Cocaine-Induced Platelet Defects
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
1352
Cocaine-Induced Platelet Defects
Lisa K. Jennings, PhD; Melanie M. White, BS; Curtis M. Sauer, MD;
Alvin M. Mauer, MD; James T. Robertson, MD
Background and Purpose: Numerous studies have demonstrated an association between acute cardiac
events, cerebrovascular accidents, and cocaine use. The underlying mechanisms leading to these
complications have not been well defined. Using various in vitro model systems, it has been reported that
cocaine, up to or greater than an order of magnitude of the lethal dose, causes either inhibitory or
proaggregatory effects on platelet function.
Methods: To address these reported discrepancies, we examined the effect of cocaine and its carrier on
the activation and aggregation of human platelets in vitro.
Results: We found that cocaine inhibited platelet aggregation when platelets were challenged with ADP,
collagen, or arachidonic acid. This inhibition was due to a direct effect on fibrinogen binding to the
activated platelet. Cocaine also caused the dissociation of preformed platelet aggregates. At these same
concentrations, cocaine did not inhibit agonist-mediated increases in cytosolic calcium or inhibit platelet
shape change, suggesting that its effect on platelet aggregation was a selective process and not due to a
total destruction of platelet function. Interestingly, the organization of the cytoskeleton of activated
platelets, a secondary event critical to cell receptor clustering and clot retraction, was disrupted by cocaine
treatment. In addition, alterations in platelet protein electrophoretic patterns were observed on
preincubation of platelets with cocaine.
Conclusions: We conclude that cocaine may have a direct inhibitory effect on the ability of platelets to
participate in thrombus formation. The contribution of this effect as an underlying mechanism of sudden
death in cocaine abusers is unknown. (Stroke. 1993;24:1352-1359.)
KEY WoRDS * cocaine * platelet activation * platelet aggregation
C oincident with the recent dramatic increase in platelet adhesion to foreign surfaces and platelet aggre-
cocaine abuse in the United States, an increasing gation initiated by ADP and thrombin. Additional stud-
number of strokes temporally related to cocaine ies have demonstrated that exposure of blood platelets
use have been reported.' In most but not all series, the to tertiary amine local anesthetics such as dibucaine,
majority of these strokes were hemorrhagic.1-4 Underlying procaine, and tetracaine results in the inhibition of
vascular lesions, especially aneurysm and arteriovenous platelet activation and aggregation.6 18 Treatment of
malformation, have been identified in three fourths and blood platelets with dibucaine caused decreased fibrin-
one half of the cases of cocaine-related subarachnoid and ogen receptor exposure on the activated platelet and
intracerebral hemorrhages, respectively.3 It is in those decreased calcium uptake and exchange across the
cases that lack an identifiable etiology that cocaine may platelet membrane.15 Feinstein et a19 reported that local
have been directly responsible for the stroke. Further anesthetics directly inhibited aggregation, and this inhi-
strengthening this association between cocaine and cere- bition was partially antagonized by calcium; the platelet
bral hemorrhage are cases in which bleeding occurred in release reaction was also blocked by local anesthetics
multiple and unusual locations.3,5 Postulated mechanisms independent of external calcium concentrations or the
for cocaine-related cerebral hemorrhage include acute aggregation response.
hypertension and profound cerebral vasoconstriction fol- With regard to cocaine, it has been demonstrated
lowed by reperfusion.235 These mechanisms, however, that, in addition to human platelet aggregation, rabbit
remain unproven. Coagulation disorders, especially those platelet aggregation induced by ADP and thrombin was
related to platelet dysfunction after cocaine use, have, inhibited by pretreatment with cocaine.19 On the other
until recently, received little attention. hand, it has been shown that cocaine enhances the
O'Brien and Boullin6 were the first to report that response of rabbit platelets to arachidonic acid, possibly
local anesthetics at various concentrations inhibited potentiating thromboxane generation and platelet ag-
gregation.20 In view of these conflicting reports, a criti-
Received December 4, 1992; revision received April 14, 1993; cal analysis of the effect of various concentrations of
accepted April 30, 1993. cocaine on human platelet activation and aggregation is
From the Departments of Medicine (L.K.J., A.M.M.), Biochem- needed. We found that cocaine, up to an order of
istry (L.K.J.), Neurosurgery (M.M.W., J.T.R.), and Neurology magnitude greater than predicted systemic concentra-
(C.M.S.), the University of Tennessee and Baptist Health Care tions of lethal doses, inhibited human platelet aggrega-
System, Memphis.
Correspondence to Lisa K. Jennings, PhD, Department of tion induced by the agonists ADP, collagen, and arachi-
Medicine, Division of Hematology/Oncology, 956 Court Ave, donic acid and also caused disaggregation of platelets.
Room A303, Memphis, TN 38163. Interestingly, at these same concentrations cocaine had
Downloaded from http://stroke.ahajournals.org/ by guest on October 21, 2015Jennings et al Cocaine-Induced Platelet Defects 1353
no measurable effect on increases in platelet cytosolic
calcium, a primary event associated with platelet
activation.
Materials and Methods
Materials
We used the following reagents: cocaine hydrochlo-
ride (10%, Roxane Laboratories, Inc, Columbus, Ohio);
carrier for cocaine (Roxane Labs); ADP (Sigma Chem-
ical Co, St Louis, Mo); collagen (Hormon-Chemie,
Munich, Germany); arachidonic acid (Bio Data Corp,
Hatboro, Pa); indo-1 AM (Molecular Probes, Inc, Eu- O
gene, Ore); Coulter Immunoprep (Hialeah, Fla); anti-
human fibrinogen, fluorescein conjugate, fluorescein W0. 0.1 0.2 0.3
z
isothiocyanate (FITC) (TAGO, Inc, Burlingame, Calif);
Sequester-Sol (Cambridge Chemical Products, Inc, De-
troit, Mich); an anti-glycoprotein (GP) lIb-llla polyclo- w
80--
0.
nal antibody (Dr D.R. Phillips, COR Therapeutics, 60
South San Francisco, Calif). (-9-
Methods (0--
Platelet preparation. Blood was drawn into a plastic 20
syringe and mixed in plastic centrifuge tubes with 0.129
mol/L buffered citrate anticoagulant at a ratio of 9 parts
a 100
blood to 1 part anticoagulant. Platelet-rich plasma (PRP)
was obtained by centrifugation at 135g for 15 minutes at
room temperature. The PRP was diluted to 2.5 x 108/mL
with autologous platelet-poor plasma (PPP).
Platelet aggregation. Aggregation was performed using
a Payton Lumiaggregometer (Buffalo, NY) as previ-
ously described.21 To determine the effect of cocaine on
platelet aggregation, 450 ,L PRP was preincubated for
2 minutes with 50 ,uL of various concentrations of PERCENT COCAINE
cocaine (diluted in carrier) or carrier control. Agonist (5
,uL of either ADP, collagen, or arachidonic acid) was FIG 1. Line graphs show effect of cocaine on ADP- and
added, and aggregation was monitored for 5 minutes or collagen-induced platelet aggregation. A, Platelets (2.5 X108/
until maximum aggregation response was observed, mL) in plasma were incubated with increasing concentrations
whichever occurred first. Aggregation was measured as of cocaine for 2 minutes and then treated with 10 ,umol/L
percent change in light transmission. ADP. Aggregation response was measured at S minutes after
To examine the effect of cocaine on platelet aggregate agonist addition or until a maxcimum response was observed,
stability, 50 ,uL of agonist was added to 450 ,uL PRP, whichever occurred first. Data are presented as mean -+-SEM
and aggregation was measured for 3 minutes. At this (n=4). B, Platelets (2.5 x Iff/mL) in plasma were incubated
point varying concentrations of cocaine or carrier con- with increasing concentrations of cocaine for 2 minutes and
trol were added, and the effect was monitored for an then treated with 2 1£glmL collagen. Aggregation response was
additional 5 minutes. Disaggregation was measured as measured at 5 minutes after agonist addition or until a
percent change in light transmission from maximal maxcimum response was observed, whichever occurred first.
aggregation to maximum disaggregation. Data are presented as mean +.SEM (n =4).
Fibrinogen binding. To measure fibrinogen binding to
activated versus resting platelets, 15 ,L of 100 ,umol/L
ADP or saline was added to 150 ,uL PRP (7.5 x 107/mL) minutes before the addition of anti-fibrinogen antibody.
and incubated at room temperature for 10 minutes. Samples were processed as described for the fibrinogen-
Next, 15 ,uL of anti-fibrinogen antibody (FITC conju- binding studies.
gated) was added and the mixture incubated for an Flow cytometric analysis. Flow cytometric analysis of
additional 20 minutes.22 After this period, the specimen platelets was performed as previously described.23 A
was prepared for flow cytometry by adding 10 4L of the Coulter EPICS 753 flow cytometer with laser adjusted
above mixture to 75 ,L phosphate-buffered saline to deliver 800 mW at 488 nm was used to analyze
(PBS), 125 ,L solution B, and 50 ,L solution C platelets. Cells in suspension were analyzed at a flow
(Coulter Immunoprep). The sample was stored at 4°C rate of approximately 400 cells per second. The forward
until analyzed. For cocaine inhibition studies, 0.2% angle light scatter and fluorescence signals from each
cocaine, carrier, or buffer was added to the PRP and cell were collected and stored by the Coulter MDADS
incubated for 5 minutes before the addition of agonist. minicomputer analysis system as two-parameter 64x64
For fibrinogen dissociation studies, 0.4% cocaine, car- channel histograms. Routinely, fluorescence signals
rier, or buffer was added after the 10-minute incubation were collected from that portion of the light scatter
of PRP with agonist and allowed to incubate for 5 profile representing single platelets so as to exclude
Downloaded from http://stroke.ahajournals.org/ by guest on October 21, 20151354 Stroke Vol 24, No 9 September 1993
TABLE 1. Effect of Cocaine on Arachidonic Acid-Mediated
Platelet Aggregation
Percent
Treatment Aggregation
100 ,ug/mL AAJennings et al Cocaine-Induced Platelet Defects 1355
TABLE 2. Fibrinogen Binding to Carrier- and not shown). Behnke7 previously showed that platelets
Cocaine-Treated Platelets treated with cocaine at a concentration of 3000 ,ug/mL
Percentage of Total (0.3%) inhibited shape change in response to 1 ,ug/mL
Platelets With Bound ADP. Thus, it is possible that higher concentrations of
Treatment Fibrinogen anesthetics are required to inhibit this response.
None 3.0±2.3 Previous reports have suggested that low-dose co-
Carrier 5.5±2.7 caine (0.005% to 0.05%) enhances the response of
Cocaine 1.4±0.5 rabbit platelets to nonaggregating levels of arachidonic
ADP 83.6±7.4 acid, thus leading to increased platelet reactivity rather
Carrier+ADP 84.0±6.1
than inhibition.20 When we pretreated platelets with
0.2% cocaine, the aggregation response to 500 ,lg/mL
Cocaine+ADP 4.8±1.5 arachidonic acid was totally inhibited (Table 1). In
ADP+carrier 80.8±6.4 addition, cocaine was able to induce disaggregation of
ADP+cocaine 1.6±0.5 preformed platelet aggregates (data not shown). Since
Values are mean+±SD; n=5. ADP concentration, 100 ,umol/L. we observed inhibitory effects of cocaine similar to
those seen when platelets were stimulated with ADP
and collagen, we investigated the reported enhanced
immunoglobulin G horseradish peroxidase conjugate platelet response to subthreshold levels of arachidonic
for 2 hours. The sheets were washed four times in acid in the presence of cocaine (Table 1). First, we
immune incubation buffer and developed by 4-chloro- established our subthreshold level of arachidonic acid.
1-naphthol color development reagent. When platelets were challenged with 100 ,ug/mL arachi-
Platelet cytosolic calcium. Platelets were loaded with donic acid, they did not aggregate; however, treatment
indo-1, and agonist-induced changes in cytosolic free with greater than 100 ,ug/mL elicited a measurable
calcium were analyzed by ratio cytofluorimetry by the aggregation response. Based on these experiments 100
method of Jennings et al.28 PRP was loaded with indo-1 ,ug/mL was selected as our subthreshold concentration.
for 30 minutes at 37°C at the concentration that resulted in In the process of setting up our assay system we found
maximal response to agonist for that donor (15 or 20 that, in contrast to all our other aggregation assays in
,mol/L). An aliquot was diluted in a buffer consisting of 2 which carrier controls caused no effect on aggregation,
mmol/L CaCl2, 150 mmol/L NaCl, and 10 mmol/L Tris, the addition of the carrier (0.2%) to platelets caused a
pH 7.4. Cocaine or carrier was added at a final concen- marked aggregation response (82% to 89%, n=3) to
tration of 0.2%. Samples were then stimulated with either subthreshold amounts of arachidonic acid. However,
100 ,umol/L ADP or 0.2 U/mL thrombin. when the platelets were treated with 0.02% carrier
Results diluted in saline, the platelets had an aggregation
response of 5% or less when challenged with 100 jug/mL
To evaluate the effect of cocaine on platelet aggrega- arachidonic acid. As expected, platelets preincubated
tion, platelets from four medication-free donors were with 10-fold less cocaine (0.02%) diluted in saline
preincubated with increasing concentrations of cocaine rather than carrier did not aggregate in response to
or carrier and challenged with the agonists ADP (10 subthreshold amounts of arachidonic acid; however,
,umol/L) or collagen (2 ,ug/mL) in the aggregometer platelets incubated with 0.02% cocaine diluted in car-
(Fig 1). We found that the aggregation response to both rier rather than saline showed a 75% to 93% aggrega-
agonists was inhibited by cocaine in a concentration- tion response to 100 ,ug/mL of arachidonic acid. As the
dependent manner. The addition of carrier had no concentration of cocaine in the carrier was increased
effect on the aggregation response to either ADP or
collagen (data not shown). When ADP was the agonist, (greater than or equal to 0.04%), the aggregatory effect
the platelet aggregation responses were inhibited com- of the carrier was abolished. Thus, cocaine does inhibit
pletely by 0.2% cocaine (Fig 1A). With respect to arachidonic acid-mediated platelet aggregation, and it
collagen, the aggregation responses were completely appears that carrier and not cocaine is responsible for
inhibited by 0.1% cocaine (Fig iB). This inhibition of an enhanced response of human platelets to subthresh-
aggregation appeared to be specifically related to the old levels of this agonist. Based on our studies it is
ability of platelets to bind fibrinogen because shape difficult to support the hypothesis that cocaine causes an
change in response to ADP addition was observed (data enhanced aggregatory response of human platelets to
agonists in vivo. We would suggest that cocaine may
inhibit platelet function particularly when high local
TABLE 3. Effect of Cocaine on Agonist-Induced Increases in concentrations are achieved.
Platelet Cytosolic Calcium To determine whether cocaine had the capacity to
Maximum Mean disaggregate platelets and possibly affect preformed
Treatment Fluorescent Channel aggregates in the circulation in vivo, cocaine was added
Baseline 19.6±0.2 at increasing concentrations 3 minutes after the initia-
Carrier+ADP 29.9±2.2 tion of aggregation induced by ADP or collagen (Fig 2A
Cocaine+ADP 29.2±1.4 and 2B). We found that disaggregation due to the
presence of cocaine was dependent on the concentra-
Carrier+thrombin 38.8±3.4 tion of drug added. Maximal disaggregation was seen at
Cocaine+thrombin 40.2±2.7 approximately 0.4% cocaine with both agonists. Con-
Values are mean of three separate experiments±+SD. ADP centrations of 0.05% or lower did not appear to affect
concentration, 100 ,umol/L; thrombin concentration, 0.2 U/mL. platelet aggregate stability.
Downloaded from http://stroke.ahajournals.org/ by guest on October 21, 20151356 Stroke Vol 24, No 9 September 1993
ABR-,
Talin-
Myosin-'-- "- _ _
- a ~-
a-Actinirr"
Actin-r - S -
-
S
dom - ~
o"
rim.Li~~~~~~~~~~~~^
1 2 3 4 5 6 7 8 9 10 11 12 13
FIG 3. Photograph shows effect of cocaine on the cytoskeletal organization of the human platelet. Platelets were treated with either
0. 2% carrier or cocaine for aggregation inhibition studies or 0. 4% carrier or cocaine for disaggregation studies as described for lanes
1-12 and then solubilized in Triton X-100 extraction buffer. Cytoskeletal cores were isolated by centrifugation, washed twice, and
elect rophoresed through 5%-20% acrylamide gradien t gels. Separated proteins were then stained with Coomassie blue. A representative
photograph (n=3) of the isolated cytoskeletal cores is shown. Control treatments: lane 1, ADP (10 gmol/L) only; lane 2, collagen (2
gg/mL) only; lane 3, carrier only; and lane 4 cocaine only. For aggregation inhibition studies: lane 5, cocaine plusADP; lane 6, cocaine
plus collagen; lane 7, carrier plusADP; and lane 8, carrier plus collagen. For disaggregation studies: lane 9, ADP plus cocaine; la ne 10,
collagen plus cocaine; lane 11, ADP plus carrier; lane 12, collagen plus carrier; and lane 13, solubilized whole platelets. ABP indicates
actin~binding protein.
Once platelets are activated, it is the binding of the drug vehicle. We found that the addition of cocaine
fibrinogen that supports platelet aggregation. When did not inhibit increases in cytosolic free calcium levels
platelets were preincubated with 0.2% cocaine, fibrino- normally observed when platelets are challenged with
gen binding to ADP-activated platelets dropped to ADP or thrombin. Thus, it appears that cocaine at the
approximately 5% of the total platelets analyzed com- concentrations used in this study does not have a
pared with that of 84% when platelets were preincu- general effect on platelet reactivity by altering all phys-
bated with saline or with carrier (Table 2, n=5). When iological responses dependent on the availability of
platelets were activated with ADP and then 0.4% calcium ions but does specifically alter fibrinogen bind-
cocaine added, the percentage of platelets having bound ing, thereby preventing platelet aggregation.
fibrinogen dropped from 81% (in the presence of saline Following platelet activation, alterations in the plate-
or carrier) to 2% when cocaine was added. Thus, the let cytoskeleton include a net polymerization of mono-
inhibition of aggregation or the disaggregation of plate- meric actin into filaments and an increased association
lets mediated by cocaine is due at least in part to the of these filaments with other cytoskeletal proteins.24,29,30
inhibition of fibrinogen binding or to the dissociation of Platelet secretion, shape change, aggregation, and clot
fibrinogen bound to agonist-activated platelets. Similar retraction require the forces generated by the interac-
results were found when platelets were treated with 1 tion of these contractile proteins. Further examination
mmol/L dubucaine.15 of the mechanism by which cocaine affects platelet
The results from the above experiments suggest that reactivity was carried out by examining the organization
cocaine dramatically affects fibrinogen binding to acti- of the cytoskeletal core of platelets pretreated with
vated platelets. The following experiments were de- cocaine or the carrier control. In addition, the compo-
signed to determine whether cocaine specifically altered sition of the cytoskeleton when cocaine-treated platelets
ligand binding to GPIIb-IIIa, the human platelet fibrin- were challenged with the agonists ADP or collagen was
ogen receptor, or if cocaine at the same concentration analyzed. Cytoskeletal cores of platelets in plasma were
inhibited other calcium-dependent pathways that would isolated by solubilization with Triton X-100 extraction
normally lead to platelet activation, fibrinogen binding, buffer and centrifugation at 10 000g. The cores were
and platelet aggregation. It has been suggested that analyzed for protein content by SDS-polyacrylamide
cocaine, along with other local anesthetics such as gel electrophoresis. Fig 3 is a representative photograph
dibucaine, tetracaine, and lidodaine, may perturb cell of electrophoresed samples after staining with Coo-
membrane phospholipid organization by displacing cal- massie brilliant blue. Three separate cytoskeletal isola-
cium from the membrane and modifying cellular cal- tions were conducted, and identical results were ob-
cium homeostasis.9 tained. Lanes 1 and 2 show that platelets stimulated
Since platelet cytoplasmic free calcium plays a central with collagen or ADP had an increased incorporation of
role in platelet activation, we examined whether cocaine actin-binding protein (ABP), talin, c-actinin, myosin,
altered agonist-induced increases in platelet cytosolic and actin into the Triton-insoluble residue when com-
calcium. Indo-1-loaded platelets were preincubated pared with platelets treated with cocaine or carrier
with 0.2% cocaine and then challenged with either ADP alone (lanes 3 and 4). When platelets were pretreated
or thrombin (Table 3). Response was compared with with cocaine and challenged with agonist, the incorpo-
that obtained when platelets were preincubated with ration of these cytoskeletal elements did not increase
Downloaded from http://stroke.ahajournals.org/ by guest on October 21, 2015Jennings et al Cocaine-Induced Platelet Defects 1357
treated with 2 mmol/L tetracaine had major reductions
kDa of bands 1 and 2 (presumably ABP and talin) and
200 increases in the staining intensity of some bands in the
lower-molecular-weight regions. Our studies show that
the decreased ABP incorporation in the residues from
*
cocaine-treated platelets was not due to its proteolysis
by platelet calcium-dependent proteases in the intact
platelet because (1) cocaine does not stimulate calcium
- 97.4 influx and (2) ABP and talin in intact platelets were not
cc degraded by these proteases during a 5-minute incuba-
tion period with cocaine (Fig 4). It is possible that
increased susceptibility of ABP and talin to proteolysis
occurred once the platelets were lysed by Triton X-100.
Interestingly, as reported earlier with lidocaine-
- 46 treated platelets,13 we did find subtle differences in the
protein content of solubilized platelets depending on
whether they were exposed to cocaine or carrier. These
c30 alterations may in part explain the inhibitory effect of
cocaine on platelet function. Fig 4 shows the Coomassie
blue-stained gels of SDS-solubilized whole platelet
-21 proteins that were electrophoresed through a gradient
polyacrylamide gel. At concentrations of 0.02% (lanes 1
and 2), the protein staining pattern of either the co-
caine- or carrier-treated platelets appeared identical
1 2 3 4 5 and was indistinguishable from that of PBS-treated
FIG 4. Photograph shows alteration of the electrophoretic platelets (data not shown). However, when 0.2% co-
mobility of platelet proteins by incubation of intact platelets caine was incubated with the platelets (lane 3), the
with cocaine. Washed platelets were treated with two concen- staining pattern of three regions was affected in the
trations of cocaine and isolated by centrifugation (n=3). apparent molecular weight ranges of 170 to 180 kD, 75
Platelet proteins were solubilized with reducing sample buffer, kD, and 27 kD. The identification of these altered
analyzed by sodium dodecyl sulfate-gel electrophoresis, and
proteins has not been carried out, but it is possible that
their alteration may in part contribute to the profound
stained by Coomassie blue. Lane 1, 0.02% carrier; lane 2, effects that cocaine and possibly other anesthetics have
0.02% cocaine; lane 3, 0.2% carrier; lane 4, 0.2% cocaine; on platelet function. The appearance of the new 27-kD
and lane 5, molecular weight standards. Three regions on the protein in the cocaine-treated samples may represent a
gel (1358 Stroke Vol 24, No 9 September 1993
GPIlbaz. ,
GPIlia -
GPIlIb-i
1 2 3 4 5 6 7 8 9 10 11 12 13
FIG 5. Photograph shows effects of cocaine on incorporation of glycoprotein (GP) IIb-IIIa in the cytoskeletons of platelets
isolated from ADP- (10 gmolIL) and collagen- (2 gg/mL) aggregated platelets. Experiments were carried out as outlined in Fig
3. The platelet cytoskeletal proteins were isolated, washed, and electrophoresed through 5%-20% acrylamide gradient gels. The
proteins were then electrotransferred to nitrocellulose. After appropriate blocking procedures (see 'Methods"), the blots were
incubated with polyclonal anti~GPIIb-IIIa antiserum for at least 16 hours. After extensive washing, the blots were reacted with goat
anti-mouse immunoglobulin G horseradish peroxidase-conjugated antibodies for 2 hours. The blots were washed an additional
four times and developed by 4-chloro-1-naphthol color development reagent. Control treatments: lane 1, ADP (10 gmol/L) only;
lane 2, collagen (2 gg/mL) only; lane 3, carrier only; and lane 4, cocaine only. For aggregation inhibition studies: lane 5, cocaine
plus ADP; lane 6, cocaine plus collagen; lane 7, carrier plus ADP; and lane 8, carrier plus collagen. For disaggregation studies:
lane 9, ADP plus cocaine; lane 10, collagen plus cocaine; lane 11, ADP plus carrier; lane 12, collagen plus carrier; and lane 13,
solubilized whole platelets.
Discussion L), the concentrations of cocaine that caused platelet
The deleterious effects of cocaine on the cerebrovas- inhibition in our in vitro experiments are within an
cular system have been well documented. Much less is order of magnitude of the predicted systemic concen-
understood about the direct effect of cocaine on human tration of lethal doses of cocaine. An earlier attempt
platelet function and the possible contribution of this was made to examine an ex vivo effect of cocaine on
effect to cocaine-mediated stroke. To elucidate further platelet function by studying platelets from patients
the potential effects of cocaine on hemostasis, we with documented cocaine abuse and cerebral hemor-
examined the effect of cocaine on human platelet rhage. While depressed platelet function was observed,
aggregation and platelet activation responses such as interpretation of the direct effects of cocaine abuse on
platelet cytosolic free calcium levels, the organization of platelet function was not possible since each individual
the platelet cytoskeleton, and fibrinogen binding. Our had also ingested other drugs (eg, aspirin, alcohol)
data support the following conclusions: (1) cocaine known to depress platelet function. Whether depressed
prevents the binding of fibrinogen to agonist-stimulated platelet function due to cocaine abuse or due to a
platelets and can bring about the dissociation of fibrin- combination of drugs is clinically relevant to the inci-
dence of cerebral hemorrhage or to the management of
ogen bound to platelets incorporated into a platelet these patients remains to be determined.
aggregate, thus causing platelet disaggregation; (2) co-
caine does not inhibit agonist-mediated increases in Acknowledgments
platelet cytosolic calcium, and therefore a central mech- This study was supported in part by Baptist Memorial
anism that governs the reactivity of the human platelet Hospital Research Foundation and grant HL-38171 from the
is still intact; and (3) cocaine affects the organization of National Institutes of Health, Bethesda, Md. L.K.J. is an
the platelet cytoskeleton and inhibits the increased Established Investigator of the American Heart Association.
association of ABP, myosin, a-actinin, and actin as well
as GPILIb-llla in the Triton X-100-insoluble residue of References
ADP- or collagen-aggregated platelets. Our in vitro 1. Klonoff DC, Andrews BT, Obana WG. Stroke associated with
cocaine use. Arch NeuroL 1989;46:989-993.
data suggest that cocaine has a dramatic effect on 2. Mangiardi JR, Daras M, Geller ME, Weitzner I, Tuchman AM.
mechanisms responsible for mediating both adhesive Cocaine related intracranial hemorrhage: report of nine cases and
and cohesive properties of platelets. review. Acta Neurol Scand. 1988;77:177-180.
It has been difficult to determine the systemic levels 3. Green RM, Kelly KM, Gabrielsen T, Levine SR, Vanderzant C.
Multiple intracerebral hemorrhage after smoking 'crack' cocaine.
of cocaine in patients admitted to the hospital for Stroke. 1990;21:957-962.
cocaine abuse due to its short half-life (0.8 to 1.25 4. Levine SR, Brust JCM, Futrell N, Ho KL, Blake D, Millikan CH,
hours).31 The fatal dose of cocaine has been approxi- Brass LM, Fayad P, Schultz LR, Selwa JF, Welch KMA. Cere-
mated to be 1.2 g, although severe toxic effects have brovascular complications of the use of the 'crack' form of alka-
loidal cocaine. N Engl J Med. 1990;323:699 -704.
been reported from doses as low as 20 ng.32,33 Thus, 5. Ramadan NM, Levine SR, Welch KMA. Pontine hemorrhage
based on blood volumes of an average-sized person (5 following 'crack' cocaine use. Neurology. 1991;41:946-947.
Downloaded from http://stroke.ahajournals.org/ by guest on October 21, 2015Jennings et al Cocaine-Induced Platelet Defects 1359
6. O'Brien RA, Boullin DJ. Accumulation, storage and release of tacyclin in response to in vitro cocaine treatment. Haemostasis.
adrenergic neuron blocking agents and related drugs by human 1985;15:100-107.
platelets. Biochem Pharmacol. 1972;21:1817-1827. 21. White MM, Foust JT, Mauer AM, Robertson JT, Jennings LK.
7. Behnke 0. Effects of some chemicals on blood platelet micro- Assessment of lumiaggregometry for research and clinical labora-
tubules, platelet shape and some platelet functions in vivo. Scand tories. Thromb Haemost. 1992;67:572-577.
J Haemat. 1970;7:123-140. 22. Jackson CW, Jennings LK. Heterogeneity of fibrinogen receptor
8. Grant RJ, Ramanathan S, Patel N, Turndorf H. The effects of local expression on platelets activated in normal plasma with ADP:
anesthetics on maternal and neonatal platelet function. Acta An- analysis by flow cytometry. Br J Haematol. 1989;72:407- 414.
aesthesiol Scand. 1989;33:409-412. 23. Jennings LK, Ashmun RA, Wang WC, Dockter ME. Analysis of
9. Feinstein MB, Fiekers J, Fraser C. An analysis of the mechanism human platelet glycoproteins lIb and Illa and Glanzmann's throm-
of local anesthetic inhibition of platelet aggregation and secretion. basthenia in whole blood by flow cytometry. Blood. 1986;68:
J Pharmacol Exp Ther. 1976;197:215-228. 173-179.
10. Vanderhoek JY, Feinstein MB. Local anesthetics, chlorpromazine 24. Jennings LK, Fox JEB, Edwards HH, Phillips DR. Changes in the
and propranolol inhibit stimulus-activation of phospholipase A2 in cytoskeletal structure of human platelets following thrombin acti-
human platelets. Mol Pharmacol. 1979;16:171-180. vation. J Biol Chem. 1981;256:6927- 6932.
11. Nachmias V, Sullender J, Asch A. Shape and cytoplasmic filaments 25. Kouns WC, Fox CF, Lamoreaux WJ, Coons LB, Jennings LK.
in control and lidocaine-treated human platelets. Blood. 1977;50: The effect of glycoprotein lIb-llla receptor occupancy on the
39-53. cytoskeleton of resting and activated platelets. J Biol Chem.
12. Davies GE, Palek J. The state of actin polymerization in 1991;21:13891-13900.
tetracaine-treated platelets. Thromb Haemost. 1982;48:153-155. 26. Laemmli UK. Cleavage of structural proteins during assembly of
13. Nachmias VT, Sullender JS, Fallon JR. Effects of local anesthetics the head of the bacteriophage T4. Nature. 1970;227:680-685.
on human platelets: filopodial suppression and endogenous prote- 27. Burnette WN. 'Western blotting': electrophoretic transfer of
olysis. Blood. 1979;53:63-72. proteins from sodium dodecyl sulfate-polyacrylamide gels to
14. Prowse C, Pepper D, Dawes J. Prevention of the platelet alpha- unmodified nitrocellulose and radiographic detection with
granule release reaction by membrane-active drugs. Thromb Res. antibody and radiolabeled protein A. Anal Biochem. 1981;112:
1982;25:219-227. 195-203.
15. Peerschke EIB. Platelet membrane alterations induced by the local 28. Jennings LK, Dockter ME, Wall CD, Fox CF. Calcium mobili-
anesthetic dibucaine. Blood. 1986;2:463-471. zation in human platelets using indo-1 and flow cytometry. Blood.
16. Solum NO, Olsen TM. Effects of diamide and dibucaine on 1989;74:2674-2680.
platelet glycoprotein Ib, actin-binding protein and cytoskeleton. 29. Carroll RC, Butler RG, Morris PA, Gerrard JM. Separable
Biochim Biophys Acta. 1985;817:249-260. assembly of platelet pseudopodial and contractile cytoskeleton.
17. Odoom JA, Sturk A, Dokter PWC, Bovill JG, Cate JWT, Oosting Cell. 1982;30:385-393.
J. The effects of bupivacaine and pipecoloxylidide on platelet 30. Phillips DR, Jennings LK, Edwards HH. Identification of
function in vitro. Acta Anaesthesiol Scand. 1989;33:385-388. membrane proteins mediating the interaction of human platelets.
18. Gilani AH, Saeed SA. A comparative study of the effects of local J Cell Biol. 1980;86:77-86.
anaesthetics on platelet aggregation. Biochem Soc Trans. 1990;18: 31. Nolte KB, Gelman BB. Intracerebral hemorrhage associated with
288-289. cocaine abuse. Arch Pathol Lab Med. 1989;113:812-813.
19. Anderson ER, Foulks JG, Godin DV. The effect of local anaes- 32. Roxane Laboratories, Inc. Cocaine hydrochloride topical solution.
thetics and antiarrhythmic agents on the responses of rabbit Package insert 4046600, May 1988.
platelets to ADP and thrombin. Thromb Haemost. 1981;45:18-23. 33. Frishman WH, Karpenos A, Molloy TJ. Cocaine-induced coronary
20. Togna G, Tempesta E, Togna AR, Dolci N, Cebo B, Caprino L. artery disease: recognition and treatment. Med Clin North Am.
Platelet responsiveness and biosynthesis of thromboxane and pros- 1989;73:475-486.
Downloaded from http://stroke.ahajournals.org/ by guest on October 21, 2015Cocaine-induced platelet defects.
L K Jennings, M M White, C M Sauer, A M Mauer and J T Robertson
Stroke. 1993;24:1352-1359
doi: 10.1161/01.STR.24.9.1352
Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1993 American Heart Association, Inc. All rights reserved.
Print ISSN: 0039-2499. Online ISSN: 1524-4628
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://stroke.ahajournals.org/content/24/9/1352
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Stroke can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.
Once the online version of the published article for which permission is being requested is located, click
Request Permissions in the middle column of the Web page under Services. Further information about this
process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Stroke is online at:
http://stroke.ahajournals.org//subscriptions/
Downloaded from http://stroke.ahajournals.org/ by guest on October 21, 2015You can also read
