Rapid and Selective Measurement of Platelet-Activating Factor Using a Quantitative Bioassay of Platelet Aggregation

JEAN AMMIT AND CHRISO’NEILL

ALAINA

A bioassay

for the measurement

quantification analyzer

in conjunction

collected and

of platelet

with

used

to quantitate

platelet factor,

pable

occurred

aggregation

FL) of titrated

rabbit

This enabled volume

whole

multiple

of whole

of platelet-activating and adenosine

acid-

no affect method

whole

blood

and adenosine

of platelet

induced inhibited

coefficients

the

inhibitors

need

were

inhibited

is,

to the arachbut

factor.

antagonist within

acidthat

aggregation,

factor was reproducible;

of variation

for a large

measurement

respectively,

by platelet-activating

(50

period.

activation,

These

platelet

by its receptor

was ca-

of arachidonic

kinase, bioassay.

aggregation

incubation

Selective

pathways before

platelet-activating

and intrabioassay

Key Words:

without

diphosphate-induced

factor was selectively

15-min)

the rabbit.

ECSo were

in small volumes

by adding

immediately

aggregation

for measuring

13.17% and 9.75%,

Whole

from

The

The bioassay

aggregation

to be performed

to be collected

blood

diphosphate

the half maximal

respectively.

using a short (i.e.,

factor was achieved

on platelet

activating

by any activator.

acid and phosphoenolpyruvate/pyruvate

rabbit

idonic

induced

of platelet

diphosphate-dependent

acetylsalicylic titrated

bioassays

blood

and whole

acid, and adenosine

concentrations

blood,

device

on the

used a platelet

This bioassay can be nonselective

and 12.5 min,

quantification

(PAF) based

The method

micromixing

rabbits.

arachidonic

at 7.5, 10.0,

of the sensitive

inter-

White

55, and 10 ~.LM, and at these

response

factor

was developed.

a multiwelled

from male New Zealand

for platelet-activating 0.0232,

of platelet-activating

aggregation

had

Platelet-

BN 52021. This at the ECSO, the

acceptable

limits

at

respectively.

PAF; Quantitative

bioassay;

Platelet

aggregation;

Platelet

analyzer;

blood

INTRODUCTION

Platelet-activating factor (PAF) (l-O-alkyl-2-acetyl-sn-glyceryl-3-phosphocholine) is a biologically active ether phospholipid with a wide spectrum of biological

ac-

tivities.” It has been implicated as a mediator in a number of pathological conditions (Braquet et al., 1987) and, physiologically, may be involved in embryo development and implantation (O’Neill et al., 1989). A complete understanding of PAF’s role in a EC50 activator concentration required to induce 50% platelet aggregation. lCSOinhibitor concentration required to inhibit platelet aggregation by 50%. From the Human Reproduction Unit, Royal North Shore Hospital of Sydney, St. Leonards, New South Wales,

Australia.

Address of Sydney, Received

reprint

requests

St. Leonards, September

to: Dr. Chris

O’Neill,

2065, New South 5,199O;

revised

Wales,

Human

Reproduction

Unit,

Royal North

Shore

Hospital

Australia.

and accepted December

20,199O. 7

Journalof Pharmacological Methods 26, 7-21 (1991) 0 1991 Elsewer Science Publishing

Co., Inc., 655 Avenue of the Americas. New York, NY 10010

8

A. j. Ammit and C. O’Neill these

processes

has been

hindered

by the difficulty

in rapidly

and quantitatively

measuring the concentration of PAF in biological fluids. Common methods suring PAF are based on the ability of PAF to induce platelet aggregation. Platelet

aggregation

can be qualitatively

assessed

by turbidometric

for mea-

(Born,

1962)

or electronic impedance aggregometers (Cardinal and Flower, 19801, or quantitatively measured by platelet counters (lumley and Humphrey, 1981; Saniabadi et al., 1983; Splawinski et al., 1984). Aggregometers in platelet-rich plasma or washed platelets. large volumes

of whole

blood

generally measure platelet aggregation This involves time delay and requires

to be collected

from

the results are, at best, only semiquantitative. Therefore, the aim of this study was to develop measurement blood. welled

of PAF, which

the experimental a quantitative

was both rapid and required

This was achieved by using a platelet micromixing device.

analyzer

minimal

animal,

bioassay volumes

in conjunction

with

and

for the of whole a multi-

METHODS

Platelet Analyzer The Baker Instruments Series 810 Whole Blood Platelet Analyzer ments, Allentown, Pennsylvania, USA) is a microprocessor-controlled,

(Baker Instrusemiauto-

mated system for the enumeration and sizing of platelets in whole blood. It incorporates a volume-discriminating threshold system, preset using a suspension of uniform latex particles, which allows the instrument to discriminate between platelets and larger cells, or smaller platelets

noise

0

4

8

12

cell fragments,

or precipitates

macro platelets

16

20

CUBIC

24

28

(Figure I). The volume

micro rbcs

32

36

40

rbcs

44

48

MICRONS

FIGURE 1. Cell population distribution graph for rabbit whole blood. The solid lines on the graph represent the volume discriminators. The first line from the left of the graph is the lower threshold (3 )rm3), and the second line is the upper threshold (24 um3). (Reproduced from the Baker Instruments Series 810 Platelet Analyzer Operators Manual with permission.)

A Quantitative teflon

water

block

well

Bioassay for PAF

SDinbar

?y

I

hollow

channel

Teflon multiwelled

FIGURE 2.

I

motor

assembly

temperature-controlled

micromixing device.

discriminators are set so that predominantly single, nonaggregated are counted (3-24 km3). Upon exposure to an activator, platelets that are larger than the upper

volume

discriminator

rabbit platelets form aggregates

and are therefore

not counted.

The resulting decrease in the number of single platelets is directly proportional the degree of platelet aggregation and can be used to quantify the response. Samples titrated capillary

were

prepared

for platelet

counting

by a l-1001

manual

rabbit whole blood. The dilutions were accomplished tube with blood (Microcaps: Drummond Scientific,

dilution

of the

by filling a IO-FL Broomall, Pennsyl-

vania, USA), removing the.excess blood with lint-free tissues (Kimwipes: Clark, Sydney, NSW, Australia), and expelling it into diluent reservoirs 10 mL of sheath fluid. Sheath fluid is an azide-free 3.8 mM KCI, 16.5 mM NaZHP04, 1.9 mM KH2P04,

to

Kimberleycontaining

isotonic buffer (139 mM NaCl, 1.0 mM ethylenediaminetetra-

acetate (EDTA) disodium salt, 1.01 mM LiCI: Sigma Chemicals, St. Louis, Missouri, USA) prepared in double-distilled water and sterilized by filtration (Sterivex-CS (0.22 km):

Millipore,

Micromixing

Bedford,

USA).

Device

The micromixing containing

Massachusetts,

eight

device wells

(Figure

(height

=

2) was constructed 10 mm,

radius

=

to consist 3.2 mm,

of a teflon

volume

=

block

322 PL),

surrounded by a channel and positioned above a motor assembly that allows controlled and consistent magnetic stirring within each individual well. Heated water was pumped

through

the channels

and maintained

temperature by connecting a thermostatted GmbH, Seelbach, FRG) to the micromixing

at the appropriate

incubation

circulator (Julabo EM: Labortechnick device. Within each well, mixing was

accomplished by a 3 mm x 3 mm teflon-coated magnetic stirring rod (Spinbar: Belart Products, Pequannock, New Jersey, USA), reducing the effective volume of each well to approximately 300 FL. The stirring speed was routinely maintained at approximately 500 rpm (0.42 g). This stirring speed does not cause damage to platelets being much less than those recommended for preparation of platelet-rich plasma

9

10

A. J. Ammit and C. O’Neill by centrifugation

prior to platelet-function

al., 1981). The micromixing distilled water and applied polished

(i.e.,

150 g) (Roper-Drewinko

et

dried

with

a cotton-tipped applicator (single-ended wooden stem: Smith and Nephew, bourne, Vie., Australia) covered by a lint-free tissue. The magnetic stirring

Melrods

were

to minimize

testing

device was cleaned by extensive rinsing with doublewith a Pasteur pipette, the tip of which had been fire-

also rinsed with

tearing

of the teflon

double-distilled

well surface.

water

and dried

The wells were

with

lint-free

tissues.

Blood Collection Blood was collected bits. Following Vie., Australia)

from the marginal

ear vein of male New Zealand

White

rab-

topical application of xylene (AR grade: BDH Chemicals. Melbourne, to cause vasodilation, a 20G teflon catheter (Surflo: Terumo, Tokyo,

Japan) was inserted, and blood was collected into iced IO-mL glass tubes (Vacutainer Brand evacuated blood collection tubes: Rutherford, New Jersey, USA). These contained 0.4 mL of 3.2%

sterile, siliconized Becton-Dickinson, (wt/vol) citrate (tri-

sodium

(4 parts blood

salt-dihydrate:

part citrate). The titrated croscope

rabbit whole

examination

abnormalities,

Sigma)

and were

blood

showed

filled

was suitable

no significant

and if the platelet

count

to a 2-mL mark for bioassay platelet

was greater

to 1

if a phase-contrast

aggregation

or blood

mismear

than 200 x 103/~L.

Activators PAF (L-alpha-phosphatidylcholine,

beta-acetyl-gamma-0-alkyl:

pared as a I-mg/mL stock solution in chloroform aliquants were removed, the solvent was evaporated

Sigma)

was

pre-

(AR grade: BDH). Before use, with NZ, and PAF was dissolved

and diluted in phosphate-buffered saline (PBS) (140 mM NaCl, 26 mM KCI, 8.2 mM NazHP04, 1.5 mM KH2P04: Sigma) containing 0.25% (wt/vol) bovine serum albumin (PBS-BSA) (CSL, Melbourne, Vie, Australia). Arachidonic acid (AA; from porcine liver: NaCl

Sigma) was dissolved in IO mM Na2C03 (Sigma). Adenosine diphosphate (ADP;

muscle:

Sigma) was dissolved

and diluted

(Sigma) and diluted sodium salt, grade

in NaCl.

in 0.9% (wt/vol) I, from equine

See text for concentrations

used.

Inhibitors BN 52021 (9H-1,7a-(epoxymethano)-lH,6a-H-cyclopenta-(c)furo(2,3-b)furo(3’,2’: 3,4) cyclopenta (1,2-d)furan-5,9,12-(4H)-trione,3-tert-butylhexahydro-4,7b,-ll-hydroxy-g-methyl: IHB-IPSEN, Le Plessis Robinson, France) was dissolved in methanol (AR grade: BDH) and diluted in PBS. Acetylsalicyclic acid (ASA; aspirin: Sigma) was dissolved with heat and diluted in NaCl. Phosphoenolpyruvate (PEP: trisodium saltheptahydrate: Sigma) and pyruvate kinase (PK; type II, from rabbit muscle, E.C. 2.7.1.40: Sigma) were dissolved and diluted in PBS-BSA. See text for concentrations used.

Activator Dose-Response A 50-FL aliquot of titrated rabbit whole blood was placed micromixing device. After a 30-see temperature equilibration

into each well of the period, 50 FL of PAF,

A Quantitative AA, or ADP (or as a control, was taken immediately single, nonaggregated

an equal volume

of vehicle)

and after 15 min of incubation platelets by the platelet analyzer.

was added.

Bioassay for PAF A lO+L

sample

at 37°C for counting of the The results were expressed

as the platelet aggregation index, that is, 1 - P15/P0, where PI5 and POare the platelet count at 15 and 0 min, respectively. Therefore, an index of zero signifies no platelet aggregation Activator

and conversely, Time

1 is total platelet

aggregation.

Kinetics

A IOO-~.LL aliquot of titrated rabbit whole blood was placed into each well of the micromixing device. After a 30-set temperature equilibration period, 100 PL of the ECso for PAF, AA, or ADP was added and a lO+L sample was taken immediately to give the platelet count at 0 min. Additional lO+.L samples were taken at 2.5, 5.0, 7.5, 10.0, 12.5, and 15.0 min. As a control, the same protocol was performed following the addition of 100 FL of PBS-BSA. The single, nonaggregated platelets were counted on the platelet aggregation index. Inhibitor Potency

analyzer

and the

results

were

expressed

as the

platelet

Potency was measured

PEP/PK on platelet tively.

by observing

aggregation

A 50+L aliquot of titrated micromixing device followed

the inhibitory

induced

effect

of BN 52021, ASA, and

by the ECso for PAF, AA, and ADP,

respec-

rabbit whole blood was placed into each well of the by either 25 FL of BN 52021, or ASA, or 12.5 PL each

of PEP and PK (or as a control, an equal volume of vehicle). After a 30-set temperature equilibration period, 25 PL of the E&, for PAF, AA, or ADP (or as a control, an equal volume viously.

of vehicle)

Inhibitor Selectivity aggregation

was added,

and the bioassay

was processed

as described

pre-

Selectivity was assessed by monitoring the effect of the inhibitors induced by PAF, AA, and ADP each used at their E&,.

A 50-PL aliquot

of titrated

rabbit

whole

blood

was placed

on platelet

into each well

of the

micromixing device followed by the I&,,, of each inhibitor, that is, 25 ~J,Lof BN 52021, or ASA, or 12.5 ~.LLeach of PEP and PK (or as a control, an equal volume of vehicle). After a 30-set temperature equilibration period, 25 ~.LLof the activator ECso (or as a control, an equal using the t test.

volume

of vehicle)

was added.

Statistical

analysis was performed

Effect of ASA and PEPIPK on the PAF Dose-Response To inhibit AA- and ADP-induced platelet aggregation, ASA and PEP/PK were prepared as stocks. Aliquots (IO FL) were added to 500 PL of titrated rabbit whole blood, immediately before bioassay, to obtain the appropriate I&, final concentrations. To determine whether these inhibitors had an effect on PAF-induced aggregation,

the

response

of platelets

to PAF (9.3-929.4

nM)

was examined

in the

11

12

A. J. Ammit and C. O’Neill 09

-

08

-

07

-

06

-

05

-

2 s 5 5 :

0001

0.0 I

0.1

IO

I ACTIVATOR

100

1000

control

(PM)

FIGURE 3. Platelet aggregation induced by PAF (m), AA (0) and ADP (+I, after 15 min of incubation with titrated rabbit whole blood. Each point is the mean f SEM of greater than seven replicates. The effect of the vehicle alone is represented by the histogram (control), which is the mean + SEM of 141 replicates.

presence/absence

of ASA and PEP/PK. Statistical

analysis was performed

by two-way

analysis of variance.

RESULTS Bioassay Validation PAF was a potent were several orders ADP were

0.023,

~.LM, respectively

activator, inducing platelet aggregation at concentrations that of magnitude less than AA and ADP. The ECso for PAF, AA, and

55, and 10 ~.LM, and the detection (Figure

limits were

0.00093,

10, and 1

3).

At ECso, PAF was the most rapid activator, taking 7.5 min to reach half maximal response, compared with 10.0 and 12.5 min for AA and ADP, respectively (Figure 4). Platelet aggregation induced by the ECso for PAF was inhibited by BN 52021 (I&.,, 2.4 FM) (Figure 5). Aggregation induced by the ECso for AA was inhibited by ASA (I& 0.1 mM) (Figure 6), and the ADP EC5,-,was inhibited by PEP and PK (ICsO 0.25 mM and 2.5 iu/mL, respectively) (Figure 7). At these concentrations, the inhibitors were selective (Figure 8, 9, IO). When the lCsO for ASA and PEP/PK were established in the titrated rabbit whole blood they had no significant inhibitory effect on the

A Quantitative

Bioassay for PAF

0.7 -

z 9 g z Kl w Yi 2 L ti % 2

0.6

-

0.5

-

0.4

-

0.3

-

0.2

-

0.1

-

0.0 -

1

1

1

I

1

2.5

5.0

7.5

10.0

12.5

1

15.0

TIME (minutes) FIGURE 4. Time kinetics of platelet aggregation induced by the El% for PAF (23.2 nM; n ), AA (55 pM; 0) and ADP (10 PM; +) or vehicle alone (control; V), during 15 min of incubation with titrated rabbit whole blood. Each point is the mean + SEM of less than six replicates.

E c a Cl w OL z a l5 K a d

0.4

-

0.3

-

0.2

-

0.1

-

.

o.oL

1 I 0 0.6

J

I

1.2

f3N 5202

2.4

control

1 (PI-I)

FIGURE 5. Effect of increasing concentrations of BN 52021 on platelet aggregation induced by the ECsofor PAF (23.2 nM; n ) after 15 min of incubation with titrated rabbit whole blood. Each point represents the mean + SEM of greater than three replicates. The effect of the vehicle alone is represented by the histogram (control), which is the mean f SEM of 11 replicates.

13

25 s? z G ;: hii is a IY g a iii!

0.6

-

0.5

-

0.4

-

0.3

-

0.2

-

0.1

-

+

00.05

0.1

0.25

control

ASA (mt+l) FIGURE 6. Effect of increasing concentrations of ASA on platelet aggregation induced by the E&, for AA (55 PM; 01, after 15 min of incubation with titrated rabbit whole blood. Each point is the mean + SEM of greater than three replicates. The effect of vehicle alone is represented by the histogram (control), which is the mean 2 SEM of three replicates.

0 0,05

0.125

0.25

0.625

PEP CrnM) 1 0

I 0.5

I

t

L

J

.25

2.5

6.25

PK (iu/mL)

control

A Quantitative

Bioassay for PAF

EC50

IEI

EC,, + ‘Go

06

-

is os25 -

I5 F

0.4 -

s E 0.3 iii a IY 0.2 E a d

01

-

00 i

-!#zL PAF

AA

ADP

control

FIGURE 8. Effect of the I&, for BN 52021 on platelet aggregation induced by the ECso for PAF (22.2 nM), AA (55 PM), and ADP (10 FM), after 15 min of incubation with titrated rabbit whole blood. Each histogram represents the mean + SEM of greater than three replicates. The effect of vehicle alone is represented by the histogram (control), which is the mean * SEM of three replicates. * denotes significant difference (p < 0.05).

PAF dose-response curve (Figure 11). The bioassay was reproducible as at the PAF EC&, the coefficient of variation was 13.17% (n = 208) interbioassay and 9.75% (n = 8) intrabioassay. DISCUSSION The development of the Born (1962) turbidometric aggregometer for measuring platelet aggregation facilitated the laboratory investigation of platelet physiology and is one of the most common methods used for studying platelet behavior. Although it has provided much valuable information on platelet function and physiology, and has proven diagnostic value, this system is only qualitative and has other inherent limitations. The most notable of these is the need for preparation of large volumes of translucent platelet suspensions. Platelet aggregation can only be measured in platelet-rich plasma, or after isolation of platelets from plasma proteins . FIGURE 7. Effect of increasing by the EC& for ADP (10 PM; +), Each point is the mean + SEM is represented by the histogram

concentrations of PEP/PK on platelet aggregation induced after 15 min of incubation with titrated rabbit whole blood. of greater than three replicates. The effect of vehicle alone (control), which is the mean -C SEM of six replicates.

15

16

A. J. Ammit and C. O’Neill

EC,,

I23

EC50

06

-

z is -

OS-

2g

04-

z E

03

-

2 t;

02

-

01

-

00

-

d kd

+ ‘C50

PAF

BL b AA

ADP

control

FIGURE 9. Effect of the KS0 for ASA (0.1 mM) on platelet aggregation induced by the ECSO for PAF (29.2 nM), AA (55 PM), and ADP (10 PM), after 15 min of incubation with titrated rabbit whole blood. Each histogram represents the mean f SEM of greater than three replicates. The effect of vehicle alone is represented by the histogram (control), which is the mean f SEM of three replicates. * denotes significant difference (p < 0.05).

using either gel filtration (Tangen et al., 1971), albumin density gradient separation (Walsh, 1972) or further centrifugation to obtain washed platelets (Ardlie et al., 1970). Centrifugation of the blood sample is required for the production of platelet-rich plasma. This involves time delay, platelet trauma and possible removal of up to 30% of the platelets. Most of the lost platelets are removed with the erythrocyte fraction, suggesting that they are the heaviest (Zwierzina and Kunz, 1985). Haver and Gear (1981) reported that heavier platelets have greater metabolic and functional capabilities and are more responsive to platelet aggregating stimuli. Methods used to isolate platelets from plasma proteins first require platelet-rich plasma preparation and, therefore, they suffer all the drawbacks associated with platelet-rich plasma preparation plus their own unique problems. In the case of gel filtration, alterations in platelet characteristics can occur from contact with the gel matrix (Linden et al., 1976). Using the albumin density gradient separation technique, one must wash the platelets free of the gradient material and variable amounts of albumin can contaminate the platelet preparation (Parker et al., 1984). Washed platelet preparation involves further centrifugation to pellet platelets. The pellet is then resuspended in a physiological buffer. This procedure has been reported to cause platelet activation (Parker et al., 1984). The use of prostacyclin throughout the procedure has been suggested to prevent this activation (Vargas et al., 1982).

A Quantitative

cl q 06 x

g

OS-

Bioassay for PAF

EC,, EC,, + 'C50

1

5 I= g

04-

5 E

03-

5 L

02

-

d k d

Ol-

00

PAF

AA

ADP

control

FIGURE 10. Effect of the I&, for PEPlPK (0.25 mM and 2.5 iu/mL) on platelet aggregation induced by the ECso for PAF (22.2 nM), AA (55 uM), and ADP (10 PM), after 15 min of incubation with titrated rabbit whole blood. Each histogram represents the mean zk SEM of greater than three replicates. The effect of vehicle alone is represented by the histogram (control), which is the mean + SEM of six replicates. * denotes significant difference (p < 0.05).

The electrical

impedance

for the measurement

aggregometer

of aggregation

(Cardinal

in opaque

and Flower,

platelet

1980) can be used

suspensions,

such as whole

blood, and, therefore, circumvents most of these problems. Measurements gregation could be made on volumes as small as 100 ~.LLenabling multiple to be performed without the need for large volumes of blood to be collected sell-smith et al., 1981). A disadvantage of this method is that only qualitative pretation of the data is possible. can be observed microscopically (Swart et al., 1984). There were a number

of reports

of agassays (Rusinter-

It also has limited sensitivity, as small aggregates where no change in impedance has occurred (Lumley

and Humphrey,

1981; Saniabadi

et al.,

1983; Splawinski et al., 1984) describing the use of platelet analyzers to measure platelet aggregation. These methods were disadvantaged, however, because they used conventional tubes requiring large volumes (i.e., 0.4-I mL) of whole blood. To overcome this limitation, we used a platelet analyzer in conjunction with a multiwelled micromixing device. The unique design of the multiwelled micromixing device (Figure 2) allows the quantification of platelet aggregation in small volumes (i.e., 50 FL) of the whole blood. This enables multiple bioassays to be performed without the need for a large volume of whole blood to be collected from the experimental animal. Also, this method can process a larger number of samples be-

17

18

A. j. Ammit and C. O’Neill

I

1

I

I

929

92.9

23.2

9.3

PAF

4

(M-l)

FIGURE 11. Effect of titrated rabbit whole blood containing ASA and PEP/PK (at I&, final concentrations) on platelet aggregation induced by PAF after 15 min incubation. 0 represents whole blood with vehicle alone, whereas W represents whole blood with ASA and PEP/PK. Each point is the mean + SEM of >ll replicates.

cause the micromixing

device

time. This is an advantage

has eight wells allowing

over the turbidometric

multicomparisons

and electronic

at the same

aggregometers

that

are either single or dual channel. The platelet analyzer incorporates a volume-discriminating threshold system allowing the instrument to discriminate between platelets and larger cells, or small cell fragments, or precipitates (Figure nators were set so that predominantly

1). For this application single, nonaggregated

the volume discrimirabbit platelets were

counted (3-24 km3). The method was based on the measurement of %P151P0, that is, the percentage of single, nonaggregated platelets in the region 3-24 pm3 at 15 min (PI,) compared with those in the same region at 0 min (P,). Therefore, the method does not determine absolute platelet size and number, rather a platelet aggregation index. If, for example, two small platelets (say, 3 pm3) aggregated, then they would not be excluded by the upper volume discriminator (24 km3), but would be counted as one 6-pm3 platelet. This would lower the %PIS/PO, indicating platelet aggregation, but would result in a loss of bioassay sensitivity. To address this problem, the volume discriminators could There are large species differences

be narrowed. in platelet sensitivity

to PAF, the source

of

A Quantitative whole

blood

rabbits,

dog,

is important. horse,

The most sensitive

platelets

and cat have approximately

are guinea

equal

Bioassay for PAF

pig. Platelets

sensitivity.

Human

from

platelets

are less sensitive. Rat and mouse platelets are thought to be totally unresponsive (Namm et al., 1982). The species variation has been correlated with differences in the number

of PAF receptors

on platelets

(Inarrea

et al., 1984). We found

that the

rabbit was a convenient donor of blood, giving samples that provided adequate sensitivity (0.93 nM) in the bioassay, yet rarely underwent nonspecific clotting or aggregation. The use of the teflon catheter for blood collection, together with direct collection

into cold citrate,

gation. In rabbit aggregation

was important

to help prevent

such nonspecific

aggre-

whole blood, PAF was a potent activator (Figure 3), inducing platelet at concentrations that were several orders of magnitude less than AA

and ADP. The results were consistent with those obtained using turbidometric aggregometry with rabbit platelet-rich plasma, that is, EG for PAF, AA, and ADP were 0.06 PM (Vargaftig et al., 1981a), 50-100 et al., 19861, respectively. In agreement platelet

aggregation

the bioassay tivators.

(Figure 4). Although

was capable

By investigating

~.LM(Vargaftig et al., 1981 b), and 5 PM (Galvez with lnarrea et al. (19841, PAF rapidly induced the responses

of quantifying

the dose-response

platelet and time

to AA and ADP were

aggregation kinetics

induced

of platelet

slower,

by these

aggregation

acin-

duced by PAF, AA, and ADP, it was shown that the bioassay could sensitively quantify platelet aggregation in small volumes (50 FL) of titrated rabbit whole blood, using a short (i.e., 15-min) incubation period. Therefore, a quantitative bioassay of platelet aggregation was developed and validated. Platelet activation involves a number of cellular

responses,

including

1) the pro-

duction and release of activators, such as PAF and AA, and 2) the discharge of ADP upon granule secretion. PAF can be inhibited by receptor antagonists, such as the terpene BN 52021 (Braquet et al., 1987), the effects of AA are blocked by the cyclooxygenase inhibitor ASA (Chignard version to ATP using PEP/PK (Haslam, PEP/PK are selective PAF-induced

platelet

PEP/PK to the titrated

inhibitors.

et al., 1979), and ADP is removed by its con1964). In this bioassay BN 52021, ASA, and

To allow

aggregation, rabbit whole

The results show that PAF-induced

the rapid

the bioassay blood

and selective

was modified

immediately

platelet

aggregation

before

quantification

by adding

of

ASA and

bioassay.

can occur in the presence

of cyclooxygenase inhibitors and ADP-removing systems. This is in agreement with previous reports (Cazenave et al., 1979; Chignard et al., 1979). The apparent activator independence observed with rabbit platelets led these early authors to hypothesize that platelets were activated by three distinct pathways; the first involving the secretion of ADP; the second mediated by products of cyclooxygenase action on AA; and the third pathway was due to PAF. However, when platelet aggregation was examined in human platelets, conflicting results were obtained, questioning the independence of these pathways (reviewed by Sturk et al., 1987). It is now generally agreed that activators released and secreted upon platelet activation act in a synergistic Altman

manner to amplify et al., 1986).

A radioimmunoassay

the response

to the initial

for PAF is now available

activator

(NEK-062:

NEN

(Sturk

et al., 1985;

Du Pont,

Boston,

19

20

A. J. Ammit and C. O’Neill MA,

USA).

It is yet to be fully validated

ison with this bioassay allowed (Ammit and O’Neill, in press). Because

platelet

analyzers

for biological

attempts

samples.

at validation

can enumerate

platelets

Therefore,

of the

in whole

compar-

radioimmunoassay blood,

this method

eliminates the time delay involved in the preparation of platelet-rich plasma and washed platelets required for most other indirect methods for measuring PAF. The disadvantage

of using whole

blood

is that acetylhydrolase

would

be present

in the

plasma (Blank et al., 1981). This enzyme removes the acetyl moiety from the C2 position of PAF, resulting in nonactive lyso-PAF (Stafforini et al., 1987). Therefore, the platelet response to PAF in whole blood and platelet-rich plasma may be less sensitive than that in washed platelets. As previously discussed, the sensitivity of this method is adequate, however, in order to minimize the effect of this enzyme a short (i.e., 15-min) incubation period is used, whole blood is collected from rabbits because

their platelets

PAF controls

are highly sensitive

are performed

The PAF receptor

to PAF (Namm

in every bioassay

antagonist

BN 52021 can be used in the PAF selective

as a convenient means of confirming that platelet providing rapid pharmacological characterization. Therefore,

et al., 1982), and synthetic

run.

as the quantitative

bioassay

of platelet

bioassay

aggregation

was due

to PAF,

aggregation

developed

in this

paper is capable of the rapid and selective measurement of PAF, it may be an important tool in the continuing research into the role of PAF in health and disease. The appropriate use of inhibitors means that it can be routinely used for monitoring platelet aggregation by many agents, for example, antibodies, drugs, and so on. Its use of small volumes of whole blood would make it particularly suitable for monitoring platelet aggregation ex vivo when repeated sampling is necessary and should be valuable for scale-up determination when multiple samples are required. Other blood constituents are known to produce both activators and inhibitors that can act to modulate platelet reactivity in vivo. Therefore, this bioassay may have clinical relevance a whole

because blood

activator-induced

in vitro bioassay

modulation

mimic

of platelet

responses

the in vivo situation

more

obtained

in

closely.

REFERENCES Altman

R, Scazziota

(1986) Synergistic dium tion. Ammit

arachidonate Thromb

tion,

osine

J, Cacchione

of PAF-acether

in human

platelet

C (1991) Comparison

R

and soaggrega-

of a radi-

and bioassay for embryo-derived

platelet-activitating

Ardlie

Rouvier

factor.

Human

Reproduc-

in press. NC,

in suspensions

MA,

Mustard

of washed

JF (1970) Aden-

platelet rabbit

aggregation platelets.

Br /

Haematol19:7-17.

specific

platelet-activating

(a hypotensive

lipid).

1 Biol Chem

and

256:175-

178.

Lee TC,

Born GVR (1962) Aggregation adenosine

P, Touqui

Perspectives search.

Fitzgerald

V, Snyder

F (1981) A

for I-0-alkyl-2-acetyl-

of blood

platelets

and its reversal.

Cardinal

L, Shen TY, Vargaftig

in

platelet-activating

Pharmacol DC,

gregometer:

acethylhydrolase

diphosphate

by

Nature

194:927-929. Braquet

Packham

diphosphate-induced

Blank ML,

sn-glycero-3-phosphocholine

Res 43:103-111.

AJ, O’Neill

oimmunoassay

A,

actions

behaviour 158.

Flower

BB (1987) factor

re-

electronic

ag-

Rev 39:97-145. RJ (1980) The

A novel device for assessing platelet in blood.

1 PharmacolMethods

3:135-

A Quantitative Cazenave

JP, Benveniste

gregation

J, Mustard

of rabbit platelets

factor is independent the

arachidonate

Chignard

of the release

drugs.

M, Le Couedic

Benveniste factor

and

reaction

and

is inhibited

by

Lab Invest 41:275-285.

JP, Tence

M, Vargaftig

aggregation.

Nature

BB,

Splawinski

A, Badimon

Electrical human,

L, Badimon

aggregometry

ter.

guinea

pig and rabbit.

blood

Thromb

from

Haemost

56:128-132. Haslam

RJ (1964)

Role of adenosine

thrombin

of human

disphosphate

blood

platelets

and by fatty acids. Nature

of platelets. lnarrea

/ Lab C/in Med

Crespo

platelet-activating animal Lindon

factor

species.

matrix

of

of different

Eur 1 fharmacol105:309-315.

JN, Rodvien contact

platelets

of the binding

to platelets

R, Waugh

during

filtration

Thromb

of

Haemost

human 36:311-

318.

Swart

P, Humphrey

quantitating

platelet

drug-receptor vitro. Namm

PPA (1981)

interactions

/ Pharmacol DH,

blood

in

6:153-166.

AS, High

of the platelet

for

and analyzing

in whole

Methods

Tadepalli

specificity

A method

aggregation

Species

to I-O-alkyl-

2-acetyl-sn-glycero-3-phosphocholine. Res 25:341-350. O’Neill

M, Ryan JP, Spinks NR (1989) Em-

bryo-derived

platelet-activating

Fert [Suppl.]

37:19-27.

Parker

RI, Rick ME, Gralnick minimize

isolation.

platelet

Thromb

Roper-Drewinko Johnston

D,

/ Reprod

HR (1984) A method

activation

during

platelet

Res 36:265-270.

PR,

Drewinko

McCredie

Standardization

factor.

Corrigan

KB, Freireich

of platelet

function

G,

EJ (1981) tests. Am /

Hemato/11:183-203. Russell-Smith

NC,

Measuring adhesion whole Saniabadi CRM,

platelet responses

blood. AR,

and

RJ, Cardinal leukocyte

in very

/ Pharmacol Lowe

Barbenel

IC

GDO, (1983)

small

Methods Forbes Platelet

DC

(1981)

Evaluation

volumes

of

6:315-333. CD,

Prentice

aeereeation

platelet

of the

M,

of plateletagonists

and release:

in

The role

cycle-oxygenase

ten Cate JW, Heymans factor:

H,

Borst

mediator

HSA,

P (1987)

of the third

aggregation?

1 C/in

D, Wood

alpha

aggregation

of electrical

0,

JK, Barnett

and

invest

DB (1984)

adrenergic in whole

impedance

an-

blood:

aggregome-

Res 36:411-418.

Berman

HJ, Marfey

P (1971) Gel-filtra-

tion: A new technique

for separation

from

Diath

plasma.

Thromb

of platelets

Haemorrh

25:268-

278. BB, Chignard

Wal F (1981a)

M,

J, Lefort

and present

platelet-activating

J,

status of

factor

(PAF-

Ann NY Acad Sci 370:119-137. BB, Chignard

noux B, Benveniste

M, Mencia-Huerta

metabolotes

ing factor

(PAF-acether).

and Pathology2. Science,

of ar-

and of platelet-activatIn Platelets

Ed., JL Gordon.

in Biology

Amsterdam:

El-

pp. 373-406.

of prostacyclin washing

JM, Ar-

J (1981 b) Pharmacology

achidonate

sevier

Benveniste

Background

on

acether).

of

M, Moncada

S (1982) The use

in the separation

from plasma and

human

platelets.

Prostaglandins

23:929-945. PN (1972)

ration platelet

aggregation/

Van Maanen

effects

Przyembel

on platelet

try. Thromb

Walsh Flower

MCL,

RBH,

Vargas JR, Radomski B,

factor

Res 40:359-372.

of platelet

tagonists

Vargaftig

C, Collier

products

Thromb

research

Thromb

other

of adrenaline

Vargaftig

JA (1982)

responses

and

SS, Pearson

Tangen

Lumley

of

coun-

79:34&350.

Effects

DF (1976) Effects of

gel

in plasma.

and

pathway

M,

262:4215-4222.

MCL,

aggregation

Platelet-activating

J, Nieto M, Sanchez

M (1984) Characteristics

Schapp

factor

platelet

Schutgens

97:187-204.

P, Comez-Cambronero

by platelet

ten Cate JW (1985) Synergistic

pathway.

Ziajor activity

platelet-activating

/ Biol Chem

Sturk A, Asyee GM, activating

W,

DE, Carter ME, Prescott SM

plasma

Sturk A, Schaap

fractionation

measured

DM, McIntyre

of ADP by

202:765-768.

Haver VM, Gear ARL (1981) Functional

Res

Res 34:93-102.

Human

human

in the aggregation

Thromb

Anti-aggregatory

blood

Thromb

(1987)

J-J, Fuster V (1986)

in whole

blood.

B, Ziembicki

B (1984)

acetylhydrolase.

Galvez

human

J, Gwodz

Splawinski

Stafforini

279:799-

800.

to

in whole

PGb in whole

J (1979) The role of platelet-activating

in platelet

studies

30:625-632.

by platelet-activating

pathway

membrane-active

IF (1979) Ag-

Bioassay for PAF

Albumin

and washing

density

of platelets

coagulant

activities.

gradient

sepa-

and the study of Br

/

Haematol

22 :205-217. Zwierzina platelet Thromb

WD,

Kunz

F (1985) A method

aggregation Res 38:91-100.

in

native

whole

of testing blood.

21