The effect of platelet-activating factor on the generation of superoxide anion in human eosinophils and neutrophils

The effect of platelet-activating factor on the generation of superoxide anion in human eosinophits and neutrophib Edward M. Zoratti, MD, Julie 9. Sedgwick, and William W. Busse, MD Madison, Wis.

PhD, Rose R. Vrtis, BS, MT,

The precise role of platelet-activating factor (PAF) in asthma has yet to be established. Nonetheless, the potential relationship between PAF and asthma appears to include the eosinophil (EOS) as an important link. Thus, to evaluate the effect of PAF on leukocyte-dependent inJammation, puriJied populations of human blood EOSs and neutrophils were isolated from the same subject. The two granulocyte populations were then incubated with PAF, and superoxide anion (O,-) generation was measured by reduction of cytochrome c in a microassay system. Both granulocyte cell types generated O,- when they were incubated with PAF. However, the generation of O,- was 3.4 times greater with EOSs (9.8 +- 1 .S nmole of cytochrome c reduced per 5 X 1oI cells) than neutrophils (2.9 5 0.4 nmole of cytochrome c reduced per 5 X 1t.Y cells; p < O.OOOl). When the effect of PAF on [Ca’ ‘1, was measured with the fluorescent label, MO-I, PAF caused similar increases in cellular fluorescence in both neutrophils and EOSs, but the increase in [Ca+ +I1 of neutrophils occurred with lower concentrations of PAF. Furthermore, when similar experiments were conducted in the presence of an extracellular calcium chelator, ethylene glycol-bis-(P-aminoethylether)-N,N,N’, N’-tetraacetic acid, there was partial suppression in both the cellular jiuorescence and O,generation to PAF; this suggests that full expression of EOS generation of O,- by PAF requires both intracellular mobilization and a transmembrane in&x of Ca’ + Our data indicate that PAF can stimulate leukocyte 02- generation, but this response is greater in the EOS than the neutrophil. Therefore, our findings support the observation that the EOS is more responsive to PAF activation than other granulocytes and that this difference muy contribute to participation of PAF in asthma (.I ALLERGY CLIN IMMUNOL 1991;88:749-58.) Key words: PAF, eosinophils,

superoxide generation,

PAF is a potent inflammatory mediator that is generated by a wide variety of celkL4 Although the role of PAF in the pathophysiology of asthma has yet to be established, certain properties of this mediator suggest that it is important. For example, administration of PAF to animals results in pulmonary inflammation and edema that is analogous to the histology of asthma.5. 6 Similarly in humans, PAF inhalation proFrom the Department of Medicine, Section of Allergy and Immunology, University of Wisconsin Medical School, Madison, Wis. Supported by National Institutes of Health Grants AI-23181, AI15685, and AI-26609. Received for publication Nov. 26, 1990. Revised June 12, 1991. Accepted for publication June 12, 1991. Reprint requests: William W. Busse, MD, University of Wisconsin Hospital, H6/360 CSC, 600 Highland Ave., Madison, WI 53192. l/1/31692

neutrophils,

Abbreviations

intracellular

calcium

used

PAF: EOS: O,- : SOD: CB: PMA: FMLP: [Ca+‘Ii: AR: EGTA:

Platelet-activatingfactor Eosinophil Superoxideanion Superoxidedismutase CytochalasinB Phorbolmyristateacetate N-formyl-methionyl-leucyl-phenylalanine Intracellularcalciumconcentration Allergic rhinitis Ethyleneglycol-bis(P-aminoethylether) N,N,N’,N’,-tetraacetic acid HBSS: Hanks’balancedsaltsolution

duces bronchoconstriction7-9 and the development of airway hyperresponsiveness.gx ‘O Although the mechanisms for these pulmonary changes have yet to be determined, instillation of PAF into the airways of 749

750

Zoratti et al.

primates causes eosinophilic infiltration of bronchial mucosa.” These reported effects of PAF on lung function, and their association with the presence of EOSs, suggest a linkage between PAF and EOSs in the development of airway dysfunction in asthma. PAF is a receptor-specific, physiologic granulocyte activator. Evidence that PAF may have an important effector relationship with EOSs has already been explored by other investigations. For example, analysis of EOS membrane phospholipids reveals high levels of PAF precursor lipids,” and activated EOSs synthesize PAF. I3 Furthermore, PAF also enhances human EOS cytotoxicity,‘4 causes granular content release,‘5T I6 and stimulates O,- generation.16 Moreover, PAF is the most potent in vitro chemoattractant for EOSs thus far described” and causes in vivo EOS accumulation in the skin’* and lung.” Finally, PAF promotes EOS adherence to human vascular endothelium.*‘z *I To evaluate further whether EOSs are more responsive to PAF than other granulocytes, we compared O,- generation to this agonist in isolated human EOSs and neutrophils. Furthermore, since increases in intracellular Ca’ + may be an important messenger for cell activation with PAF, this activity was also determined.

J. ALLERGY

CLIN. IMMUNOL. NOVEMBER 1991

subjectswithout asthma(threemaleand threefemalesubjects, rangingin agefrom 26 to 46 years)wereincludedin the study. Medication for the treatmentof asthmaor AR waswithheldfor at least12hoursbeforebloodwasobtained; noneof the patientsreceivedoral corticosteroidtherapyfor at least4 weeksbeforethe time of study. Isolation of EOS and neutrophil suspensions

Ethylenediaminetetraacetic acid anticoagulatedblood (0.5 ml of 2% ethylenediaminetetraacetic acid per 9.5 ml of blood)wassedimented in 4.5% dextranin T&-buffered salinesolution,pH 7.4 (4 vol of bloodper 1 vol of dextran), for 45 minutesat 22” C. The plasma-leukocyte suspension wasremoved,layeredover Ficoll-Hypaque,andthen centrifuged (400 g for 15 minutesat 22” C) as describedby B6yum.22After centrifugation,the mononuclearcell-rich bandwasremoved,andthe granulocyte-redbloodcell pelletswereresuspended in Ca’ +-freeHBSS.Thegranulocytes werepelleted,washedtwice, andresuspended in HBSS(1 mmol/L of Ca++) with 5% normalcalf serum.Two milliliters of the granulocytesuspension (10 to 20 X lo6cells per milliliter) werethencarefully layeredonto a discontinuousdensityPercollgradient*”andcentrifugedfor 20 minutesat 700g at 22” C. The centrifugedcellsseparated into five distinctdensitybandswithin thePercollgradient;bands containingpurified (>95%) neutrophils(densityof 1.085 to 1.090gm/ml) and EOSs (>90%, density of 1.095 to 1.100gmlml) werecarefully collected,counted,and differentiatedafter the applicationof Wright stain. Residual METHODS red blood cells were lysed by hypotonic shock, and the Reagents purifiedgranulocytefractionswerewashedandresuspended Percollwaspurchasedfrom PharmaciaFine Chemicals, in HBSS with 0.1% gelatin(HBSSlgel). Piscataway,N.J. HBSS (containing1 mmol/L of Cat+) wasobtainedfrom GIBCO, GrandIsland,N.Y. Horse-heart 02- generation ferricytochromeC (type VI), EGTA, Tris, Triton, PMA, The functionof purifiedneutrophilsandEOSsfrom each FMLP, Cl8 PAF (L-a-phosphatidylcholine-B-acetyl-y-o- blood donor was assessed simultaneously.Generationof octadec-9-cis-enyl), andSOD werepurchased from Sigma 02- wasmeasured asthe SOD-inhibitablereductionof ferChemicalCo., St. Louis, MO. Cl6 PAF (1-o-hexadecyl-Z ricytochromeC as previouslydescribed.*”Briefly, with a acetyl-sn-glycero-3-phosphorylcholine) andlyso PAF (l-o96-wellmicrotiterplate(ImmulonII, Dynatech,Alexandria, hexadecyl-sn-glycero-3-phosphorylcholine) werepurchased Va.) and a 200 l.r,lreactionvolume, 1 X 105cells were from BacchemCorp., Monterey, Calif. Indo-l/AM was addedto 100 pmol/L of cytochromec in HBSS/gel. To purchased from MolecularProbesCorp., Eugene,Ore. The initiate the reaction,the cellswere incubatedwith FMLP PAF antagonist,WEB 2086, wasa gift of BoehringerInor PAF in the presence of 5 pg/ml of CB (or PMA). Immegelheim(Rigefield, Conn.). Stock solutionsof PMA (5 diately after the additionof activator, the absorbance of the mg/ml), FMLP (lo-* mol/L) andCB (10 mg/ml; Aldrich cell suspensions in the individualreactionwellswasmeaChemicalCo., Milwaukee, Wis.) were madein dimethyl suredat 550 nm in a microplateautoreader(EL 309, Biosulfoxideand storedin aliquotsat -80” C. Immediately Tek Instruments,Burlington, Vt.), followed by repetitive beforeuse, aliquotswere thawedand adjustedto the ap- readingsduringthe next 60 minutes.Betweenreadings,the propriateconcentrations with HBSS. The concentrationof plateswereplacedin a 5% CO, incubatorat 37” C. Each dimethylsulfoxidepresentwasalways95% after 60 minutesof incubationwith PAF or theother documented asthma(two femaleand six malepatients,ranging in agefrom 21 to 44 years)andsix nonallergicnormal activators.

VOLUME NUMBER

PAF stimulates

88 5

+

-Q-

NEUTROWILS

= P-c.05,

Y I

Ol 0

3E-9

-

superoxide

generation

751

EOSINOPHILS

n=8)

I

I

1 E-8

3E-8 PAF

I 3E-7

CONCENTRATION

Measurement of intracellular calcium concentrations

I

I

1 E-7

FIG. 1. The dose-response relationship of O,- generation after EOSs and neutrophils. Values are expressed as mean 2 SEM.

I

lE-6

3E-6

I

I

1 E--5

(MOLAR) activation

with

PAF in isolated

asdescribedby Grynkiewicz,26wherekd = 250 nmol/L,

Purifiedpopulationsof EOSsandneutrophils,obtained by methodspreviouslydescribed,wereloadedwith the fluorescentcalciumindicator,Indo-l /AM (37” C for 30 minutes)at a cell concentrationof 5 X lo6cellsper milliliter.z6 After the loadingprocedure,the cellswere washedtwice and resuspended in HBSS at a concentrationof 6 X 106 cells per milliliter. Before the measurements of fluorescence,the cells were transferredto a cuvette and diluted with HBSS to concentrationsrangingfrom 2 to 5 x lo6 cellsper milliliter. The final cell concentrationwasdependent on the total numberof EOSsavailable.Varying cell concentrationsover this rangedid not affect the interpretation of fluorescence (datanot presented).The contentsof the cuvettewerewarmedto 37” C andmagneticallystirred. Changesin fluorescencewere measuredwith an SLM 8000C spectrofluorometer (SLM, Inc., Urbana,Ill.) with an excitation wavelengthof 355nm. The ratio of fluorescenceat 405 and 485 nm was measuredsimultaneously. Maximal andminimalfluorescence valuesweredetermined aftercell lysisin thepresence of excessextracellularcalcium (HBSS)or absence of extracellularcalcium(EGTA-chelated HBSS), respectively.Measurements of fluorescence were determinedafter stimulationwith PAF at concentrations of 1 X lo-” to 1 X 10e6mol/L. Theseexperimentswere alsoperformedin the presenceandabsence of extracellular calcium. [Ca+‘Ii wascalculatedfrom the following formula: [Ca++], = kd (s)

eosinophil

(2)

R = fluorescencemeasureratio of $$

nm, R,,. =

fluorescenceratio in the absenceof Ca”, R, = fluorescence ratio in the presenceof excessCa++, Sf, = fluorescenceof the Indo-l at 485 nm in the absenceof calcium,and Sb, = fluorescence of the Indo-l at 485 nm in the presenceof calciumexcess.

Statistical

analysis

The resultswereanalyzedby a two-tailedStudent’st test for comparisonof neutrophilsand EOSs. The analysisof cellularresponses betweenpatient groupswas performed with a Mann-WhitneyU test.

RESULTS Superoxide

generation

Concentration-dependentresponseto PAF. Purified human neutrophils and EOSs from the same subject were incubated (37” C) with increasing concentrations of PAF, 3 x 10m9to 1 x 10m5mol/L, and O,- generation was measured during a 60-minute incubation (Fig. 1). Peak neutrophil generation of Oz- occurred at 1 X 10m6 mol/L of PAF; however, the small amount of O,- generated by neutrophils with PAF made calculation of a median effective dose difficult. The maximal EOS response was observed with 3 x IO-’ mol/L of PAF with a median effective dose value of 1.23 (k 1.55) x lo-’ mollL (mean + SEM). Although EOSs generated more O,-

752

Zoratti et al.

J. ALLERGY

CLIN. IMMUNOL. NOVEMBER 1991

125r

FIG. 2. The effect of WEB 2086 (mol/L) on PAF (1 x 1O-6 mol/L) stimulation eration. Values are mean + SEM; n = 4.

than neutrophils, the differences were significant only at PAF concentrations > 1 X lo-’ mol/L because of wide individual variations in O,- generation at lower PAF concentrations. Lyso-PAF (1 X 10e9 to 1 X 10s6 mol/L), a biologic precursor and metabolite of PAF, did not stimulate more O,- generation than spontaneous release levels from either neutrophils or EOSs (data not presented). The potency of Cl6 and Cl8 PAF (1 X low6 mol/L for 60 minutes) was compared and did not reveal significant differences between these two compounds in either EOS or neutrophil O,- generation. (EOS: 6.61 + 1.72 [Cl61 versus 5.44 + 1.08 [Cl81 mmol/L of cytochrome c reduced per 5 X lo5 cells; n = 5; p > 0.05; neutrophils: 0.47 + 0.52 [Cl61 versus 1.5 + 0.4 [Cl81 nmol/L of cytochrome c reduced per 5 x lo5 cells; n = 5;p > 0.05). In these experiments, EOSs still generated more 02- with both Cl6 and Cl8 PAF than did neutrophils. The effect of the PAF antagonist, WEB 2086,“~ *’ was evaluated on PAF (1 x 10m6 mol/L) generation of 02- (Fig. 2). EOSs were incubated with WEB 2086 for 15 minutes before the addition of PAF (1 X 10m6 mol/ L) . There was concentration-dependent WEB 2086 inhibition of PAF stimulation of O,- generation with complete suppression at equimolar levels of agonist and antagonist. Comparisonof EOS and neutrophil responses.The amount of O,- generated by EOSs and neutrophils from all 20 subjects (combined normal and atopic

of EOS O,- gen-

subjects and subjects with asthma) was measured and compared after stimulation with 1 X 10m6 mol/L of PAF. This concentration of PAF was selected because it was associated with a maximal response in neutrophils and would therefore not prejudice the comparison in favor of EOSs. When granulocytes from the same subject were compared, EOSs generated approximately three times more O,- than did neutrophils: 9.77 + 1.49 versus 2.92 k 0.43 nmol per 5 X 10’ cells, respectively; p < 0.0001 (Fig. 3). EOSs achieved peak generation of O,- with PAF 30 minutes after activation. In contrast, peak neutrophil generation of O,- with PAF was noted within 10 minutes of activation. Both cell suspensions maintained viability (>95%) during incubation with 1 x lOA mol/ L of PAF. We also compared generation of O,- by neutrophils and EOSs with two other activators, PMA and FMLP (Fig. 4). EOSs generated significantly more 02- in response to PMA (1 rig/ml) than did neutrophils (41.52 k 1.14 nmol of O,- per 5 X lo5 cells versus 29.35 + 2.23; p < 0.0001). In contrast, O,- generation to FMLP (1 X lo-’ mol/L) was similar in EOSs and neutrophils, 15.23 + 7.98 nmol of O,per 5 X 10’ cells versus 11.90 +- 4.09; p = 0.12. Of the three activators, PAF stimulated less O,- generation from EOSs and neutrophils than either PMA or FMLP. However, when the relative potency of these activators on the two granulocyte populations was compared, the O,- response to PAF was proportion-

PAF stimulates

l2 0 -

0

NEUTROPHILS

(n=20)

-

l

EOSINOPHILS

(n-20)

IO- l

*

T/I

8-

0

5

10

15

20

25

+

eosinophil

superoxide

+

+

35

40

753

+ I)

Iw--- 0I I I

T T eye I

30

generation

45

50

55

60

TIME (MINUTES) FIG. 3. Comparison (1 x 1O-6 mol/L);

of EOS and neutrophil 02*p < 0.001; tp < 0.0001.

ally greater in EOSs than neutrophils (see inset, Fig. 4). The ratio (EOS / neutrophil) of O2- generation with PAF was 3.4; in contrast, the ratio of response to PMA and FMLP was 1.4 and 1.3, respectively. Comparisonof granulocyte responseto PAF in subjects with AR and asthma and in normal subjects.

PAF-stimulated 02- generation by EOSs and neutrophils from the different subject groups (normal, N = 6; atopic, N = 6; asthma, N = 8) was compared (Fig. 5). In each of the three study groups, PAF stimulated more 02- production from EOSs than from neutrophils. Although there was a trend of increased 02- generation by PAF-stimulated EOSs from patients with asthma, the difference did not achieve signifiCalKX.

Intracellular calcium responseto PAF. Intracellular mobilization of calcium has been identified as a messenger for many membrane-receptor activated responses. Therefore, we speculated that differences in neutrophil and EOS generation of O,- may be reflected in calcium mobilization after cell activation. Consequently, the response to PAF was determined with neutrophils and EOSs with the calcium-binding fluorescent indicator, I&o-l I AM. The baseline, or resting, [Ca* ‘]i values were similar for neutrophils and EOSs (126 -+ 17 versus 156 k 24 nmol/L; N = 6, respectively). In an initial series of experiments, the intracellular [Ca+ +Jj response to PAF

production

during

a l-hour

incubation

with

PAF

(1 x lo-’ to 1 X 1O-6 mol/L) was determined in neutrophils and EOSs (Table I). Although similar maximal responses to PAF were noted between neutrophik (1757 1 237 nmol/L, N = 5) and EOSs (1.543 -+ 43 nmol! L; N = 3), it was apparent that the initial concentration of PAF tested (I X lo-’ mol/L) had already caused a peak increase in neutrophi1 [Ca’ ‘Ii. To determine the threshold response to PAF, as measured by an increase in [Ca’ +]i, a subsequent series of experiments was performed with lower concentrations of activator. In the neutrophil, no increase in [Ca+ +]; was noted to 1 x 10-l’ mol/L of PAF, but [Ca’+], rose (42% t 48% over baseline; N = 5) with I X lo-” mol/L of PAF. In contrast, eosinophil [Ca’ ‘Ii values did not change with I x IO-” moI/L of PAF, however, 1 x 10-l’ m&L of PAF caused an increase of 55% +- 23% over baseline values; N = 3. These data suggest a tenfold difference between the neutrophil and EOS sensitivity of the [Ca* ‘1, response to PAF. The peak [Ca++& responses to PAF were similar in both cell types but were achieved in neutrophils at 1 X lo-’ mol/L, whereas a concentration of 1 X lO-‘j mol/L was required to reach the same plateau in EOSs. When extracellular Ca+ + was chelated by EGTA, there was reduction (>70%), but not complete suppression, of the Indo- 1-dependent fluorescence response to PAF (Fig. 6). Also, when eosinophil 02-

Zoratti

J. ALLERGY

et al.

=

EOSINOPHILS

&fd NEUTROPHILS

FMLP (N n16)

41.6 +/- 1.2

29.4 +/- 2.2

16.2 +/- 2.0

11.0 */-to

9.8

l /-1.6

P’MA

0 -

l

12-

+/-

2.9 0.4

FklLP ’

FIG. 4. A comparison of neutrophil and EOS total 1 nglnl of PMA, 0.1 kmol/L of FMLP, or 1 pmol/L

15T

CLIN. IMMUNOL. NOVEMBER 1991

0

l

O,- production during 1 hour incubation of PAF. (Values are mean k SEM.)

NE UTROPHILS/NORhlAL EOSINOPHILS/NORMAL

-A

NEUTROPHILS/ALLERCIC

-A

EOSINOPHILS/ALLERGIC

0

NEUTROPHILS/ASTHMA

l -

l

EOSINOPHILS/ASTHMA

-m

D--D

n’

9-

0 -

with

m

6-

0

5

10

15

20

25

30

35

40

45

50

55

60

TIME (MINUTES) FIG. 5. Comparison of 02- generation to 1 x 10m6mol/L from allergic subjects (N = 6). subjects with asthma

generation to PAF was determined in the presence of EGTA, the response was diminished but not completely inhibited (Table II). There was also a difference in the fluorescent pattern to PAF when neutrophils and EOSs were compared (Fig. 6). Both neu-

of PAF by neutrophils (N = 8), and normal

and EOSs obtained subjects (N = 6).

trophils and EOSs elicited a distinct immediate fluorescent spike with the addition of PAF; in neutrophils, however, the increase in fluorescence to PAF was sustained longer. These observations suggest a discordance in the respiratory burst and

VOLUME NUMBER

PAF stimulates

88 5

IN THE PRESENCE

m Iz

OF EXTRACELLULAR

eosinophil

superoxide

generation

755

CALCIUM

-

5 WO

I

t-PAF I

I

I

50

100

150

200

ii

8

IN THE ABSENCE

E

OF EXTRACELLULAR

CALCIUM

Neutrophils

Eosinophils

8 3

i;;l

EGTA

0

I

I

50

100

I

150

200

I

I

J

0

100

I

J

150

200

TIME (SECtIN% FIG. 6. A representative

tracing to illustrate the change in neutrophil and EOS fluorescence with Indo-l-loaded cells after activation with PAF (1 x lo-’ mol/L). The changes in [Ca++], in the presence (top) and absence (bottom) of extracellular calcium are calculated with the formula described in the METHODS section.

[Ca’ +li response to PAF between the EOS and neutrophil . DISCUSSION

If PAF contributes to the pathophysiology and pathogenesis of asthma, 7-1othere appears to be an important link to EOS activation.‘4-2’ Our observations are additional evidence and support for the hypothesis that the ability of PAF to generate leukocyte inflammation is more apparent with the EOS than the neutrophil, a phenomenon shared with other activators (Fig. 4). From our experiments, a number of observations were made on the ability of PAF to stimulate O,generation by EOSs and neutrophils. First, stimulation of 02- generation by PAF with both EOSs and neutrophils is similar in normal subjects and in patients with either AR or asthma. Second, compared to other activators, PMA and FMLP, the absolute amount of O,- generated after PAF activation is less. However, with PMA and FMLP, the ratio of EOS to neutrophil 02- response is 1.4 and 1.3, respectively; PAF, in contrast, generates “relatively” more O,- in the EOS than neutrophil by a factor of 3.4. Our data do not establish the mechanism by which EOSs produce more O,- after PAF activation than

TABLE

I. The

human

neutrophils

with

[Ca++l,

response and

EOSs

to PAF

in

loaded

Indo-l

[Ca++], after PAF activation Neutrophil [Ca’ +li*

Conditions Baseline 1

x

lo-* mol/L

PAF

1

x

lo-’

1

x

10m6 mol/L PAF

mol/L PAF

126 (N 1453 (N 1757 (A’ 1624 (N

5 = ? = += k =

17 6) 204 3)t 237 5)t 249 3)-t

EOS [Ca++l, 156 (N 505 (N 889 (N 1543

~fr 24 = 6) c 71 = 311 -c 118 = 4)‘f +- 43 (N = 3Yt

Values are mean -t- SEM. *Expressed as nanomolars per liter. tp < 0.05 compared to baseline values.

neutrophils produce. There are, however, a number of potential explanations. Both high- and low-affinity PAF receptors have been described.29-3’ Thus, although both neutrophils and EOSs have PAF receptors, the subtypes of these receptors, or their relative distribution on EOSs and neutrophils, may explain the observed differences in activity. However, there are

756

Zoratti et al.

J. ALLERGY

TABLE II. PAF (1 x 1O-6 mol/L) stimulated EOS O,- generation in the presence and absence of extracellular calcium (mean +- SEM; N = 3) Conditions

1 mmol/L of Ca++ 1 mmol/L of Ca++ plus 2 mmol/L of EGTA

nmol/L Cytochrome c reduced/5 x lo5 cells

21.7 f 4.26 3.49 f 1.13

multiple transduction pathways that link membrane receptor activation and intracellular regulation of stimulus-response coupling to PAP’, 31;consequently, the EOS and neutrophil may have important dissimilarities in this regard. PAF also causes an increase in [Ca’ +]i in both EOSs and neutrophils. The initial rise and peak in [Ca’ ‘Ii response occurred at lower PAF concentrations in the neutrophil, but maximal values were equivalent to the EOS. Although the increase in [Ca++li was concentration dependent, it did not necessarily correlate with the generation of O,- by these cells. Moreover, EGTA chelation of extracellular Ca’ + inhibited but did not totally suppress either the PAF-stimulated rise in [Ca’ ‘Ii or O,- generation. The apparent discordance between the rise in [Ca’ ‘Ii and O,- generation agrees with an earlier study that a dissociation exists between calcium influx and 02- production by guinea pig EOSs.*’ Furthermore, this finding suggests that, although increased [Ca’ ‘11 alone may not be sufficient to stimulate O,- generation (as observed with neutrophils), it does appear to be necessary to initiate or to obtain optimal activation of the granulocyte respiratory burst by PAF (Table II). This discrepancy underscores the necessity to investigate further other calcium-independent biochemical transduction events that are responsible for receptor-mediated cellular activation.30, 32,33The generation of cell membrane-derived phosphoinositide metabolites is a prime candidate for this alternative second messenger.34 Circulating concentrations of PAF are estimated to be 1 x 10-l’ to 1 x 1O-‘2 mol/L3’ and, in this range, promote adhesion to vascular endothelium.*’ Consequently, the concentrations of PAF used in our experiments are unlikely to be present in circulation. However, as leukocytes accumulate at inflammatory sites (i.e., the airways) and undergo activation, it is possible that concentrations of PAF will be greater than concentrations present in circulation, since numerous airway cells generate PAF. IT3The concentration of PAF used in our experiments did not reduce either neutrophil or EOS viability and was similar to

CLIN. IMMUNOL. NOVEMBER 1991

those evaluated in other in vitro analyses.16 Moreover, the response to PAF was blocked by the antagonist, WEB 2086, indicating an effect on the cell via its receptor and not a nonspecific stimulation of O,generation. Thus, although the PAF concentrations required to achieve maximal stimulation of O,- were large, they may still be of physiologic relevance. Generation of O,- by neutrophils and EOSs after stimulation by PAF was enhanced by the addition of CB. Whether there is an in vivo correlate to in vitro activity of CB is not established. However, the properties of CB allowed us to simultaneously compare cellular activation to PAF with the limited number of EOSs available from patients without hypereosinophilia, as was the case of our study subjects, particularly the normal control subjects. Furthermore, since the actions of CB were similar on EOS and neutrophil generation of 02- to PAF (data not presented), we believe that these in vitro comparisons are valid. Nonetheless, the conditions of incubation and activation must be considered in making in vivo projections . The differential response of EOSs and neutrophils to PAF was similar in normal subjects and patients with either AR or asthma. For a number of reasons, these observations do not diminish the potential importance of PAF to EOS activity in asthma. EOSs are a heterogenous population of cells when they are evaluated by density and activity.36”8 Since patients with asthma and AR have an increased percentage of circulating, low-density EOSs, which may have increased activity,23, 3gadditional analyses of these subpopulations will be necessary to determine fully the effect of PAF activation of EOSs in allergic diseases. However, blood EOS subpopulations may not be the best cell source for these comparisons. In a previous study, we found only subtle differences in peripheral blood EOS generation of 02- between normal and low-density cells from patients with asthma.4o More importantly, preliminary experiments have found an interesting inverse relationship between generation of 02- to PAF by airway EOSs and cell density. Lowdensity cells generate less 02- .4’ Our preliminary observations raise the following possible explanation for the absence of a difference between the EOS response to PAF between normal subjects and subjects with asthma. PAF rapidly causes receptor desensitization. 16*42 Consequently, if increased endogenous PAF concentrations are present in asthma, both in the airway and circulation, it would be anticipated that the exposed cells would have a “down regulated” response to PAF. Based on this speculation, the similar response to PAF with EOS in control subjects and patients with AR and asthma may actually represent a desensitized

VOLUME NUMBER

PAF stimulates

88 5

response in the latter population. This possibility will, obviously, require additional study, and its focus will be on airway rather than circulating EOSs. The role of PAF in the pathogenesis of asthma remains undefined. However, our data provide additional information as to the interdependency of PAF and EOSs. Continued studies into PAF regulation of EOS function promise to elicit not only insight into contribution of PAF to asthma via the EOS but also an increased understanding of the biology of this cell.

16.

17.

18.

19.

REFERENCES 1. Amoux B, David D, Benveniste J. Release of plateletactivating factor (Paf-acether) from alveolar macrophages by the calcium ionophore A23187 and phagocytosis. Eur J Clin Invest 1980;10:437-41. 2. Sanchez-Crespo M, Alonso F, Figaro J. Platelet-activating factor in anaphylaxis and phagocytosis. I. Release from human peripheral polymorphonuclear and monocytes during the stimulation by ionophore A23187 and phagocytosis but not from degranulating basophils. Immunology 1980;40:645-55. 3. Jouvin-Marche E, Grzych JM, Boullet C, Capron M, Benveniste J. Formation of Paf-acether by human eosinophils [Abstract]. Fed Proc 1984;43:1924. 4. Camussi G, Aglietta M, Malavasi F, et al. The release of platelet-activating factor from human endothelial cells in culture. J Immunol 1983;131:2397-403. 5. Camussi G, Pawlowski I, Tetta C, et al. Acute lung inflammation induced in the rabbit by local instillation of l-o-octadecyl-2-acetylsn-glyceryl-3 phosphorylcholine or of native PAF. Am J Path01 1983;112:78-88. 6. Lavis JC, O’Flahetty JT, McCall CE, Wykle RL, Bond MG. Platelet-activating factor effects on pulmonary ultrastructure in rabbit. Exp Mol Pathol 1983;38: 100-8. 7. Rubin AE, Smith LJ, Patterson R. The bronchoconstrictor properties of platelet-activating factor in man. Am Rev Respir Dis 1987;163:1145-51. 8. Gateu 0, Amoux B, Deriaz H, Viors P, Benveniste J. Acute effects of intratracheal administration of PAF-acether in humans [Abstract]. Am Rev Respir Dis 1984;133:129. 9. Cuss FM, Dixon CMS, Barnes PJ. Effect of inhaled plateletactivating factor on pulmonary function and bronchoresponsiveness in man. Lancet 1986;2:189-92. 10. Chung KF, Dixon CMS, Barnes PJ. Platelet-activating factor and asthmatic airways: effects on caliber, responsiveness, and circulating cells [Abstract]. Am Rev Respir Dis 1987;135: 159. 11. Amoux B, Dergean A, Page CP, Morley J, Benveniste J. Pulmonary effects of platelet-activating factor in a primate are inhibited by ketotifin [Abstract]. Am Rev Respir Dis 1985;131:2. 12. Ojima-Uchiyama A, Masuzawa Y, Sugiura T, et al. Phospholipid analysis of human eosinophils: high levels of alkylacylglycerophosphocholine lipids. Lipids 1988;23:815-7. 13. Lee TC, Lenihan DJ, Malone B, Roddy LL, Wasserman SI. Increased biosynthesis of PAF in activated human eosinophils. J Biol Chem 1984;259:5526-30. 14. MacDonald AJ, Moqbel R, Wardlaw AJ, Kay AB. Plateletactivating factor (PAF-acether) enhances eosinophil cytotoxicity in vitro [Abstract]. J ALLERGY CLIN IMMUNOL 1986;77:227. 15. Kroegel C, YukawaT. Dent G, Chung KF, Barnes PJ. Platelet-

20.

21.

22.

23.

24.

25. 26.

27.

28.

29.

30.

31.

32.

33.

34.

eosinophil

superoxide

generation

757

activating factor induces eosinophil peroxidase release from human eosinophils. Immunology 1988;64:559-61. Kroegel C, Yukawa T, Dent G, Venge P, Chung KF, Barnes PJ. Stimulation of degranulation from human eosinophils by platelet-activity factor. J Immunol 1989;142:3518-26. Wardlaw AJ, Mogbel R, Cromwell 0, Kay AB. Platelet-activity factor-a potent chemotactic and chemokinetic factor for human eosinophils. J Clin Invest 1986;78:1701-6. Archer CB, Page CP, Paul W, Morley J, MacDonald DM. Inflammatory characteristics of platelet-activating factor (PAFacether) in human skin. Br J Dermatol 1984;110:45-50. Takizawa H, Ishii A, Suzuki S, Shiga J, Miyamoto T. Bronchoconstriction induced by platelet-activating factor in the guinea pig and its inhibition by CV3988, a PAF antagonist: serial changes in findings of lung histology and bronchoalveolar lavage cell population. Int Arch Allergy Appl Immunol 1988;86:375-82. Kimani G, Tonnesen MG, Henson PM. Stimulation of eosinophil adherence to human vascular endothelial cells in vitro by platelet-activating factor. J Immunol 1988;140:3161-6. Dillon PK, Fitzpatrick MF, Rittar AB, Duran WN. Effect of PAF on leukocyte adhesion to microvascular endothelium. Inflammation 1988;12:563-73. Boyurn A. Isolation of mononuclear cells and granulocytes from human blood. Stand J Clin Lab Invest 1968;21(suppl):7783. Frick WE, Sedgwick JB, Busse WW. The appearance of hypodense eosinophils in antigen-dependent late-phase asthma. Am Rev Respir Dis 1989;139:1401-6. Sedgwick JB, Vrtis RF, Gourley MF, Busse WW. Stimulusdependent differences in superoxide anion generation by normal human eosinophils and neutrophils. J .~LLERGY CLIN IMMUNOL 1988;81:876-83. Van Gelder BF, Slater EC. The extinction coefficient of cytochrome c. Biochim Biophys Acta 1962;58:593-8. Grynkiewicz G, Poenie M, Tsien RY. A new generation of calcium indicators with greatly improved fluorescent properties. J Biol Chem 1985;260:3440-50. Casals-Stenzel J, Muacevic G, Weber K-H. Pharmacological actions of WEB 2086, a new specific antagonist of plate-activating factor. J Pharmacol Exp Ther 1987;241:974-8 1. Casals-Stenzel J. Effects of WEB 2086, a novel antagonist of platelet-activating factor in active and passive anaphylaxis. Immunopharmacology 1987;3:7-24. Kroegel C, Yukawa T, Westwick J, Barnes PJ. Evidence for two platelet-activating factor receptors on eosinophils: dissociation between PAF-induced intracellular Ca’ + mobilization, degranulation, and superoxide production in eosinophils. Biothem Biophys Res Commun 1989;162:511-21. Hwang SB. Identification of a second putative receptor of platelet-activating factor from human PMNs. J Biol Chem 1988;263:3225-33. Shaw JO, Pinckard N, Ferrigni KS, McManus LM, Hanahan DJ. Activation of human neutrophils with platelet-activating factor. J Immunol 1981;127:1250-5. Koenderman L, Yazdanbaksh M, Roos D, Verhoeven AJ. Dual mechanisms in priming of the chemoattractant-induced respiratory burst in human granulocytes. J Immunol 1989;142: 623-8. Apeldorf WJ, Melnick DA, Meshulam T, Rasmussen H, Malech HL. A transient rise in intracellular calcium is not a sufficient stimulus for respiratory burst activation in human PMNs. Biochem Biophys Res Commun 1985;132:674-80. Berridge MJ. Inositol triphosphate and diacylglycerol as second messengers. Biochem J 1984;220:34560.

Zoratti

et al.

J. ALLERGY

35. Yamada K, Osamu A, Yoshimura T, Katayama K. Highly sensitive gas chromatographic-mass spectrometric method for the determinationof platelet-activating factor in human blood. J Chromatogr 1988;433:243-7. 36. Prin L, Capron M, Tonne1 AB, Bletry 0, Capron A. Heterogeneity of human peripheral blood eosinophils: variability in cell density and cytotoxic ability in relation to the level and the origin of hypereosinophilia. Int Arch Allergy Appl Immunol 1983;72:336-46. 37. Winqvist I, Olofsson I, Persson A-M, Hallberg T. Altered density, metabolism, and surface receptors in eosinophilia. Immunology 1982;47:531-9. 38. Kajita T, Yui Y, Mita H, et al. Release of leukotriene C, from

39. 40. 41. 42.

CLIN. IMMUNOL. NOVEMBER 1991

human eosinophils and its relation to cell density. Int Arch Allergy Appl Immunol 1985;78:406-10. Frick WE, Sedgwick JB, Busse WW. Hypodense eosinophils in allergic rhinitis. J ALLERGY CLIN IMMUNOL 1988;82: 119-25. Sedgwick JB, Geiger KM, Busse WW. Superoxide generation by hypodense eosinophils from patients with asthma. Am Rev Respir Dis 1990;142:120-5. Sedgwick JB, Calhoun WJ, Busse WW. Enhanced superoxide (Or-) generation by airway eosinophils (EOS) [Abstract]. J ALLERGY CLM IMMUNOL 1991;87:240. Henson PM. Activation and desensitization of platelets by PAF derived from IgE-sensitized basophils. J Exp Med 1976;143:937-44.

Precision of skin prick and puncture nine methods

tests with

Pascal Demoly, MD,* Jean Bousquet, MD,* Jean-Claude Manderscheid, MD,** Sten Dreborg, MD,*** Henriette Dhivert, MD,* and Franqois-Bernard Michel, MD* Montpellier,

France,

and Linkoping,

Sweden

New devices for puncture tests have been proposed recently, but their precision by comparison to the prick test method is poorly known. Seven puncture tests (Allerprick, Morrow Brown standardized needle, Phazer, Pricker, Stallerpointe, Stallerkit, and Wyeth btficated needle) were compared with the modified prick test pegormed with hypodermic or intradermal needles in eight carefully selected normal volunteers. Skin tests with histamine hydrochloride (10 mglml) were only pe$ormed when there was no factor that might interfere with their interpretation. The site of skin tests on the forearm was demonstrated not to significantly influence the reaction size. The coeficient of variation of the tests ranged from 8.4% to 21.7%. Modt$ed skin prick tests are satisfactory since they are highly reproducible (coeficient of variation: 13.4% and 16.5%) and there is no subject effect. Phazet was found to be more reproducible without subject effect. Pricker is satisfactory since it has no subject effect and a reproducibility similar to that of modified prick tests. Other tests are less reproducible (Stallerkit or Morrow Brown) or vary between subjects (Allerkit, Stallerkit, Stallerpointe, and Wyeth Needle). (J ALLERGY CLIN IMMUNOL 1991;88:758-62 .) Key words: Histamine,

skin test, prick test, reproducibility,

From the *Clinique des Maladies Respiratoires, Hopital l’Aiguelongue, Montpellier, and **lXpartement d ‘Informatique M&licale, Montpellier, France, and ***Department of Pediatrics, University Hospital, Linkoping, Sweden. Received for publication Nov. 27, 1990. Revised June 13, 1991. Accepted for publication June 13, 1991. Reprint requests: Jean Bousquet, MD, Clinique des Maladies Respiratoires, Hop&al PAiguelongue, 34059, Montpellier Cedex, France. 111131693 758

human

SPTs are among the skin test methods from which to choose and, at least in Europe, they tend to be preferred to intradermal test.‘, ’ The modified SPT introduced by Pepys3 is the current reference method, although the variability of this test has been demonstrated to be greater than that of the intradermal reaction.4, 5 Investigators have therefore attempted during the past 10 years to decrease the variability of SPT by introducing puncture tests with various devices. The most popular instruments are the MB stan-