecipitated with an anti-3 antibody (AP3) and analyzed by autoradiography of an SDS-7% polyacrylamide gel run under reducing conditions. Although preIIb and Leu262Pro3 synthesis and association are normal in lysates from LD, the appearance of a mature IIbLeu262Pro3 complex is markedly delayed. To study the effect of the Leu262Pro3 on ligand binding and specificity, the mutant 3 subunit alone was stably expressed in human embryonal kidney 293 cells. Previous studies using this cell line23 showed that transfected 3 is expressed on the cell surface in a complex with endogenous v and can mediate fibrin-clot retraction. A further advantage of this system is that v3, unlike transfected IIb3, does not require activation to bind ligand.24 The expression of v-Leu262Pro3 in 293 cells was assessed by flow cytometry and biotin surface labeling. As shown in Figure 7, the mutant complex was expressed less effectively than wild-type v3 (40%-50% levels), as expected, but intact heterodimer was detected on the cell surface by both AP3 and the complex-specific antibody, LM609. View larger version (42K): [in this window] [in a new window]   Fig 7. Leu262Pro3 is expressed in a complex with v in 293 cells. Untransfected (C) 293 cells or cells stably transfected with WT 3 or Leu262Pro3 (LD) were analyzed by flow cytometry. The mean fluorescence for each cell line is shown. The 2 left-hand panels show binding of a negative-control antibody (nonimmune mouse IgG) and a positive-control antibody (polyclonal anti-1). The 2 right-hand panels show the results of flow cytometry and immunoprecipitation of biotin surface-labeled cells with antibodies against 3 (AP3) or the v3 complex (LM609). The positions of v and 3 are indicated. The ability of vLeu262Pro3 cells to retract a fibrin clot was compared with that of untransfected and v3 cells (Figure 8). Wild-type v3 cells mediated rapid retraction of fibrin clots from both fibronectin-depleted plasma (Figure 8A) and purified fibrinogen (Figure 8B). Despite the reduced receptor density on their surface, vLeu262Pro3 cells showed nearly normal retraction of fibrin clots (Figure 8). In both v3 and vLeu262Pro3 cells, retraction could be enhanced by 0.5-mmol/L Mn2+ and inhibited by EDTA or ethylene glycol tetraacetic acid (data not shown), inhibited by RGDW but not by RGEW or fibrinogen -chain peptides, and inhibited by the complex-specific antibody, LM609 (Figure 8C). This suggests that a similar, cation-dependent mechanism underlies the interaction of both the wild-type and mutant integrin complex with an RGD-like site on fibrin. View larger version (26K): [in this window] [in a new window]   Fig 8. vLeu262Pro3 on 293 cells retracts a fibrin clot. Untransfected 293 cells (solid circles) and cells expressing WT v3 (solid squares) or vLeu262Pro3 (open circles) were assessed for their ability to retract clotted fibronectin-depleted plasma (A) or purified fibrinogen (B). The estimated clot volume is expressed as a function of time (minutes). The data shown are representative results from 3 separate experiments. vLeu262Pro3-mediated clot retraction was comparable to that of WT v3. (C) WT v3 cells (open columns) or vLeu262Pro3 cells (solid columns) were incubated for 15 minutes at 22°C with peptides RGDW, RGEW, or H12 (all at a final concentration of 2 mmol/L) or with the indicated antibodies (final concentration, 10 µg/mL). Only RGDW and the v3-specific antibody LM609 fully inhibited clot retraction. Finally, transfected 293 cells were assessed for their ability to bind purified fibrinogen and other ligands immobilized in a 96-well microtiter plate (Figure 9). Untransfected cells and 3-transfected lines bound equally to the control ligand fibronectin, suggesting a primary role for v1 and other endogenous integrins. The degree of cell spreading on fibronectin after adhesion was similar for all cell lines. Similarly, all 293 cell lines adhered to immobilized vitronectin (data not shown). As expected, both 3-transfected cell lines, but not untransfected cells, bound to a 3-specific antibody, AP3. However, whereas wild-type v3 cells showed significant adhesion and spreading on immobilized fibrinogen, the vLeu262Pro3 complex was unable to interact with immobilized fibrinogen. Therefore, the expression of a vLeu262Pro3 mutation in a cultured cell line recapitulated the ligand-binding specificity of IIbLeu262Pro3 on platelets in that it was unable to bind fibrinogen but showed nearly normal binding and retraction of a fibrin gel. View larger version (41K): [in this window] [in a new window]   Fig 9. vLeu262Pro3 does not support cell adhesion to immobilized fibrinogen. Untransfected 293 cells (open columns) and cells expressing WT v3 (hatched columns) or vLeu262Pro3 (solid columns) were labeled with calcein AM and allowed to adhere to immobilized substrates (bovine serum albumin, AP3, fibronectin, or fibrinogen) for 1 hour at 37°C. Adhesion is shown as the ratio of fluorescence per well before and after washing. Each column represents the mean (± SD) value from triplicate samples, and the results are representative of 4 separate experiments.     Discussion Top Abstract Introduction Patients and methods Results Discussion Acknowledgments References In this study, we characterized the molecular defects responsible for a novel case of GT. The patient's history and the absence of platelet aggregation in response to multiple agonists allowed the clinical diagnosis of GT. On the basis of our detection by flow cytometry and immunoblotting of platelet IIb3 levels that were 5% to 30% of normal values, we found this patient to have type II thrombasthenia. Although flow data with subunit-specific mAbs showed surface expression to be 30% of normal levels, there was no binding of the complex-specific mAb, AP2 (Figure 2A). This indicates that either the conformation of the mutant complex is abnormal or that the complex itself is unstable after surface expression. Such an unstable complex was described previously in another 3 point mutation (serine 162 to Leu), in which dissociation of the mutant IIb3 complex was demonstrated by abnormal migration through a sucrose gradient.33 Genotypic analysis of the patient's family identified 2 novel mutations of 3 only 16 bases apart in exon 5. The first was a dinucleotide deletion of bases 867 to 868 that resulted in a frameshift and substitution of a termination codon for amino acid residue 267. The 867-868 deletion was found in both the patient and her father and is predicted to result in the synthesis of a severely truncated form of 3 encoding only residues 1 to 267. </