Impaired platelet activation in familial high density lipoprotein deficiency (Tangier disease).

ATP binding cassette transporter A1 (ABCA1) is involved in regulation of intracellular lipid trafficking and export of cholesterol from cells to high density lipoproteins. ABCA1 defects cause Tangier disease, a disorder characterized by absence of high density lipoprotein and thrombocytopenia. In the present study we have demonstrated that ABCA1 is expressed in human platelets and that fibrinogen binding and CD62 surface expression in response to collagen and low concentrations of thrombin, but not to ADP, are defective in platelets from Tangier patients and ABCA1-deficient animals. The expression of platelet membrane receptors such as GPVI, alpha2beta1 integrin, and GPIIb/IIIa, the collagen-induced changes in phosphatidylserine and cholesterol distribution, and the collagen-induced signal transduction examined by phosphorylation of LAT and p72syk and by intracellular Ca2+ mobilization were unaltered in Tangier platelets. The electron microscopy of Tangier platelets revealed reduced numbers of dense bodies and the presence of giant granules typically encountered in platelets from Chediak-Higashi syndrome. Further studies demonstrated impaired release of dense body content in platelets from Tangier patients and ABCA1-deficient animals. In addition, Tangier platelets were characterized by defective surface exposure of dense body and lysosomal markers (CD63, LAMP-1, LAMP-2, CD68) during collagen- and thrombin-induced stimulation and by abnormally high lysosomal pH. We conclude that intact ABCA1 function is necessary for proper maturation of dense bodies in platelets. The impaired release of the content of dense bodies may explain the defective activation of Tangier platelets by collagen and low concentrations of thrombin, but not by ADP.

The ATP binding cassette (ABC) 1 transporter superfamily contains multispan transmembrane proteins that translocate a variety of substrates across extra-and intracellular membranes in an energy-dependent manner (1). Genetic defects in ABC transporters are associated with various disorders, including cystic fibrosis, retinal degeneration, defects of bile transport, and anemia (2).
ABCA1 is a member of the ABCA subfamily with high expression levels in hepatocytes, adrenal glands, liver, lung, intestine, placenta, and fetal tissues (3,4). Several functions have been attributed to ABCA1, including the engulfment of apoptotic cells, secretion of leaderless proteins, and transplasma membrane anion transport (5)(6)(7). More recently, work from our and other laboratories demonstrated ABCA1 mutations as the underlying defect in Tangier disease (familial an-alfalipoproteinemia) (8 -10). Tangier disease is a monogenic disorder characterized by the virtual absence of high density lipoprotein cholesterol, deposition of cholesteryl ester in cells of reticuloendothelial systems, and premature atherosclerosis in a subset of patients (11). In vivo models with targeted inactivation of ABCA1 demonstrated a pivotal role of this transporter in the trafficking of lipids, high density lipoprotein biogenesis, and overall cholesterol homeostasis as demonstrated by deposition of cholesterol in macrophages (12).
Thrombocytopenia is typically encountered in patients with Tangier disease (11). Furthermore, Tangier platelets have been occasionally reported to hypo-or hyper-respond to various agonists (13)(14)(15). ABCA1-deficient mice presented with mild hemorrhagic diathesis (16). In addition, abnormalities of Golgi apparatus and open canalicular systems were noted in platelets from these animals (16). These observations prompted us to systematically investigate the role of ABCA1 in platelet activation. We have provided evidence that ABCA1 is expressed in platelets and that its absence results in defective formation and release of dense bodies, followed by defective reactivity to collagen and low doses of thrombin.

EXPERIMENTAL PROCEDURES
Materials-Bovine serum albumin, fluorescein isothiocyanate (FITC)-isomer I on celite (10%), mouse IgG-FITC, ADP, and ␣-thrombin were from Sigma. Sepharose CL-2B was from Amersham Biosciences. Annexin V-FITC (FITC to protein (F/P) molar ratio 1.0) was from Bender Medical Systems. Human fibrinogen (von Willebrandt-factor (vWf) and plasminogen-free) was from Enzyme Research Laboratories (South Bend, IN). Monoclonal anti-CD62P and anti-CD63 antibodies and FITC-labeled antibodies against CD29 and GPIIb/IIIa were from Coulter-Immunotech Diagnostics. Monoclonal antibodies recognizing CD68, LAMP1, and LAMP2 were from BD Biosciences. The monoclonal antibody against ␣ 2 integrin clone Ab-1 was from Merck Biosciences. Phospho-specific antibodies against LAT and p72 syk were from Cell * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Patients-All studies were performed with blood donors who gave informed consent. Three male patients with Tangier disease were examined in this study: SB (56 years old), NA (brother of SB, 53 years old), and AF (65 years old). All patients presented with dyslipidemia, including total cholesterol Ͻ100 mg/ml, high density lipoprotein cholesterol Ͻ2 mg/ml, and mild hypertriglyceridemia. Patient NA presented with splenomegaly, patient SB suffered from peripheral neuropathy, and patient AF from coronary heart disease. Moderate thrombocytopenia was present in all three patients (SB: 107,000/l; NA: 97,000/l; AF: 104,000/l). Prolonged in vitro bleeding time (PFA-100; DADE-Behring) in collagen/epinephrine stimulation was observed in patients SB and NA. Coagulation parameters (PT, APTT, TT) were normal in all patients. The defect of ABCA1 was characterized on the molecular level in all patients. In patient AF, a 1-bp deletion in exon 14 led to a stop codon at amino acid position 635, resulting in the deletion of most of the protein sequence, including both ATP cassettes. In patients SB and NA, large chromosomal deletions starting in intron 39 led to a partial deletion of the second ATP cassette and the cytoplasmic C-terminal fragment. For controls, samples were obtained from age-matched, male, healthy volunteers. All subjects included in the study did not take any medication that could affect platelet function for at least 2 weeks before the study.
ABCA1 "Knock-out" Animals-Gene targeting of ABCA1 was performed using B6 embryonic stem cells according to standard procedures. ABCA1 genomic sequences were isolated from a -Fix mouse genomic library (provided by Stratagene) made of leukocyte DNA from mouse strain C57BL/6. A neo R -gene was introduced into exon 2 of the gene by cloning an ϳ4.4-kb DNA fragment containing part of exon 2, intron 2, exon 3, and part of intron 3 into the NotI-XhoI site of the vector pPNT (left arm). A 2.6-kb DNA fragment extending from the KpnI site located upstream of exon 2 to the middle part of exon 2 was cloned into the KpnI-BamH1 sites of the vector pPNT (right arm). After electroporation followed by two-step selection with G418 and ganciclovir, four clones showing homologous recombination were identified. Three of these were used to generate chimeric mice that were subsequently crossed with C57Bl/6 mice, resulting in germ line transmission of the targeted allele.
Flow Cytometric Analysis of Fibrinogen Binding or Antigen Expression on Platelet Surface-Flow cytometric analysis of fibrinogen binding or antigen expression on platelet surface was performed exactly as described previously (18,19). Briefly, human fibrinogen was conjugated with FITC using the FITC-celite method according to Xia et al. (20). Platelet suspensions (0.2 ml) preincubated for 3 min at room temperature with 0.15 mg/ml fibrinogen-FITC were added to 20 l of a solution containing increasing concentrations of collagen, thrombin, or ADP. The reaction was stopped after 180 s by fixation in 1% formaldehyde for 30 min. For antigen expression analysis, FITC-coupled monoclonal antibodies recognizing CD62, CD63, CD68, LAMP1, or LAMP2 were added after fixation for 1 h at room temperature. The concentrations required for saturated binding of antibodies were determined beforehand.
Platelet Aggregation-Platelet aggregation was monitored as described by Born (21). Platelet-rich plasma (PRP) was prepared by centrifugation at 250 ϫ g for 10 min at room temperature. The platelet count was adjusted with autologous plasma. Aggregation from PRP platelets was monitored by assessing light transmission in an aggregometer (Lumitec), with continuous stirring at 1100 rpm at 37°C.
Assay for Inward Movement of Phosphatidylserine (PS) and Cholesterol-Fluorescent PS and cholesterol analogs were inserted into the plasma membrane of platelets as described by Sleight and Pagano (22). Briefly, NBD-PS or NBD-cholesterol dissolved in chloroform was mixed with egg yolk phosphatidylcholine in a molar ratio 1:3, transferred to a glass tube, dried under nitrogen, and solubilized in absolute ethanol. The solution (1% of the final volume) was injected under vortexing into phosphate-buffered saline containing 3.5% albumin (v/v). The final concentration of fluorescent analogs in the labeling solution was 20 mol/liter. Platelets were suspended in the labeling solution (10 8 cells/ ml), incubated for 1 h at 4°C in the dark, centrifuged, and resuspended in buffer A. For determination of the inward transport of NBD-PS, platelets were incubated for 30 min in the dark at 2 and 37°C, the former temperature corresponding to the passive and the latter to the active transmembranous transport. For determination of the inward transport of NBD-cholesterol, platelets were incubated for 2 min with the agonist. The amount of internalized fluorescent lipid analogues was determined by the selective reduction of NBD-lipids on the outer monolayer by dithionite according to McIntyre and Sleight (23). Dithionite was added to the cell suspension to the final concentration of 5 mmol/ liter, and the decrease of fluorescence reflecting the amount of NBD analogues remaining in the outer monolayer was monitored at 540 nm (excitation 480 nm) with a Hitachi 2000 fluorescence spectrophotometer. Thereafter, platelets were permeabilized with 5% digitonine, and the further decrease of fluorescence was monitored. This decrease reflected the amount of fluorescent analogues taken up by the cell during active inward transport. The amount of NBD-lipids residing inside the cells was expressed as percentage of total fluorescent labeling.
Analysis of Dense Granule Release with Mepacrine-Gel-filtered platelets were diluted to 2.5 ϫ 10 7 platelets/ml in buffer A. 0.8 ml of platelet suspension was incubated for 20 min with 5 mol/liter mepacrine. Platelets were activated with 50 l of solution containing increasing concentrations of thrombin and collagen. The reactions were stopped after 180 s by fixation with 1% formaldehyde. Platelets were washed and analyzed in a flow cytometer.
Estimation of pH in Lysosomal Compartment-Gel-filtered platelets were diluted to 2.5 ϫ 10 7 platelets/ml in buffer A and incubated for 30 min with LysoSensor Green DND-189. The fluorescence of platelets was then analyzed by flow cytometry.
Determination of the Intracellular Ca 2ϩ Concentration-Intracellular Ca 2ϩ measurements were performed using the Ca 2ϩ -sensitive fluorescence probe Fura-2/AM according to established methods (24).
Western Blotting-Western blotting of proteins from cell lysates was performed exactly as described previously (25). Loading controls were performed with an antibody against a ubiquitously expressed protein (␣-actin).
Electron Microscopy of Platelets-Electron microscopy of platelets was performed exactly as described previously (26).

ABCA1 Expression in Human
Platelets-Initially we examined whether ABCA1 is expressed in human platelets. As shown in Fig. 1A, ABCA1 was recognized in platelet extracts by a specific anti-ABCA1 polyclonal antibody. The amount of ABCA1 present in platelets appeared to be comparable with that in macrophages and fibroblasts stimulated with liver X receptor (LXR)/retinoid X receptor (RXR) agonist and higher than in unstimulated fibroblasts. No ABCA1 expression could be observed in platelets from Tangier patients (Fig. 1A). The expression of ABCA1 in normal platelets was further confirmed by flow cytometry (Fig. 1B, left panel). The binding of the anti-ABCA1 antibody to the cell surface was concentration-dependent and exhibited saturation at higher concentrations, indicating that, similar to nucleated cells, ABCA1 is present in plasma membranes of platelets. By contrast, anti-ABCA1 binding to the cell surface was much reduced in platelets from Tangier patients (Fig. 1B, right panel).
Agonist-induced Activation of Tangier Platelets-To assess platelet activation in ABCA1 deficiency, we examined the agonist-induced increase in the number of bound fibrinogen molecules and the exposure of CD62 on the platelet surface. These parameters reflect the rate of platelet aggregation and the extent of platelet secretory responses, respectively. As shown in Fig. 2, stimulation of control platelets with low concentrations of collagen, thrombin, and ADP led to an abrupt increase in fibrinogen-FITC and anti-CD62 binding. Both responses reached a plateau at higher agonist concentrations. ABCA1deficient platelets were markedly less responsive to collagen across the whole agonist concentration range, both with respect to fibrinogen-FITC binding and CD62 surface exposure. Similarly, thrombin-induced fibrinogen-FITC binding and CD62 surface exposure were reduced in Tangier platelets stimulated with low concentrations of thrombin. However, higher thrombin concentrations led to comparable responses both in control and Tangier platelets. No significant differences in fibrinogen-FITC binding and CD62 surface exposure were found between control and Tangier platelets stimulated with ADP.
To prove unequivocally whether collagen-induced platelet activation is defective in ABCA1 deficiency, we extended our study to platelets from ABCA1 knock-out animals. As shown in Fig. 3, the collagen-induced fibrinogen-FITC binding and CD62 surface exposure were much reduced in ABCA1-deficient platelets. However, both responses to thrombin were comparable in ABCA1-deficient and control platelets.
Agonist-induced Aggregation in Tangier Platelets-The impaired fibrinogen binding and CD62 exposure in response to collagen observed in ABCA1-deficient platelets would be expected to translate into differences in platelet aggregation. Thus, we next compared collagen-induced aggregation in Tangier and normal platelets. As shown in Fig. 4, both the extent and the rate of platelet aggregation were severely impaired in Tangier platelets stimulated with 0.5 or 1.0 g/ml collagen.
Expression of Membrane Receptors in Tangier Platelets-Abnormal expression of platelet membrane receptors could account for the inadequate responses to agonists seen in ABCA1deficient platelets. Therefore, we next examined the expression of collagen receptors GPVI and integrin ␣ 2 ␤ 1 as well as fibrinogen receptor GPIIb-IIIa in platelets from Tangier patients and control subjects. No significant differences in the surface expression of integrin ␣ 2 ␤ 1 , GPIIb-IIIa, and GPVI were noticed between platelets from persons with and without Tangier disease.
Phospholipid and Cholesterol Translocation in Tangier Platelets-ABCA1 has been postulated to act as phospholipid and/or cholesterol translocase. Because dysregulation of lipid distribution could account for the impaired response of ABCA1deficient platelets to collagen, we next examined the translocation of PS and cholesterol across the plasma membrane in agonist-stimulated platelets. No differences in the exposure of negatively charged phospholipids on gel-filtered platelets as determined by annexin V-FITC binding could be observed between control and Tangier platelets stimulated with increasing concentrations of collagen or A23187, a calcium ionophore. To follow the inward PS movement across plasma membranes, platelets were loaded with NBD-PS and incubated for 30 min at 37 or 2°C to sustain or inhibit active NBD-PS transport, respectively. As expected, substantially less NBD-PS was accessible to dithionite in platelets incubated at 37°C, indicating that that portion of NBD-PS was translocated from the exo-to the endofacial leaflet of cell membrane. However, the redistribution of NBD-PS within platelet membranes was comparable in Tangier and control platelets (not shown). To investigate the agonist-induced translocation of cholesterol, platelets loaded with NBD-cholesterol were stimulated with collagen for 2 min, and than NBD-associated fluorescence was quenched with dithionite. Similar amounts of NBD-cholesterol were moved out of the outer monolayer in control and Tangier platelets (not shown). Collectively, these results show that phosphatidylserine and cholesterol translocation mechanisms are not affected by ABCA1 deficiency.
Collagen-induced Signaling in Tangier Platelets-Collagen- induced platelet activation is preceded by induction of intracellular signaling pathways. To investigate whether signal transduction by collagen is impaired in Tangier platelets, several signaling events located both proximal and distal to collagen receptor activation were examined. The time course of LAT and p72 syk phosphorylation was similar in collagen-activated platelets from Tangier and control platelets (not shown). Furthermore, no differences between Tangier and control platelets could be observed with respect to collagen-induced calcium mobilization (not shown). Recent studies revealed that expression of several small G proteins with a potential to regulate platelet activation is dysregulated in cells from Tangier disease (27). However, no major differences in the expression levels of Cdc42 and Rac1 could be observed between Tangier and normal platelets in this study (not shown).
Electron Microscopy of Tangier Platelets-To gain further insight into the origins of the activation defect associated with ABCA1 deficiency, ultrastructural studies were performed. Fig. 5 shows platelet sections obtained from one patient with Tangier disease and a control subject. It is evident that several structures that are typical for platelet morphology, including ␣-granules, microtubules, the dense tubular system, and the surface-connected canalicular system, were normal in Tangier cells. A distinct feature of Tangier platelets was a relatively low number of dense bodies. In addition, unusually large and electron-dense structures closely resembling the giant granules typically encountered in platelets from Chediak-Higashi syndrome were noted.
Function of Dense Bodies in Tangier Platelets-The presence of reduced and/or irregular dense bodies (␦-granules) as a result of ABCA1 deficiency could account for the defective platelet activation. Therefore, we attempted to more closely characterize the function of dense bodies in Tangier platelets. To this end, platelets were loaded with mepacrine, a fluorescent dye known to accumulate specifically in ␦-granules. As shown in Fig. 6, A and B, when stimulated with increasing concentrations of collagen and thrombin, Tangier platelets released mepacrine to a much lesser extent than their normal counterparts. To further characterize the release of dense granules from platelets upon agonist stimulation, we determined the surface exposure of CD63, a glycoprotein regarded as a specific marker of organelles of lysosomal origin (i.e. ␦and -granules). As shown in Fig. 6, C and D, activation of control platelets led to a concentration-dependent increase in anti-CD63 binding to platelet surfaces in collagen-and thrombin-stimulated platelets. By contrast, CD63 surface exposure was severely impaired in Tangier platelets stimulated with collagen and reduced when thrombin was used as an agonist. Similar to Tangier platelets, the impaired release of mepacrine could also be ob-served in collagen-stimulated platelets from ABCA1-deficient animals (Fig. 6E). The functional abnormality of lysosomal granules in Tangier platelets was further documented by estimating their pH with LysoSensor, a fluorescent pH indicator that specifically partitions into acidic organelles. As shown in Fig. 6F, Tangier platelets incubated with LysoSensor displayed Open symbols, Tangier patients; closed symbols, controls. C and D, platelet-rich plasma was treated with increasing concentrations of collagen (C) or thrombin (D). Platelets were fixed and incubated with FITC-labeled monoclonal antibodies against CD63. Fluorescence of 5000 platelets/measuring point was determined by flow cytometry. Data are the mean Ϯ S.D. from at least three independent experiments with platelets from different donors. Open symbols, Tangier patients; closed symbols, controls. E, platelets from ABCA1-deficient and control animals were loaded with mepacrine as described under "Experimental Procedures" and stimulated with collagen (10 g/ml). Fluorescence of 5000 platelets/measuring point was determined by flow cytometry. Data are the mean Ϯ of the range from two independent determinations. F, platelets were loaded with LysoSensor Green DND-189 as described under "Experimental Procedures." Fluorescence of 5000 platelets was determined by flow cytometry. Shown are histograms representative for one of three separate determinations in platelets from different donors. much lower fluorescence intensity than normal platelets, indicating a higher pH value of ␦and -granules and other organelles of lysosomal origin.
Agonist-induced Lysosome Release in Tangier Platelets-In addition to dense bodies, lysosomes (-granules) are also released from platelets upon activation with agonists. By analogy to ␦-granules, we were interested whether ABCA1 deficiency is associated with defective release of lysosomes from platelets. Therefore, we examined the surface expression of several lysosomal markers in platelets stimulated with thrombin and collagen. As shown in Fig. 7, both agonists induced concentrationdependent binding of antibodies against LAMP1, LAMP2, and CD68 to the platelet surface. By contrast, the exposure of all lysosomal markers was severely impaired in Tangier platelets. The defective externalization of lysosomal markers was seen both in collagen-and thrombin-activated platelets.

DISCUSSION
The present study has demonstrated for the first time that ABCA1 is expressed in human platelets. The amount of ABCA1 in platelets as estimated by Western blotting was comparable with that found in cells known to efficiently export cholesterol. In addition, we found that, similar to nucleated cells, ABCA1 is present on the surface of platelets. These observations are consistent with the notion that ABCA1 may be involved in the regulation of platelet function, which prompted us to investigate systematically agonist-induced platelet activation in ABCA1 deficiency. We have demonstrated that platelets from both Tangier patients and ABCA1 knock-out animals are impaired in their ability to respond to collagen and, to a much lesser extent, to thrombin stimulation. By contrast, ADP-induced activation remained intact in platelets from Tangier patients.
Previous investigations of platelet activation in Tangier disease led to conflicting results. Similar to our findings, Harmon et al. (15) observed an impaired activation of platelets in the presence of low, but not high, concentrations of thrombin. Shastri et al. (13) found normal platelet activation by collagen but an impaired response to ADP and epinephrine. In contrast to these studies, hyper-responsiveness to aggregating agents and increased thromboxane B 2 production were reported by Vergani et al. (14). Although reasons for these discrepancies remain unknown, it must be stressed that the present study is the first one in which platelet function was tested in humans or animals with defined molecular defects. Phenotypes closely resembling Tangier disease are known to be caused by defects of proteins other than ABCA1 (28). Thus, it cannot be excluded that some contradictory results reported previously were a consequence of misidentification of Tangier subjects. Alternatively, the hypo-or hyper-responsiveness of Tangier platelets to various agonists might reflect the wide diversity of molecular defects encountered within ABCA1 that may have inhibiting or activating influences on platelet function.
It is generally assumed that ABCA1 functions as a translocase involved in the movement of phospholipids such as phosphatidylserine and/or cholesterol across the plasma membrane (29). The distribution of lipids between both monolayers of the plasma membrane undergoes tight regulation in platelets; its perturbation may result in an impaired platelet activation. For instance, the defect of the transmembranous PS transport seen in platelets from Scott syndrome is associated with defective signaling from collagen receptors, abnormal platelet function, and prolonged bleeding time (30,31). Therefore, we considered the possibility that the defective response of Tangier platelets to collagen stimulation might be related to a distorted plasma membrane asymmetry followed by a dysregulation of intracellular signaling. However, we failed to observe any changes in PS externalization or the inward cholesterol translocation in agonist-stimulated Tangier platelets as compared with normal platelets. Neither was the inward PS movement disturbed in these cells. In addition, treatment of Tangier platelets with collagen initiated regular signal transduction, both with respect to signaling events proximal (phosphorylation of LAT and p72 syk ) and distal (Ca 2ϩ mobilization) to collagen receptors. Cdc42 and Rac1 are important elements of collagen-induced signaling in platelets, and subnormal expression of Cdc42 in cells from patients with Tangier disease has been reported recently (27). However, normal small G-protein levels were found in Tangier platelets. Finally, normal expression levels of collagen receptors, integrin ␣ 2 ␤ 1 , and GPVI were observed in Tangier platelets. Taken together, our results argue strongly against the notion that the impaired platelet activation by collagen in Tangier disease results from abnormal lipid distribution in plasma membrane and/or dysregulation of collagen receptor-dependent intracellular signaling.
The functional impairment of dense bodies is a distinguishing feature of platelets in ABCA1 deficiency. It is well established that liberation of several potent agonists, including ADP and serotonin, from dense granules in response to agonists provides a major contribution to platelet activation. Compared with other agonists, collagen appears to have particularly strong dependence upon the co-secretion of ADP for full activation. Several studies demonstrated suppression of collageninduced platelet aggregation and fibrinogen binding in the presence of apyrase, an enzyme that specifically destroys ADP (32,33). In addition, collagen-induced platelet reactivity could be completely blocked by a soluble form of CD39, an endothelial ecto-enzyme with ADPase and ATPase activity (34). Chediak-Higashi platelets, which similar to Tangier platelets have a decreased number of dense granules, are characterized by defective aggregation and fibrinogen binding to collagen but not to ADP (35). In light of these observations, it seems reasonable to assume that the impaired collagen-induced platelet reactivity in ABCA1 deficiency is a consequence of the defective dense body function and decreased liberation of agonists from ␦-granules during activation.
The presence of reduced dense bodies and giant granules in Tangier platelets suggests a defect in the biogenesis of lysosome-related organelles in these cells. Similar giant granules seen in Chediak-Higashi syndrome likely arise as a consequence of impaired fission/fusion events that determine the size of lysosomes and related organelles (36,37). By analogy, defective vesicle fusion could arise in Tangier platelets as a consequence of ABCA1 deficiency. The results of recent studies on cellular localization and trafficking of ABCA1 provide some evidence to support this contention. In a fibroblast cell line, ABCA1 was shown to reside on the surface of both small and large vesicles likely representing early and late endosomes, respectively (38). The latter compartment is a direct precursor of lysosomes and is equivalent to lamellar bodies in platelets, where ␦-granules are thought to be sorted (39). Furthermore, ABCA1-containing early endosomes were found to undergo continuous fusion and fission events and to interact transiently with late endocytic vesicles and with the plasma membrane (38). Thus, it is seems likely that ABCA1 is involved in endosomal sorting and, as a consequence, ABCA1 defects might result in distorted maturation of lysosomes and/or lysosomerelated organelles. Actually, large and morphologically abnormal lysosomes were found in macrophages from Tangier patients (40). In addition, impaired activity of lysosomal enzymes (i.e. phospholipase A1) because of abnormally high pH was observed in Tangier fibroblasts. 2 Taken together, these observations support the hypothesis that the presence of reduced dense bodies and giant granules in Tangier platelets is a consequence of defective ABCA1-mediated regulation of endosomal sorting and distorted ␦-granule maturation.
In conclusion, ABCA1 is expressed in human platelets, and its proper function appears to be necessary for normal ␦-granule maturation and release. Consequently, ABCA1-deficient platelets are defective in their ability to respond to collagen and low concentrations of thrombin, but not to ADP.