From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
Heterogeneity in Filamentous Actin Content Among Individual Human Blood
Platelets
By Atsushi Oda, John F. Daley, Claudia Cabral, Joonghee Kang, Marianne Smith, and Edwin W. Salzman
The content of filamentous actin in individual platelets was
measured by flow cytometry, using a fluorescent probe
specific for filamentous actin (F-actin), 7-nitrobenz-2-oxa-l.3phallacidin (NBD-phallacidin). NBD-phallacidin binding t o
fixed platelets was specific in that either pretreatment of
platelets with unlabeled phallacidin or absorption of NBDphallacidin by rabbit skeletal F-actin, but not globular actin
(G-actin), resulted in a significant loss in the bound fluorescent probe. Mean NBD-phallacidin binding t o fixed platelets
varied with the agonist and paralleled the changes in F-actin
reported with the DNAse I inhibition assay. (1) NBDphallacidin binding increased with stimulation by ADP,
U46619 (a prostaglandin H, analogue), or collagen and paralleled shape change. (2) Epinephrine did not increase NBDphallacidin binding. (3) Platelets treated at 4°C contained
more F-actin than did platelets kept at 37°C. (4) Cytochalasin
D ( I O pmol/L) inhibited the increase of phallacidin binding t o
individual platelets stimulated by either ADP or U46619. In
measurementsof cytosolic free calcium concentration ([Ca"],)
by flow cytometry in Indo-1-loaded platelets, ADP's doseresponse for actin polymerization was similar t o that for
calcium mobilization. As shown by flow cytometry, a tail
population that had a minimal increase in F-actin upon
stimulation with ADP or U46619 also contained the platelets
with the least forward and right angle light scattering, which
are functions of platelet size and shape. When platelets
treated with NBD-phallacidin were incubated with S12murine monoclonal antibody (a marker of agranule secretion
detected by phycoerythrin-conjugated antimouse IgG second antibody), phallacidin fluorescence paralleled S I 2 binding. Thus, human blood platelets are heterogeneous in
regard t o actin polymerization at rest and in association with
platelet activation; different degrees of phallacidin binding
may identify functionally different platelet populations.
o 1992by The American Society of Hematology.
A
fibrinogen binding,14 fibrinogen receptor expressi~n,'~
and
granule ~ecretion.'~.'~
Recently, we reported that low concentrations of thrombin produce a heterogeneous distribution
of elevation of [Ca'+], in platelets and that this is reflected in
parallel inhomogeneity in secretion." This finding suggests
that heterogeneity in different aspects of platelet activation
may be related. To address the question of the possible
heterogeneity of F-actin content among individual platelets, we used 7-nitrobenz-2-oxa-l,3-phallacidin
(NBDphallacidin) for flow cytometric studies. This probe is a
fluorescent derivative of mushroom (Amanita phalloides)
toxins and has been successfully used for flow cytometric
studies of fixed neutrophiles and other cell^.'^-'^ With
measurement of F-actin content in individual platelets by
NBD-phallacidin, we confirmed the increase in F-actin
content upon platelet stimulation and, further, found that
there was heterogeneity in F-actin content among individual platelets. This method, which requires as little as 10 pL
of platelet-rich plasma (PRP), may be useful to answer
other questions of basic or clinical interests.
CTIN COMPRISES about 20% of the total platelet
protein mass and its polymerization in filamentous
form (F-actin) is an early event in platelet activation,
apparently involved in shape change, granule centralization, and clot retraction.'"
Recently, it was suggested that actin polymerization
might play important roles in collagen-induced platelet
activation' and translocation of glycoprotein (GP) Ib from
the platelet s ~ r f a c eFurthermore,
.~
F-actin content in platelets was reported to be increased in diabetic patientssz6and
during platelet tora age.^ Because of the high content of the
protein and relatively easy availability of the cell, the
platelet has served as a model system for studies of actin
and its polymerization.'.' Polymerized actin can be measured by high speed centrifugation.'.' Alternatively, globular actin (G-actin) can be measured by DNAse I inhibition
assay and then F-actin content can be calculated by
subtracting the amount of G-actin from the total actin
However, these biochemical methods are blind to
the possible heterogeneity of platelets in F-actin content
and are often not suitable for a small sample.
Platelets are heterogeneous in many respects, including
size,' density," free calcium concentration ([Ca2+]J,1'-'3
From the Departments of SurgeT and Medicine, Beth Israel
Hospital; and Division of Tumor Immunology, Dana Farber Cancer
Institute, Harvard Medical School, Boston, MA.
Submitted June 13, 1991; accepted October 1,1991.
Supported by National Heart, Lung, and Blood Institute Grants No.
HL-38820 and HL-344014.
Address reprint requests to Edwin W. Salzman, MD, Department of
Surgev, Beth Israel Hospital, 330 Brookline Ave, Boston, MA 02215.
The publication costs of this article were defiayed in part by page
charge payment. This article must therefore be hereby marked
"advertisement" in accordance with 18 U.S.C. section 1734 solely to
indicate this fact.
0 1992 by The American Society of Hematology.
0006-49711921 7904-0016$3.00/0
920
MATERIALS AND METHODS
Materials were obtained as follows. NBD-phallacidin and Indo-1
acetoxymethylester (AM) were from Molecular Probes (Eugene,
OR); ADP, U46619 (9, ll-dideoxy-9a, lla-methoepoxy prostaglandin F2a), epinephrine, cytochalasin D, phallacidin, aspirin,
phycoelythrin (PE)-conjugated antimouse goat antibody, rabbit
skeletal actin, and mouse IgG2b (MOPC 141) were from Sigma (St
Louis, MO); horse type I collagen was from Hormon-Chemie
(Munchen, Germany); glutaraldehyde was from Fisher Scientific
(Fairlawn, NJ); green fluorescent standard beads for flow cytometry were from Flow Cytometry Standards (Research Triangle Park,
NC); S12 antibody? which reacts with GMP-140 molecule,2' was
the kind donation of Dr Rodger P. McEver (University of Oklahoma).
Measurement of NBD-phallacidin binding to individual platelets.
Blood was obtained from healthy donors by venipuncture and was
anticoagulated with 0.38% trisodium citrate. PRP was obtained by
centrifugation (200g for 20 minutes). Only the upper half of PRP
Blood, Vol79, No 4 (February 15). 1992: pp 920-927
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
FILAMENTOUS ACTIN IN PLATELETS
was collected to avoid contaminating red cells. In such preparations, fewer than 2% of cells were contaminating red cells, as
determined by phase microscopy. PRP was warmed at 37°C for 30
minutes before experiments. Except in experimentswith collagen,
aspirin (1 mmol/L) was added 30 minutes before platelet agonists.
Fifteen seconds before each experiment, 5 mmol/L EDTA was
routinely added to prevent aggregation of platelets. An aliquot (10
to 50 pL) of PRP was fixed with freshly prepared 2.5% glutaraldehyde (1.5 mL) in phosphate-bufferedsaline (PBS; pH 7.4) at 37°C.
After fixation (at least 2 hours), platelets were centrifuged (8OOg
for 5 minutes) and washed twice in PBS. Fixed platelets (106 cells)
were incubated with NBD-phallacidin and resuspended in PBS at a
final concentration of 3.3 pmol/L. In preliminary experiments,
NBD-phallacidinbinding reached a plateau in less than 30 minutes
with occasional gentle shaking. Fixation for up to 24 hours did not
change the extent and distribution of phallacidin fluorescence.
Permeabilizationwas not necessary, as was also shown to be true of
much larger cells, neutrophils, probably because of the small
molecular size of NBD-phallacidin (around 800 d).17.” In some
experiments, NBD-phallacidinwas preincubated with rabbit skelebefore
tal G-actin or F-actin (2 mg/mL), prepared as de~cribed,’~
the addition to platelet suspensions in G-buffer (5 mmol/L Tris,
pH 7.4, 1 mmol/L GTP) or in F-buffer (100 mmol/L KCl, 1
mmol/L MgCI,), respectively,at a final NBD-phallacidinconcentration of 3.3 pmol/L.
After incubation, platelets were washed twice and examined
with the FacStar Plus flow cytometer (Becton Dickinson, Braintree, MA). Both NBD and PE were excited with a 5 W argon laser
at 200 mW power at a wave length of 488 nm. NBD fluorescence
was detected using a 530 f 11 nm band pass filter, and PE
fluorescence was detected with a 575 f 11nm band pass filter. The
platelet population was defined by light scattering of 488 nm beam
(forward and right angle) to gate out contaminatingred cells or cell
debris from contamination as described.” Five thousand platelets
were examined in each sample at a rate of less than 1,OOO cells per
second. Log-NBD-fluorescence was converted to a linear scale and
calibrated by green standard beads as de~cribed.’~
Aliquots of the
platelet samples were pretreated with unlabeled phallacidin (33
pmol/L, 10 times excess of the final concentration of NBDphallacidin) for 1 hour before the addition of NBD-phallacidin.
We assumed the mean fluorescence of these pretreated platelets to
be a measure of mean nonspecificbinding of NBD-phallacidin plus
mean autofluorescence. We subtracted this mean fluorescence
from the mean fluorescence of counterparts without pretreatment
with nonfluorescent phallacidin to estimate the specific mean
fluorescence of NBD-phallacidin bound to platelets.
Measurement of platelet [Caz+],by flow cytometry. Loading of
Indo-1 and measurement of Indo-1 fluorescence of individual
platelets (in diluted PRP, at a final platelet concentration of
2 x lo6cells/mL) was performed with an EPICS V model 753 flow
cytometer (Coulter Electronics, Hialeah, FL) as described in
detail.” Indo-1 violet/blue fluorescence ratio (0 to 1.0) was used as
a measure of calcium change. The difficulty in calibration of [Ca*’],
was described in detail by Jennings et a1.’*
Detection ofplatekt s h p e change. “Shape change” of platelets
was investigated by a lumiaggregometer (Chronolog Corp, Havertown, PA) as described using initial rate of decrease of light
transmission as a measure of “shape change.”22Platelet concentrations were adjusted to 3 X 10scells/mL by dilution with autologous
platelet-poor plasma (PPP). Aggregation was inhibited by the
addition of 5 mmol/L EDTA. Although it is not entirely clear what
aspect of change of platelet shape is reflected by this measurement
or by “Schlierren,” there is widespread acceptance of this technique as a measure of platelet activation.2)The use of this method
921
facilitated the comparison of our experimentswith those previously
reported.
RESULTS
Because NBD-phallacidin has not previously been used
for platelet studies, particular attention was paid to determination of the specificity of NBD-phallacidin to fixed platelets. Platelets in PRP instead of gel-filtered platelets were
used, because Fox et a1 reported that gel-filtered platelets
with a minimal amount of F-actin can be obtained only in
the presence of prostacyclin, a potent platelet inhibitor,”.=
at 37”C’32;
such platelets are not suitable for subsequent
stimulation. Spangenberg and Till also described difficulty
in the preparation of gel-filtered or washed platelets with a
minimal amount of F-actin.6Fixed resting platelets exposed
to NBD-phallacidin (3.3 Fmol/L) increased NBD-fluorescence (excitation 488 nm, emission 530 nm) (Fig lA),
although there was significant overlap between the autofluorescence level and the fluorescence level of NBDphallacidin-treated cells. The increase was not affected by
pretreatment of NBD-phallacidin with rabbit G-actin (Fig
lB), but was markedly inhibited by pretreatment of the
fluorescent probe with rabbit F-actin (Fig 1C). Pretreatment of platelets by unconjugated phallacidin also largely
eliminated the binding of NBD-phallacidin(Fig 1D). Taken
A
Intact tixed plateleis
150
1
B
G-actin treated NBD-ph8llacidin
binding
.-.
FLUORESCENT UNIT
FLUORESCENT UNIT
Fig 1. NED-phallacidin binding to unstimulated cells: the effects of
G and F actins or pretreatmentof plateletswith unlabeled phallacidin.
Unstimulated fixed platelets were incubated with 3.3 pmol/L NEDphallacidin alone (A), or in the presence of rabbit skeletal G-actin (1
mg/mL) (E) or rabbit skeletal F-actin (1 mg/mL) (C). Alternatively,
fixed plateletswere pretreatedwith unlabeled phallacidin(33 pmol/L,
10 times excess of final concentration of NED-phallacidin) for 1 hour
before the addition of NED-phallacidin (D). The abscissa expresses
log-NBD fluorescenceintensity, convertedto linear channel number (0
to 255). The ordinate shows cell number. (*
.) The histogram of
autofluorescence of the platelet population. Data are from a study
typical of seven different experiments.
..
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
922
ODA ET AL
together, these data indicate the specificity of NBDphallacidin binding to platelets through F-actin.
It has previously been reported that cytochalasins prevented the polymerization of a~tin.~.’.~,’~
When platelets in
PRP (pretreated with dimethyl sulfoxide [DMSO], the
vehicle of cytochalasin D; final concentration, 0.1%) were
stimulated by ADP (10 pmol/L, Fig 2A) or U46619 (1
pmol/L; Fig 2C) for 15 seconds, they had increased
phallacidin binding. After either ADP- or U46119-induced
stimulation, there were platelets that apparently failed to
increase their F-actin content. These platelets constituted a
shoulder on the left side of the curve, while the main
population, which showed a log normal distribution, shifted
to the right. There was significant overlap of the main
population and the cells refractory to stimuli, especially in
the resting state. When anti-GPIb monoclonal antibody
(MoAb; AF’1, the kind gift of Dr Thomas J. Kunicki, South
East Wisconsin Blood Center, Milwaukee, WI) and PEconjugated antimouse antibody were used to identify the
NBD-phallacidin-pretreated platelet population, their patterns of distribution of NBD-fluorescence were unchanged,
indicating that the “shoulder population” was indeed
composed of platelets. When platelets were pretreated with
cytochalasin D (10 pmol/L) for 1 minute, they did not
increase their phallacidin binding with either ADP or
U46619 (Fig 2B and D).
We also examined light scattering by the platelet population, which, when aggregation is inhibited, is considered to
be a function of platelet size and ~ h a p e . ’ ~ ,In~ the
~ * ~resting
**~
state, glutaraldehyde-fixed platelets had a triangular light
scattering distribution in the contour gTaph of their log
forward and right angle scattering of a 488 nm beam (Fig
2E, left panel), which is consistent with the report of Holme
et
Within 15 seconds after stimulation by ADP (10
pmol/L) or U46619 (1 pmol/L), the platelet population
increased their forward scattering, and at the same time
forward scattering and right angle scattering became more
linearly correlated than in the resting state (Fig 2E, middle
and right panels). When NBD-phallacidin fluorescence was
compared with forward scattering (Fig 2F) and right angle
scattering (Fig 2G), it became clear that cells with the least
phallacidin binding were the cells with the least forward
and right angle light scattering, especially after stimulation
with ADP or U46619.
The effects of various stimuli on actin polymerization
were examined. As stated above, U46619, ADP, and
collagen, which are known to induce shape change,=
increased phallacidin binding (Table 1). Epinephrine, which
does not induce shape change in aspirin-treated platelets
(ie, during the primary phase of epinephrine-induced
platelet activation),= did not increase phallacidin binding
A
B
control + ADP
150
0
100
Fig 2. NED-phallacidin binding t o DMSO-pretreated (A and C) or
cytochalasin D-pretreated (10 pmol/L) (B and D) platelets. (-)ADP
(10 pmol/L, 15 seconds)-stimulated cells (A and B) or U46619 (1
pmol/L, 15 seconds)-stimulated cells (C and D). The abscissa and
ordinate are the same as in Fig 1. (E) Contour graphs of light scattering
of the resting (left panel) and activated platelet populations (middle
and right panels), in which the abscissa is log forward scattering and
the ordinate is log right angle scattering, recorded at the same time as
the results of (A and C) (unstimulated), (A) (ADP-stimulated), (C)
(U46619-stimulated). The pixels with more cells are surrounded with
more lines. (F) and ( G )are contour graphs of forward scattering versus
NBD-fluorescence (F) and of right angle scattering versus NBDfluorescence. The platelet populations and lines were as described in
(E). Data are representative of seven different experiments.
1
0
200
c
control
D
+ U4661S
150
100
200
FLUORESCENT UNIT
FLUORESCENT UNIT
Cytochalasin treated
+
w6619
150
0
100
200
FLUORESCENT UNIT
Unstlmulated
0
100
200
FLUORESCENT UNIT
ADP 10 pM
U46619 1 pM
-5
W
-
W
E
.-UW
Forward Scattering
F
Unstimulated
ADP 10 pM
U46619 1 pM
m
C
U
!I
LL
-
.-E
>
treated
150
1
I
E
+Cyio;halh.l.aln
-m
Forward Scattering
c
m
Unstimulated
ADP 10 pM
m
C
U
I!
P
.-
-a
c
m
n
Right Angle Scattering
U46619 1 pM
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
FILAMENTOUS ACTIN IN PLATELETS
923
Table 1. Increase of F-Actin Content induced by Platelet Stimulation
ADP
U46619
Collagen*
Epinephrine
4°C. 10 min
(10 kmol/L, 15sec)
(1 pmol/L, 15 sec)
(10 FglmL)
(10 Frnol/L, 1 min)
1.68 f 0.20
1.89 f 0.30
1.52 ? 0.18
1.04 e 0.05
1.34 2 0.03
(N = 10, P < ,005)
(N = 10, P < ,005)
(N = 6,P < ,005)
(N = 7, P > .05)
(N = 3, P < ,005)
Times increased over resting 2 standard deviation. The significance
of the increase over resting levels was tested by paired t-test.
*Collagen-induced shape change was delayed in the presence of 5
mmol/L EDTA. Samples were taken approximately 1.5 minutes after
the addition of collagen, when the decrease of light transmittance
reached its maximal level, as detected by aggregometry.
(Table 1). However, cold treatment (4°C for 10 minutes),
which is a known stimulant of platelet shape change and
actin polymerization,1~2
induced a 1.3-fold increase in mean
phallacidin binding to platelets.
ADP-induced actin polymerization, as indicated by phallacidin binding, reached a peak within 15 seconds and then
decreased continuously from 1 minute after stimulation;
the effects of U46619 were more prolonged than those of
ADP (Fig 3). Figure 4A shows the dose dependency of
ADP-induced phallacidin binding from 0.1 to 10 Fmol/L,
the concentration range in which ADP induced dosedependent shape change (Fig 4B).22323,30
It was possible by flow cytometry to measure [Ca’+], in
diluted PRP, without gel-filtration or further centrifugation
that might cause refractoriness to ADP.31After the addition
of ADP (10 pmol/L) in the presence of extracellular
calcium (1 mmol/L), all the platelets elevated their [Ca’+],
(Fig 5A). The removal of extracellular calcium by 5 mmol/L
EDTA attenuated the elevation of [Ca”], induced by ADP
(Fig 5B). However, ADP concentrations of 0.5 Fmol/L or
more increased [Ca’+], even in the absence of extracellular
calcium (Fig 5B and C), albeit transiently. This finding is
consistent with our previous observation^,'^ but is in con-
trast to those of Jennings et a],” who found calcium
elevation induced in only a limited number of platelets
stimulated at room temperature, possibly because of the
differences in temperature of these experiment^.'^ Both the
dissociation constant of calcium indicators and the cell
calcium response may be altered in experiments performed
at room temperat~re.~’
ADP (0.1 Fmol/L) had no effect on
[Ca’+], (Fig 5D). Thus, ADP concentration required to
increase [Ca2+],was similar to that necessary for actin
polymerization.
U46619 but not ADP is known to induce granule secretion in the absence of extracellular calcium; ADP-induced
secretion requires aggregation and secondary thromboxane/
endoperoxide production?’ We compared the extent of
a-granule release (indicated by binding of S12 MoAb to
platelets) with actin polymerization. After stimulation with
an agonist, platelets were fixed and exposed to S12 MoAb
(20 pg/mL) and, after washing, to PE-conjugated antimouse antibody (goat, 30 pg/mL) and NBD-phallacidin
and were then analyzed by flow cytometry. Figure 6 shows a
two-color analysis of PE and NBD-fluorescence. Fifteen
seconds after the addition of U46619, an increase in
NBD-phallacidin fluorescence from the level at 0 seconds is
indicated as a rightward shift (compare Fig 6A with 6B) of
the population with the maximal phallacidin binding (the
population, shown in Figs 1 and 2C, that had a bell shaped
NBD-fluorescence distribution). There is also a small
increase in S12 binding, implying a-granule secretion,
which is progressively more prominent from 15 seconds
through 5 minutes after stimulation (Fig 6C and D). At 5
minutes after stimulation, the cells that bound the most S12
were the ones that bound the most phallacidin. The
proportion of S12-negative cells was higher in the population with the least phallacidin binding, especially in the
early stages of platelet stimulation.
DISCUSSION
2.0
U46619 1 p M
\
1.4
1.2
ADP 10 p M
1 .o
0
200
400
600
Time (sec)
Fig 3. Time course of mean NBD-fluorescence change with ADP
(10 pmol/L, ). or U46619 (1 pmol/L,
The mean ? standard
deviation (SD) of three independent experiments from different
donors. The ratio of the increase of NBD-fluorescence after stimulation to fluorescence in an unstimulated sample (time = 0) is on the
ordinate.
e).
Heterogeneity in the amount of F-actin in platelets has
not been described previously. Measurement of F-actin
content by phallacidin binding suggests the presence of
heterogeneities in the state of actin polymerization in both
resting and stimulated platelets (Figs 1and 2). Distribution
of autofluorescence level (Fig 1, shown by dotted lines) or
autofluorescence plus nonspecific binding (the fluorescence
of phallacidin-pretreated cells, Fig 1D) showed a log
normal pattern. However, the pattern of distribution of
NBD-phallacidin was clearly asymmetrical in both resting
and stimulated cells (Figs 1A and 2A and C), and had a
leftward shoulder with minimal NBD-phallacidin binding.
NBD-phallacidin binding was specific in that it was inhibited by exogenous F-actin but not G-actin (Fig 1B and C),
and the heterogeneity in NBD-phallacidin binding appeared to be due to differences in the degree of specific
NBD-phallacidin binding to individual platelets, which, in
turn, most likely reflected heterogeneous F-actin content
among individual platelets. Inadequate exposure to NBD
phallacidin, an alternative possibility as a cause of the
heterogeneity, is unlikely, because stimulated platelets
showed a clear increase of NBD-phallacidin binding in a
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
ODA ET AL
924
B
’7
T
“
-7.5
-7.0
-6.5
-6.0
-5.5
-5.0
-4.5
Log molar ADP concentration
I
-7.5
.
I
-7.0
.
I
-6.5
.
I
-6.0
.
I
-5.5
.
I
-5.0
.
I
-4.5
Log molar ADP concentration
Fig 4. (A) Dose dependency of phallacidin binding induced by ADP. ADP was added 15 seconds earlier. The abscissa is the log-molarconcentration of ADP. The ordinate is the same as in Fig 3. (B) Initial change of light transmittance, measured as described,= induced by ADP. Rate
was normalized to that induced by ADP (10 pmol/L). The abscissa is the same as in (A). The light transmission change was recorded when the
aliquot of the platelet sample for (A) was taken. Values are the mean f standard deviation (SD) of three independent experiments from different
donors.
majority of platelets. Such heterogeneity was also previously described for neutrophils’721s
and may be a general
feature of blood cells.
Most of the current knowledge of actin polymerization
has been obtained from changes in F- and G-actin content,
measured biochemically (by DNAse 1 inhibition or by high
speed centrifugation) after the lysis of platelets in EDTA or
EGTA containing buffers’3239
to minimize the concentration
of ionized calcium; without a calcium chelator, F-actin may
be rapidly depolymerized. Although the occurrence of
increased actin polymerization was also observed by electron microscopic studies,’,’ the circumstances have been
somewhat puzzling, because, as shown by Fig 5 and reported by
actin polymerization occurs at the same
time as calcium mobilization. The situation has been
termed the “Ca’+ paradox.”’ It was suggested that gelsolin
and calcium elevation, both of which stabilize profilactin
complexes in vitro and together sever actin filaments,%may
account for the apparent paradox.’ If so, depolymerization
instead of polymerization of actin could occur in platelets
with an elevation of [Ca’+Ii,at least theoretically, because
platelet cytoplasm also contains gelsolin and profilin.19
More recent studies suggest that there is no 1:1 stoichiometry between profilin and actin in platelet cytoplasm35and
that these provide no quantitative evidence that EGTA- or
EDTA-lysis buffers actually “freeze” the actin strand as it is
found within the cells. The results of our studies are
consistent with those previously obtained by biochemical
techniques.’,’ We found that U46619, ADP, and collagen
increased mean phallacidin binding 1.5 to 1.9 times over the
resting levels. Biochemical techniques indicate that F-actin
constitutes 30% to 50% of total actin in resting platelets,
and it increases to 60% to 80% of the total after stimulation’.’; these values are similar to those detected by our
measurement of NBD-phallacidin binding. The effects of
cold treatment and of cytochalasin D, examined by flow
cytometry, were also consistent with the results of studies
performed with biochemical technique^.^^^^^^^^ Thus, there
must be mechanisms in vivo that dissociate profilactin in
spite of the presence of elevation of ionized calcium
concentration and gelsolin, permitting freed G-actin to
polymerize.
Our data also confirm the previously described relation
of actin polymerization to shape change. Dose-dependent
ADP-induced “shape change,” verified by light transmission in aggregometry,= was observed (Fig 4B) at ADP
concentrations that agreed with previous r e p ~ r t s . ’ ’ ~ ~ ~ ’ ~
U46619, collagen, ADP, and cold treatment, all of which
induce “shape change,”= polymerized actin (Table 1).
Epinephrine, which does not cause “shape change,” did not
increase F-actin (Table 1). Feinberg et a1 reported a good
agreement between actin polymerization and “shape
change” in stored platelets.’ A causal relation between
shape change and actin polymerization is likely; the phosphorylation of myosin light chain may also play a role.’3
We found that ADP-induced actin polymerization was
largely reversible, while U46619 induced a more prolonged
actin polymerization (Fig 3). U46619 (1 kmol/L) is in
general a more potent stimulus for platelets than ADP, with
respect to functions such as granule secretion (Fig 6). In
neutrophils, too, it has been reported that stronger stimuli
induced more prolonged actin polymeri~ation.”~’~
Other examples of platelet heterogeneity have been
documented by flow cytometry,”-I6but the origin of the
heterogeneity has not been clarified. We found a correlation of granule secretion with the degree of specific phallacidin binding (Fig 6), consistent with our view that many
aspects of platelet heterogeneity may be interrelated.13
Further, we found that the population with the least
phallacidin binding was found among the cells with the least
forward and right angle light scattering (Fig 2F and G).
Jackson and Jennings reported that the platelets with the
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
FILAMENTOUS ACTIN IN PLATELETS
925
A
+ 1 mMCa2+
TIME
0
' A
300 sec
1 .o
300s
.....
.......
.
...:..
.....-.. ...........
. .....
......
ADP 10 uM
c + 5mMEDTA
D + 5mMEDTA
0
L
Q,
0
I
ADP 10 uM
c
d)
5mMEDTA
TIME
1 .o
.-0
+
0
85
TIME
TIME
=3
c
6
v
c 1.0
0
300 sec
1 .o
I
Fig 5. Two-dimensional histogram of
calcium distribution (frequency plot).
Platelets examined at a rate of 600 cells/s.
The darkest area contains the majority of
platelets. Platelets were stimulated by
ADP 10 pmol/L (A and B), 0.5 pmol/L (C),
0.1 pmol/L (D). EDTA 5 mmol/L was
added except in (A), in which 1 mmol/L
calcium was added instead. Arrows indicate the time of addition of ADP. The
ordinate is the violet/blue ratio of Indo-1
fluorescence, as a measure of [CaZ+],. The
abscissa is time. Data are representative
of five different experiments.
300 sec
0
O t
ADP 0.5 uM
O
f
ADP 0.1 uM
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
ODA ET AL
926
A
Resting platelets
B U46619 1 JIM, 15 sec
-
NBD-phallacidin binding
c U46619 lyM,1 min
D U46619 1 pM, 5 min
Fig 6. Two-color analysis of log NBD-fluorescence(phallacidin binding) and log PE fluorescence (S12MoAb binding) as a measure of a-granule
release. Each figure was divided into four quadrants (1 through 4, identified on the figure) according t o the intensity of fluorescence. The level of
PE fluorescence dividing the upper and lower quadrants was selected so that the binding of nonspecific murine lgG2b (recognized by
PE-conjugatedantigoat second antibody) would remain inthe lower quadrants. NBD-fluorescence was divided according t o the autofluorescence
level. An appropriate compensation for overlapping fluorescence between NBD and PE was performed by computer, so that a PE fluorescence
increase results in upward shift in the figure and an increase in NBD-fluorescence in the rightward shift. The percentage of platelets in each
quadrant was as follows (100% = the total, approximately 5,000 cells examined): A (14.42%; 2-3.95%; 3-11.20%; 444.43%). B (14.38%;
2 4 3 8 % ; 3-9.56%; 441.68%). C (14.05%; 247.78%; 3-10.13%; 449.04%). D (1467%; 247.26%; 34.83%; 4-22.74%). Each dot represents a
pixel that includes one t o nine platelets. Enclosed areas indicate that the pixels in them contained at least 10 platelets and, thus, indicate
maximum platelet density. About 70% of the total platelet population is included in the enclosed area. Data are representative of five different
experiments.
least fibrinogen binding were the cells with the least
forward ~cattering.’~
They interpreted forward scattering as
an indicator of platelet size and found a linear relation
between the amount of GPIIb-IIIa on the platelet surface
and forward ~ c a t t e r i n g . Holme
’ ~ ~ ~ et a1 observed a positive
correlation between platelet size estimated by Coulter
counter and forward and right angle light ~cattering.’~
Thus,
it is possible that the platelets with the least forward and
right angle scattering are indeed the smallest platelets.
However, we found that light scattering changed within 15
seconds of platelet activation by ADP and U46619, so that
in experiments in which aggregation was inhibited by
EDTA (5 mmol/L), there was an increase in forward
scattering (Fig 2E). Right angle scattering and forward
scattering became more linearly related than in the resting
state with either ADP or U46619. These changes are also
consistent with the report of Holme et alZ9and probably are
related to platelet “shape change,” although it is not clear
what aspects of a change in shape are reflected in the
scattering of light. They do not appear to be a simple
function of platelet size.
In any event, the heterogeneity in F-actin content is not
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
FILAMENTOUS ACTIN IN PLATELETS
927
independent of heterogeneity in other aspects of platelet
function.’ l6 Results obtained through averaging techniques
such as biochemical analysis should be interpreted with
caution, just as was true for heterogeneity in [Ca2+],.11-13
in
This relatively simple method may be
studies to detect a subtle prothrombotic state, such as that
which may occur in diabetic patients?6 or to monitor the
effects of storage of platelets, in which the F-actin content
of individual platelets may serve as an index of their
functional integrity.’
ACKNOWLEDGMENT
We thank Dr Stuart Schlossman for generously allowing us to
use an EPICS v flow cytometer in his laboratory. The kind
donations of monoclonal antibodies by Drs Rodger P. McEver and
Thomas J. Kunicki are also gratefully acknowledged.
REFERENCES
1. Lindberg U, Markey F Platelet microfilaments and motility,
in MacIntyre DE, Gordon JI (eds): Platelets in Biology and
Pathology 111. Amsterdam, The Netherlands, Elseveier/North
Holland, 1987, p 205
2. Fox JEB, Phillips DR: Polymerization and organization of
actin filaments within platelets. Semin Hematol20:243,1983
3. Nakano T, Hanasaki K, Arita H: Possible involvement of
cytoskeleton in collagen-stimulated activation of phospholipases in
human platelets. J Biol Chem 264:5400,1989
4. Michelson AD, Ellis PA, Bamard MR, Matic GB, Viles AF,
Kestin AS: Downregulation of the platelet surface glycoprotein
Ib-IX complex in whole blood stimulated by thrombip, adenosine
diphosphate, or in vivo wound. Blood 77:770,1991
5. Spangenberg P, Schymik CH, Hofmann B, Ostermann G,
Ruhling K, Till U: Blood platelet behaviour in patients with type I
diabetes mellitus. Exp Clin Endocrinol93:329,1988
6. Spangenberg P, Till U: Actin filament content in platelets-A
sensitive index of cellular activity. Biosci Rep 9307,1989
7. Feinberg H, Sarin MM, Batka EA, Porter CR, Miripol JE,
Stewart M: Platelet storage: Changes in cytosolic Caz+, actin
polymerization and shape change. Blood 72:766,1988
8. Blikstad I, Markey F, Carlsson L, Persson T, Lindberg U:
Selective assay of monomeric and filamentous actin in cell extracts,
using inhibition of deoxyrybonuclease I. Cell 15:935,1987
9. Thompson CB, Moloy RL, Skelly RR, Quin PG, Jakubowski
JA. The biology of platelet volume, in Martin J, Trowbridge A
(eds): Platelet Heterogeneity-Biology and Pathology. Amsterdam, The Netherlands, Elseveier/North Holland, 1990, p 25
10. Corash L: Platelet density, heterogeneity and platelet ageing, in Martin J, Trowbridge A (eds): Platelet HeterogeneityBiology and Pathology. Amsterdam, The Netherlands, Elseveier/
North Holland, 1990, p 1
11. Davies TA, Drotts DL, Weill GJ, Simmons ER: Flow
cytometric measurement of cytoplasmic calcium changes in human
platelets. Cytometry 9:138,1988
12. Jennings LK, Docker ME, Wall CD, Fox CF, Kennedy DM:
Calcium mobilization in human platelets using Indo-1 and flow
cytometry. Blood 74:2674,1989
13. Oda A, Daley JF, Kang J, Smith M, Ware JA, Salzman EW:
Quasi-simultaneous measurement of ionized calcium and a-granule release in individual platelets. Am J Physiol260:C242,1991
14. Jackson CW, Jennings L K Heterogeneity of fibrinogen
receptor expression on platelets activated in normal plasma with
ADP: Analysis by flow cytometry. Br J Haemetol72407,1989
15. Shattil SJ, Hoxie J, Cunningham M: Detection of activated
platelets in whole blood using activation-dependent monoclonal
antibodies and flow cytometry. Blood 70:173, 1987
16. Johnston GI, Picket EB, McEver RP, George JN: Heterogeneity of platelet secretion in response to thrombin demonstrated by
fluorescence flow cytometry. Blood 69:1401,1987
17. Wallace PJ, Westtro RP, Packman CH, Lichtman M A
Chemotactic peptide-induced changes in neutrophile actin conformation. J Cell Biol99:1060,1984
18. Packman CH, Lichtman M A Activation of neutrophils:
Measurement of actin confomational changes by flow cytometry.
Blood Cells 16:193, 1990
19. Miyazaki M, Mely-Goubert B, Leiserson WM, Dudler L,
Trnka Z Actin activity is high in the immunocompetent fraction of
tymocytes. Thysmus 6:57,1984
20. McEver RP, Martin M N A monoclonal antibody to a
membrane glycoprotein binds only to activated platelets. J Biol
Chem 259:9799,1984
21. Stenberg PE, McEver RP, Shuman MA, Jacques W,
Bainton D F A platelet alpha-granule protein (GMP-140) is
expressed on the plasma surface after activation. J Cell Biol
101:880,1985
22. Daniel JL, Molish IR, Rigmaiden M, Stewart G: Evidence
for a role of myosin phosphorylation in the initiation of the platelet
shape change response. J Biol Chem 259:9826,1984
23. Frojmovic MM, Milton JG: Human platelet size, shape, and
related functions in health and disease. Physiol Rev 62:185,1982
24. Steer ML, Salzman E W Cyclic nucleotides in hemostasis
and thrombosis. Adv Cyclic Nucleotide Res 12:71,1980
25. Siess W Molecular mechanism of platelet activation. Physio1 Rev 69:58,1989
26. Fox JEB, Phillips DR: Inhibition of actin polymerization in
blood platelets by cytochalasins. Nature 293:302, 1981
27. Castella JF, Flamagan MD, Lin S: Cytochalasin D inhibits
actin polymerization and induces depolymerization of actin filaments formed during shape change. Nature 293:302,1981
28. Jennings LK, Ashman RA, Winfred C W , Dockter ME:
Analysis of human platelet glycoprotein IIb-IIIa and Granzmann’s
thrombasthenia in whole blood by flow cytometry. Blood 69:173,
1986
29. Holme S, Heaton A, Konchuba A, Hartman P: Light
scattering and total protein signal distribution of platelets by flow
cytometry as parameters of sizes. J Lab Clin Med 112:223,1988
30. Born GVR: Observations on the change in shape of blood
platelets brought about by adenosine phosphate. J Physiol209487,
1970
31. Peerschke EIB: Caz+mobilization and fibrinogen binding of
platelets refractory to ADP stimulation. J Lab Clin Med 106:111,
1985
32. Shuttleworth TJ, Thompson J L Effect of temperature on
receptor-operated changes in [Caz+],and their determination using
fluorescent probes. J Biol Chem 266:1410,1991
33. Salzman EW, Ware J A Ionized calcium in blood platelets,
in Jamieson GA (ed): Platelet Membrane Receptor-Molecular
Biology, Immunology, Biochemistry, and Pathology. New York,
NY,Lis, 1988, p 419
34. Stossel TP: From signal to pseudopod. J Biol Chem 264:
18261,1989
35. Goldschmidt-Clermont PJ, Machesb LM, Doberstein SK,
Pollard TD: Mechanism of interaction of human platelet profilin
with actin. J Cell Biol113:1081, 1991
From www.bloodjournal.org by guest on June 18, 2017. For personal use only.
1992 79: 920-927
Heterogeneity in filamentous actin content among individual human
blood platelets
A Oda, JF Daley, C Cabral, JH Kang, M Smith and EW Salzman
Updated information and services can be found at:
http://www.bloodjournal.org/content/79/4/920.full.html
Articles on similar topics can be found in the following Blood collections
Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests
Information about ordering reprints may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprints
Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American
Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.