vasoconstrictor effects of platelet-activating factor in the hamster

722

Vasoconstrictor Effects of Platelet-ActivatingFactor in the Hamster Cheek Pouch

Microcirculation: Dose-Related Relations andPathways of Action

Patricia K. Dillon, Arthur B. Ritter, and Walter N. Duran

Platelet-activating factor (PAF) has been implicated as a potential mediator of inflammatory processes.In this study, we quantified the effects of PAF on vessel diameter in a microvascular bed and investigatedthe biochemical pathways of this compound. The hamster cheek pouch microcirculation was observedwith intravital microscopy. Experiments were video-recorded and analyzed with an image shearingdevice. Vasoconstriction was the predominant vasomotor response to PAF. PAF (10~'°-10~* M) wasapplied topically to the pouch for 3 minutes. Arterioles ranging in size from 8 to 15 (xm were the mostsensitive, and they constricted completely in response to PAF 10 7 and 10"' M. Arterioles 21—40 |xmin diameter constricted to 12-17% of control after PAF at 10~7 and 10"3 M, respectively; they reopenedto about 70% of their control value after a few minutes and remained near that size throughout theexperiment. Arterioles 41-60 |xm in diameter constricted to about 20% control size in response to 10"'and 10~5 M PAF, and by the end of the experiment, these vessels had returned to about 90% controlsize. To determine the pathways of PAF actions, inhibitors of the arachidonic acid cascade and receptorblockers were used. Dexamethasone, indomethacin, OKY-046 (a thromboxane A2 synthetaseinhibitor), and kadsurenone (a PAF-receptor blocker) blocked the vasoconstrictor response to PAF.Our experiments demonstrate that PAF-produced arteriolar constriction in a microvascular bed is 1)dose-related, 2) dependent upon vessel size, 3) largely due to thromboxane A2 activity, and 4) mediatedby PAF-receptor interactions. (Circulation Research 1988;62:722-731)

Platelet-activating factor (PAF, l-0-alkyl-2-0-acetyl-sn-3-phosphoryl-choline), a polar phos-pholipid derived from membrane phospholi-

pids, is a putative mediator of inflammatory processes.It is produced by a variety of inflammatory cell typesin both animals and humans,'"4 activates the cellsinvolved in inflammation,5"7 and has been implicated inspecific inflammatory processes such as immunoglo-bulin E-induced anaphylaxis and immune complexdisease.89

PAFelicitsa diversity of hemodynamic effects in vivo.When administered intravascularry, it induces pulmo-nary hypertension,10" systemic hypotension,1Otl2J3

detrimental cardiac alterations,"14 transient neutro-penia and thrombocytopenia,1012 and increased vas-cular permeability.10"12-1516 Plasma levels of thrombox-ane B2, platelet factor 4, and 6-keto PGF^ are increasedsubsequent to intravascular administration of PAF, andthe substance also causes platelets and polymorpho-nuclear leukocytes (PMNs) to sequester in the small

From the Department of Physiology, UMDNJ-New JerseyMedical School, Newark, New Jersey.

Supported by a United States Public Health Service grant fromthe National Institutes of Health, National Heart, Lung, and BloodInstitute grant HL 25032, and a grant-in-aid from the AmericanHeart Association, New Jersey Affiliate. P.K.D. was the recipientof a fellowship from the American Heart Association, New JerseyAffiliate. W.N.D. is an Established Investigator of the AmericanHeart Association.

Address for reprints: Walter N. Durin, PhD, Department ofPhysiology, MSB H-609, UMDNJ-New Jersey Medical School,185 South Orange Avenue, Newark, NJ 07103-2757.

Received April 14, 1987; accepted October 20, 1987.

blood vessels of various organs.1718 In contrast withreports of hypotension after intravascular administra-tion, intradermal injection of PAF induces localizedvasoconstriction in human4 and guinea pig skin."

The mechanisms or pathways through which PAFalters vascular caliber are not known. Vasoactivityafter intravascular administration could be mediatedcentrally, while the vasoconstriction induced by topicalapplication or intradermal administration of PAF ismost likely a local event. Since PAF is released bymany blood-borne elements, an understanding of thelocal effects of PAF on microvascular variables is ofimportance.

In the present study, we characterized the effects ofPAF on arteriolar diameter in the microvascular bed ofthe hamster cheek pouch. Possible pathways throughwhich PAF could be exerting its effects were thenexamined using pharmacological inhibitors of enzymesin the arachidonic acid cascade as well as a receptorblocker for PAF.

Materials and Methods

Anesthesia and SurgeryAnesthesia and surgical procedures were similar for

all experiments. Male golden Syrian hamsters, weigh-ing 80-110 g, were anesthetized with sodium pento-barbital (60 mg/kg i.p.). Tracheotomy was performedto ensure clear airway passages. The left jugular veinwas cannulated for the administration of supplementaldoses of anesthetic and other drugs when appropriate.The left carotid artery was cannulated for collection of

by guest on February 14, 2018http://circres.ahajournals.org/

Dow

nloaded from

http://circres.ahajournals.org/

Dillon et al Dose-Related Vasoconstrictor Effects of PAF 723

blood samples and monitoring of blood pressure.Animals were kept on a heating pad throughout theexperiment to maintain body temperature at 37° C.

The right hamster cheek pouch was prepared fordirect visual observation and intervention according tothe methods of Greenblatt et al20 and Click et al.21

Briefly, a two-piece Lucite chamber with a 1-mlreservoir capacity was attached to a single layer of thepouch, delineating a 2.3-cm2 area for intravital obser-vation. The chamber reservoir was filled with bicar-bonate buffer solution and tested for leakage. Themillimolar composition of the buffer was 131.9 NaCl,4.7 KC1, 2.0 CaCl2, 1.2 MgSO4, and 18.0 NaHCO3.The buffer was adjusted to pH 7.35 with a Cole Palmer5987 digiphase pH meter (Chicago, Illinois) andequilibrated with 95% N-5% CO2.

Characterization of the Dose-Response Effects of PAFon Arteriolar Diameter

After surgical preparation, the hamster was posi-tioned on a Lucite board and placed on the stage of aNikon Optiphot microscope. A 1-hour stabilizationperiod ensued, during which the pouch was continu-ously suffused with bicarbonate buffer solution at a rateof 1 ml/min. A borosilicate glass suffusion systemequipped with heat and gas exchangers maintainedsuffusate temperature at 35° C with the aid of a constanttemperature circulating bath (VWR Scientific, SanFrancisco, California)

Fifteen minutes before and immediately precedingapplication of PAF, the transilluminated cheek pouchwas scanned, and selected vessels were video-recordedfor determination of control diameters. The suffusatewas then discontinued, and PAF (10~9, 10"7, and 10"5

M) was topically applied to the pouch for 3 minutes.PAF (Sigma Chemical, St. Louis, Missouri) wasdissolved in dimethyl sulfoxide (DMSO, Sigma) to aconcentration of 10~2 M and subsequently diluted to theappropriate concentration with a mixture of 1.5%bovine serum albumin (BSA, Sigma) and bicarbonatebuffer solution. Each animal received only one dose ofPAF. After 3 minutes of PAF application, suffusion wasreestablished. The video recorder was kept on contin-uously during PAF application and for 10 minutes afterinitial application. Vessels were recorded again at 20,30, 40, 50, and 60 minutes after initial application ofPAF. Both untreated pouches and pouches that receivedthe DMSO/BSA/bicarbonate buffer vehicle served ascontrols. Vessel diameters were measured from a replayof the videotape with a video image shearing monitormodel 907 (Instrument for Physiology and Medicine,San Diego, California). The instrument was calibratedwith a slide micrometer.

AnalysisVessels were categorized into one of three size

groups: 8-15 n.m, 21-40 \i.m, and 41-60 ^.m. Controldiameters were normalized to a value of one, andexperimental diameters were expressed as a ratio ofcontrol. The relative luminal diameters for each timepoint were averaged and plotted as relative luminal

diameter versus time. Regression equations for thevarious time segments were determined for the curveswithin each size group according to dose, and the slopesof the curves were compared for statistical differenceswith the IBM ANOVA statistical analysis package.

Determination of Blood Pressure and HematocritBlood samples were collected in duplicate from the

arterial cannula 5 minutes before PAF application andthen at 30-minute intervals. Hematocrits were deter-mined after centrifugation of blood samples. Arterialpressures were continuously recorded on a Beckmandynograph (Schiller Park, Illinois) with a Stathampressure transducer.

MicroscopyObservations were made with a Nikon Optiphot

microscope with 6.3 x , 10 x , 20 x , and 32 x long-working distance Leitz objectives with 10 x Nikonoculars. The microscope is equipped for both trans-illumination and epi-illumination studies. Bright-fieldtransillumination was provided by a 100 W halogenlamp. Light was delivered to the pouch through a fiberoptic system placed in the animal's mouth. Anepiscopic-fluorescent Ploem attachment was used forfluorescent microscopy. Epi-illumination was providedby a 50 W mercury arc lamp. An exciter filter forfluorescein (488 nm) was inserted between the mercurylamp and the dichroic mirror, while a barrier filter (515nm) was positioned between the dichroic minor and theoculars. The record ing system was composed of a Cohu4410 silicon intensified target (SIT) television camera(San Diego, California) coupled to an RCA timegenerator, a Sony VO 5858 video tape recorder, and anRCA monochrome video monitor TC1217. Still pho-tographs were taken either from the monitor screen witha Nikon EL 2 camera or through the trinocular head ofthe Nikon microscope with a Nikon PhotomicrographicAttachment Microflex PFX. Kodak Tri-X-Pan film wasused.

Investigation of the Pathways of PAF ActionThe select inhibitor or receptor blocker was admin-

istered in an appropriate manner, and vessel diameterwas determined as described above. PAF 10"7 M wastopically applied, and vessels with control diametersranging between 21 and 40 n-m were observed. A briefsummary for each inhibitor used follows.

Dexamethasone sodium phosphate (Decadron;Merck Sharp & Dohme, Rahway, New Jersey) wasinjected 45 minutes before application of PAF at a doseof 5 mg/kg i.p. Indomethacin was dissolved in 0.1 MTris-buflfer (pH 8.5) and given 2 hours before topicalapplication of PAF at a dose of 1 or 10 mg/kg i.p.OKY-046 [sodium (£>3-4-(l-imidazorylmethyl) phe-nyl-2-propanoate; Ono Pharmaceutical, Osaka, Japan]was dissolved in 0.9% NaCl saline solution and wasadministered at a dose of 1.5 mg/kg i.p. 30 minutesbefore the PAF challenge. BW755C (3-amino-l-/n-trifluoro methyl phenyl-2-pyrazoline hydrochloride;Burroughs Wellcome, Research Triangle Park, North


Dow

nloaded from


724 Circulation Research Vol 62, No 4, April 1988

FIGURE 1. Arteriolar constriction induced by topical application of10 7 M PAF. Top panels depict arteriole with epi-illumination:A, before PAF application; B, 2 minutes after application. Bottom panels depict an arteriole with transitlumination: A, before PAFapplication, the arteriole is approximately 40 \UN in diameter, B, after PAF application, the arteriole constricted to approximately20 \im.

Carolina) was tested at a dose of 1.5 mg/kg anddissolved in 0.9% NaCl saline solution. It wasgiven intraperitoneally 15 minutes before the topicalapplication of PAF. Kadsurenone [2-(3,4-dimeth-oxyphenyl)-2fi, 3-dihydro-3a, methoxy-3fl-methyl-5-(alfyl)-6-2//-oxobenzofuran; Merck Sharp & Dohme]was used to test the hypothesis that PAF actions arereceptor mediated. Kadsurenone was initially dis-solved in DMSO to a 100 mM stock solution andsubsequently diluted to a 0.5 mM solution in phosphatebuffer.22 Kadsurenone was administered intravenously30 minutes before chal lenge with PAF at a dose of 100(jig/kg i.v. At the administered doses, none of theinhibitors affected systemic blood pressure.

ResultsCharacterization of the Dose-Response Effects of PAFon Arteriolar Diameter

Twenty-six hamster cheek pouch preparations wereused to assess the effects of PAF 10"', 10"7, and 10"3

M on arteriolar diameter in a microvascular bed. Thethree doses tested induced vasoconstriction of the smal1arterioles observed (Figure 1). The time of onset,intensity, and duration of the vasoconstrictor responsevaried with both vessel size and the concentration ofPAF applied. Arteriolar diameters were not signifi-cantly al tered by topical appl icat ion of the PAF vehicle,DMSO/BSA/bicarbonate buffer.

The time courses of the response of arterioles to PAFare presented in Figures 2-4 . Figure 2 illustrates theeffects of PAF on the smallest group of arteriolesobserved, those ranging in size from 8 to 15 j im. PAF

10"5 M had the most profound effect on these vessels,causing an immediate constriction that lasted for 9minutes. The constriction was so severe in these vesselsthat the lumen was not detectable. Between 10 and 30minutes after application, the vessels underwent a greatdeal of vasomotion, varying between 30% and 100%control size. At 40 minutes after PAF, arteriolesassumed a diameter approximately 70% of control andremained that size through the end of the experiment.The effect of PAF 10"7 M on these vessels was lesssevere. The average initial response of arterioles was

—•— 1 0 J M PAF

- » - io-7 U PAF

- O - 10"* U PAF

O 2 4 8 8 10" 20 30 40TIME AFTER PAF (mln)

FIGURE 2. Time course of PAF-induced constriction forarterioles between 8 and 15 \un in diameter. For PAF 10'' M,22 vessels were observed in 10 animals; average luminaldiameterwas 12. 7±2.4\un. ForPAF 10''M, 25 vessels wereobserved in 6 animals; average luminaldiameter was 10.1 ±2.3\un. For PAF 10'' M, 20 vessels were observed in 5 animals;average luminal diameter was 11.6±2.7 \im. Data presentedas mean ± SD.


Dow

nloaded from



— 10"* U PAF

- 10" ' M PAF

- 10"* U PAf

~-J—}.—*

0 2 4 6 8 10 " 2 0 30 40 50 60

TIME AFTER PAF (mln)

FIGURE 3. Time course of PAF-induced constriction forarterioles between 21 and 40 \im in diameter. For PAF 10' M,28 vessels were observed in 10 animals; average luminaldiameter was 27±5.2 \un. For PAF 10~7 M, 20 vessels wereobserved in 6 animals; average luminal diameter was 28.6 ±5.8\un. For PAF 10~° M, 27 vessels were observed in 5 animals;average luminal diameter was 27.5 ±3.7 \xm. Data presentedas mean ± SD.

a slight vasodilation that occurred 1 minute afterapplication and was of very short duration. The vesselsthen constricted completely for 2 minutes. Five to 8minutes after initial application, vessels opened toapproximately 50% of their control diameters. At 10minutes, vessels were at or slightly greater than controlsize. Twenty minutes after application, vessels de-creased to approximately 80% of their control diameterand remained this size throughout the 60-minuteexperimental period. After application of 10"' M PAF,arterioles on the average retained their control sizes for2 minutes. Constriction began 3 minutes after appli-cation. Maximal constriction occurred at 7 minutes,averaging 12% of control size. Arterioles opened to anaverage value of 70% control at 8 minutes. From thistime to the end of the experimental period, arteriolesunderwent a great deal of vasomotion, ranging in sizefrom 65% to 125% control diameter.

Arterioles 21-40 \xm in diameter also constrictedafter exposure to PAF (Figure 3). In this size range, theresponses to 10"5and 10"7M PAF were similar. Vesselsbegan to constrict immediately upon application andreached maximal constriction at 3 minutes for 10"7 Mand at 4 minutes for 10"5 M PAF. Maximal constrictionvalues averaged 12% and 17% control size for PAF 10~7

and 10~5 M, respectively. The vessels then began toreopen. With 10"7 M PAF, vessels opened to anaverage of 70% control value 7 minutes after initialapplication and remained near that size throughout theexperiment. Arterioles also opened to approximately70% of their control diameter after 7 minutes and20-60 minutes after application of 10"5 M PAF.However, between 8 and 20 minutes, vessels con-stricted to 20-40% control size. After application of10"" M PAF, arterioles 21-40 ^m initially underwenta wide variation in response. Between the time ofapplication and 8 minutes afterward, some vesselsconstricted, others dilated, and still others appearedunaffected. The mean of these varied responses was aninitial constriction to 70% control followed by a

2-minute dilation to 120% control. Between 3 and 7minutes, there was a second vasoconstriction (80%control), which was followed at 8 minutes by anotherdilation (120% control). The vessels returned toapproximate control sizes 10 minutes after the initialapplication and remained at these diameters throughoutthe experiment.

PAF 10"7 and 10"5 M also induced vasoconstrictionin the largest group of arterioles tested, those rangingbetween 41 and 60 (xm in diameter (Figure 4). As withthe 21-40-u,m group, the response profiles for thesetwo doses were similar, with 10"5 M slightly moreeffective. After topical application of 10"5 M PAF,arterioles immediately began to constrict. Maximalconstriction was obtained 2-4 minutes after initialapplication, decreasing to approximately 20% controldiameter. The vessels began to open 5 minutes afterapplication and hovered at control size between 7 and20 minutes. The vessels closed again slightly to 75%control at 30-40 minutes after PAF, then opened to90% control diameter by the end of the experiment.Vessels did not respond to 10"7 M PAF the first minuteafter application. They began to constrict 2 minutesafterward and reached a maximal constriction value of20% control 4 minutes after application. The vesselsstarted to reopen 5 minutes after PAF and remained atapproximately 50% control between 5 and 8 minutes.The vessels then opened farther to approximately 75%control size 10 minutes after application and remainednear this size through the end of the experiment. Afterapplication of 10"' M PAF, vessels 41-60 (im indiameter underwent some initial vasomotion that sub-sided at 9 minutes. Throughout most of the experiment,the diameters for this size vessel ranged between 90%and 110% control.

For purposes of statistical analysis, every curveshowing vasomotor response was divided into threetime periods: 1 to 4 minutes, 4 to 10 minutes, and 10to 60 minutes. These time periods were chosenbecause, in general, they reflected the three major

I"a

\ A'"W

—•— 10"* « PAF

- - • • - 10"' M PAF

— O - 10"* U PAF

A_ •

0 2 4 6 8 10 " 2 0 X 40 50 60TIME AFTER PAF (mln)

FIGURE 4. Time course of PAF-induced constriction forarterioles between 41 and 60 \un in diameter. For PAF 10'' M,10 vessels were observed in 10 animals; average luminaldiameter was 44 ±3.5 \un. For PAF 10'' M, 13 vessels wereobserved in 6 animals; average luminal diameter was 48.8 ±4.8\mu For PAF 10'" M, 8 vessels were observed in 5 animals;average luminal diameter was 50 ±7.7 \im. Data presented asmean ± SD.


Dow

nloaded from



TABLE 1. Comparison of Regression Lines for PAF-Induced Constriction

PAF concentration

8-15 Aim

0

1 0 - ' M

10"7 M

10"3 M

21-40 /im

0

1 0 - 9 M

10" 7 M

1 0 - 5 M

41-60 urn

0

10- 9 M

10" 7 M

10" 5 M

1—4 min

0.946+ 0.010ttt§1.498-0.262t*§

1.168-0.328t*

O.Ot

1.046-0.001tt§

1.078-0.066t

0.916-0.159t

0.423-0.08U

1.006-0.004t*t§

0.965+ 0.013t*t§1.126-0.248t

0.648-0.129t

4-10 min

0.952-0.004t*§

-0.225 + 0.114t

-0 .053+ 0.096t

- l . l l + 0 . 1 6 9 t

0.917 + 0.013t§||

0.623 + 0.046t§||

-0.168 + 0.096t

0.299+ 0.020t

1.005 -0.004t*t| |

1.104-0.016t*t||

0.120 + 0.063t

0.121+0.106t

10-60 min

0.978+ 0.001t§

0.812 + O.OOlt

0.837 +0.00H

0.803 -0.002t

1.136 —0.004tt§||0.879 + 0.001t*t

0.919-0.006t

1.005-0.006t

0.671 -0.066t*t t

0.735 + 0.007t

0.871-0.003t

0.826 + O.OOlt

All diameters normalized to control for each time interval. Vessels are grouped according to control diameter inmicrometers. All table entries are the regression equation for the change in diameter with time during the time intervalshown. The first term is the intercept, and the second term is the slope.

•Slope is significantly different from 10~5 M PAF (/?<0.01); tslope is significantly different from 10~9 M(p<0.01); and tslope is significantly different from that obtained with 10"7 M PAF (p<0.01)

IDiameters are significantly larger than with 10"3 M PAF (based on their 95% confidence intervals); and || diametersare significantly larger with 10"7 M PAF (based on their 95% confidence intervals).

PAF, platelet-activating factor.

phases each vessel experienced when subjected to PAF.These phases consisted of 1) a time interval to maximalvasoconstriction; 2) a time interval for vessel reopening(rapid rate of recovery); and 3) a time interval of slowrecovery and stabilization, normally below controllevel. Regression equat ions for each of the curves weredetermined. The slopes and 95% confidence intervalsof the curves within each vessel-size group for eachtime interval were then compared for statistical dif-ferences. The slope of each curve is a measure of therate change of vessel diameter over each time intervalfor each dose of PAF. The 95% confidence interval isa measure of the relative size of the vessel over eachtime interval for each dose of PAF. The results of thesecomparisons are shown in Table 1. Since the timeintervals used were selected on the basis of the abovecriteria, some caution should be exercised whenreviewing the table, and it should be used only inconjunction with information presented in the graphs.For example, the table indicates that between 1 and 4minutes, the slope for 10"5 M PAF, vessel size 8-15ji,m, is not significantly different from control. How-ever, a comparison of their 95% confidence intervalsshows that control diameters were significantly largerthan those observed with 10"5 M PAF over this timeinterval. This is clearly shown in Figure 2.

The regression analyses, slope comparisons, andgraphs can be used to identify trends in the responsesof different sized vessels to increasing concentrationsof PAF. For every vessel-size group and time interval,the curves obtained with PAF 10~5 M differed signif-icantly from control with respect to either slope, 95%confidence interval, or both. With two exceptions, the

same differences hold true for PAF 10"7 M, while PAF10~9 M rarely induced curves that differed significantlyfrom control.

Within each size group and time interval, the slopesof the curves for PAF 10"7 and 10"5 M differedsignificantly from each other only once, between 1 and4 minutes for 8-15-u.m vessels. The graphs show that,in fact, the response of vessels to these two concen-trations of PAF are very similar. In most cases, smallerarterioles were more strongly affected by PAF thanlarger ones. In all cases, higher concentrations of PAFconstricted the arterioles to an equal or greater extentthan lower concentrations.

In separate experiments, 10~6 and 10"8 M PAF wereapplied to the pouch. Both concentrations producedarteriolar constrictions similar to 10~7 and 10"5 M PAF.Arterioles responded to PAF lO-'MO"10 M with slightvasomotion but no notable vasoconstriction.

A limited study demonstrated that venular size wasnot significantly affected by topical application of PAF.Venular flow became sluggish after PAF challenge, butsize remained fairly constant.

Effect of Reapplication of PAF on Arteriolar DiameterReapplication of PAF 90 minutes after initial ap-

plication always resulted in vasoconstriction of smallarterioles. In further experiments, it was found thatthere was a refractory period for the vasoconstrictoreffects of PAF. Reapplication 10-15 minutes afterinitial application either did not alter vessel diameteror caused a slight vasodilation. Reapplication 30, 60,and 90 minutes after initial application consistentlyinduced a second vasoconstrictor response. This re-


Dow

nloaded from



EFFECT OF INHIBITORS ONPAF-INDUCED CONSTRICTION

8 8 10 " 20 30 40TIME AFTER PAF (mln)

so to

FIGURE 5. Effect on pharmacological antagonists on PAF-induced arteriolar constriction. PAF 10'' M was applied andarterioles with luminal diameters between 21 and 40 \imobserved. KAD, kadsurenone; OKY, OKY-046; 1ND, indometh-acin; BW, BW755C; DX, dexamethasone. Data shown asmean ± SD.

sponse pattern occurred whether the same or a higherconcentration of PAF was applied. In one experiment,PAF 10"" M was applied to the pouch withoutnoticeable effect. Five minutes later, PAF 10"8 M, aconcentration that normally induced vasoconstrict ion,was administered and caused a slight dilation ofarterioles that lasted throughout the 10-minute obser-vation period.

Effect of PAF on Hematocrit and Blood PressureHematocrits ranged between 47% and 59% in

individual animals and did not change significantlyduring any one experiment as determined by theStudent's paired t test. Systemic blood pressure,ranging between 85 and 110 mm Hg, was also notinfluenced by topical application of PAF.

Effect of Pharmacological Antagonists onPAF-Induced Arteriolar Constriction

In an effort to elucidate the pathways through whichPAF induced arteriolar constriction in the hamstercheek pouch, select enzyme inhibitors and pharmaco-logical receptor antagonists were used in conjunctionwith PAF. In all experiments, PAF 10~7 M was applied,and only vessels with initial luminal diameters rangingbetween 21 and 40 u.m were observed. The dose usedwas chosen because it produced a reproducible, strong,long-lasting vasoconstrictor effect. The vessel size wasalso chosen because of the reproducibility of theresponse.

Abrogation of PAF-Induced Arteriolar Constriction byInhibitors That Prevent Formation of CyclooxygenaseProducts — Evidence That Thromboxane A2 Mediatesthe Constrictor Response

The phospholipase A2 inhibitor, dexamethasone, andBW755C, the equipotent inhibitor of cyclooxygenaseand lipoxygenase enzymes, completely blocked thevasoconstrictor response to topically applied PAF(Figure 5). Indomethacin was also found to completely

inhibit PAF-induced vasoconstrict ion (Figure 5), im-plicating cyclooxygenase products as mediators of thisevent. The specificity of the indomethacin block wastested in two experiments with norepinephrine. In thefirst experiment, norepinephrine was applied afterpretreatment with indomethacin and challenge withPAF. In the second experiment, norepinephrine wasapplied immediately after indomethacin pretreatment.In both experiments, norepinephrine induced im-mediate, intense vasoconstriction of small arterioles(Figure 6). This demonstrated that indomethacin didnot interfere with the vessels' general ability tovasoconstrict but rather was acting through a specificinterference.

To test the role of thromboxane in PAF-inducedvasoconstriction, OKY-046, a thromboxane synthetaseinhibitor, was administered. OKY-046 completelyabolished the arteriolar constrictor response to topi-cally applied PAF (Figure 5). This finding stronglyimplicated thromboxane A2 as the mediator of PAF-induced vasoconstriction.

For purposes of statistical analysis, the curves forPAF plus dexamethasone, BW755C, indomethacin,and OKY-046 were compared with the curve for 10 ~7

M PAF alone. Every curve was divided into three timeperiods as previously described; regression equationswere determined; and the slopes and 95% confidenceintervals of the lines for each time group compared forstatistical significance. The results of these compari-sons are shown in Table 2.

For the time interval between 1 and 4 minutes, boththe slopes and 95% confidence intervals of the lines fordexamethasone, BW755C, indomethacin, and OKY-046 were significantly different from the slope obtainedwith PAF 10"7 M alone. Between 4 and 10 minutes, thecurves for each inhibitor except OKY-046 were againsignificantly different from the curve for PAF 10"7 Malone. While the slopes of the lines for PAF 10"7 M andOKY-046 were similar during this time period, thediameters for OKY-046 were significantly larger thanthose obtained with 10~7 M PAF. Diameters of vesselstreated with 10"7 M PAF alone were returning towardcontrol values, while those of vessels pretreated withOKY-046 were increasing greatly above control (seeFigure 5). Between 10 and 60 minutes, the slopes forboth BW755C and kadsurenone were significantlydifferent from the slope for 10"7 M PAF alone. Thediameters for indomethacin, OKY-046, and dexameth-asone were all returning toward control values duringthis time interval after a period of vasodilation, whilethe diameters for PAF alone were decreasing away fromcontrol, which accounts for the similar slopes (seeFigure 5). However, based on their 95% confidenceintervals, the diameters obtained with all the inhibitorswere significantly greater than those of 10"7 M PAFalone over this time interval.

Abrogation of PAF-Induced Arteriolar Constriction bya Specific PAF-Receptor Antagonist, Kadsurenone

To test the hypothesis that PAF activity is receptormediated, kadsurenone, a synthetic receptor blocker of


Dow

nloaded from



A

FIGURE 6. Specificity of indomethacin blockade. Panel A, cheek pouch arterioles of a hamster pretreated with indomethacin asindicated in "Materials and Methods. " Panel B, same arterioles 2 minutes after topical application of 10~7 M PAF. Note that thearteriole did not constrict; compare with the effect of PAF 10~7 M illustrated in Figure 1. Panel C, same arterioles immediately afterapplication of norepinephrine. Note the vasoconstriction; one branch becomes so narrow that it nearly disappears. Panels A throughC are shown in the experimental temporal sequence of the blockade and tests.

PAF, was administered before PAF application. Kad-surenone pretreatment completely inhibited PAF-induced vasoconstriction (Figure 5). Statistical anal-ysis showed that the slopes for PAF plus kadsurenonewere significantly different from the curves forPAF 10~7 M alone at every time interval tested(Table 2).

Effect of Pharmacological Agents on Hematocrits andSystemic Blood Pressure

The pharmacological agents used in these experi-ments did not alter hematocrit values or affect systemicblood pressure. Values for these variables were similarto those obtained with PAF alone.

Effect of Pharmacological Agents on ArteriolarDiameters

To test whether the pharmacological agents used inthese experiments affected arteriolar diameter, vesselswere monitored for 1 hour after administration of theinhibitors and receptor blockers in separate experi-ments. BW755C had an adverse effect on flow in allpreparations observed. No other inhibitor decreasedorstopped flow. Dexamethasone, BW755C, indometh-acin, and OKY-046 had a slight vasodilatatoryeffect.

DiscussionIn this study, we investigated the direct effects of

PAF on a microvascularbed. The relation between PAFand arteriolar diameter was characterized, and thepossible pathways through which PAF could be exert-ing its vasoconstrictor effects were investigated. Ourfindings demonstrate that 1) PAF induces a constrictionof arterioles that varies with both the concentration ofPAF applied as well as the size of the arteriole affected;2) a refractory period for PAF-induced vasoconstrictionexists, lying between 15 and 30 minutes after initialapplication; 3) arteriolar constriction is blocked byenzyme inhibitors that prevent the formation of cyclo-oxygenase products; more specifically, PAF-stimulatedvasoconstriction appears to be mediated by thrombox-ane A2; and 4) vasoconstrictor responses are partiallymediated by PAF-receptor interactions.

Effect of PAF on Vessel DiameterTopical application of PAF to the microvascular bed

of the hamster cheek pouch induced constriction ofsmall arterioles. The times of onset, intensities, anddurations of the vasoconstrictor responses varied withboth vessel size and the concentration of PAF applied.PAF-induced vasoconstriction has been described qual-itatively by other investigators. Using the hamster


Dow

nloaded from



TABLE 2. Comparison of Regression Lines for Effects of Inhibitors on PAF-Induced Constriction of Arterioies(21^0 Mm)

Inhibitor

None (PAF 10"7 M)

BW755C

OKY-O46

Indomethacin

Kadsurenone

Dexamethasone

1—4 min

0.92-0.16t

0.95 + 0.O2t*t

0.97 + 0.12t*$

1.21+0.01t*t0.92+0.07t*t

0.95 + 0.02t*t

4-10 min

-0.17 + 0.10t

0.98 + 0.01t*t

1.03 + 0.07tt

1.19 + 0.01t*t

1.23-0.02t*t

1.03-0.01t*t

10-60 min

0.92-O.Olt

1.07-0.01t*t

1.32-O.Oltt

1.24 —O.Oltt

1.01+0.01t*t

0.%-O.Oltt

All diameters normalized to control for each time interval. All table entries are the regression equation for the change indiameter with time during the time interval shown. The first term is the intercept, and the second term is the slope.

•Slope is significantly different from PAF 10"7 M.tDiameters are significantly larger than with PAF 10~7 M (based on their 95% confidence intervals); and texcept for

the first minute, these diameters were significantly larger than PAF 10~7 M (based on their 95% confidence intervals).PAF, platelet-activating factor.

cheek pouch as a model, Bjork and Smedegard23

reported transient constriction of arterioies after topicalapplication of 10"' and 10~8 M PAF. Vasoconstrictionwas also observed upon intradermal injection of PAFin humans4 and guinea pigs." In vitro, PAF constrictedsmooth muscle strips of guinea pig ileum24 and lung.25

To the best of our knowledge, this is the first report toquantify the vasoconstrictor effects of PAF in amicrovascular bed and to study the time course. It isimportant to keep in mind that all vessels in themicrovascular bed are exposed to PAF at the same time.The results therefore reflect the combined direct effectsof PAF on vessel diameter as well as the indirect effectsresulting from alterations in other vascular segments.Results should be interpreted as the relation betweenPAF concentration and vascular diameter in a micro-vascular bed rather than as a strict dose-responserelation between PAF and individual arteriolar diam-eter. Perhaps the most significant contribution of thistime-course study is the finding that PAF-inducedconstrictor effects are long lasting. Between 10 and 60minutes after topical application of 10~9 and 10"7 MPAF, arterioies 8-40 jim in diameter remain con-stricted to 70-80% control size.

The mechanisms through which PAF alters vascularcaliber are not known. In contrast with consistentreports of vasoconstriction after topical application orintradermal injection, PAF has been described as ahypotensive agent, a vasoconstrictor, and a materialthat increases total peripheral resistance when admin-istered intravascularry. Vasodilation or constrictionafter intravascular administration could be centrallymediated through the stimulation of adrenergic recep-tors, secondary to cardiac alterations, or result from theproduction or inhibition of vasoactive compounds. Theconsistent arteriolar constriction caused by topicalapplication or intradermal administration of PAF ismost likely mediated locally. In our experiments,hematocrit and blood pressure were not affected bytopical application, indicating that the systemic cir-culation was not involved. The central nervous systemwas also likely unaffected.

Since PAF is produced by many blood-borne ele-ments as well as endothelial cells, it is possible that thisneutral phospholipid plays an important role in local

microcirculatory control mechanisms. In this study, wedemonstrate the vasoactive effects of PAF on a micro-vascular bed, and in a separate report, we show thatPAF also increases clearance of macromolecules.Under normal physiological conditions, vasomotortone and vascular permeability could therefore bemodulated by PAF in accordance with the particularneeds of the tissue. During inflammation, the regula-tory role of PAF in micTOcirculatory dynamics may beobliterated. PAF could be involved in the initiationand/or mediation of normal physiological inflamma-tory events, or if excessive or unbalanced amounts ofthe compound are released, it could play an importantrole in the pathophysiology of inflammation. ThePAF-induced constriction of arterioies shown in ourstudies may be of significance in the inflammatoryprocess. Immediately upon injury, local arterioiesconstrict for a few seconds to several minutes, whileperipheral arterioies often remain narrowed for muchlonger periods.26 It has been suggested that thisvasoconstriction serves a protective function.27 PAFcould therefore be involved in the initiation and/ormaintenance of this initial protective inflammatoryevent.

Pathways of PAF ActivityRole of enzymes in the arachidonic acid cascade. To

test whether the arachidonate pathway was involved inPAF activity, the effects of dexamethasone were firstassessed. Dexamethasone is a glucocorticoid that bindsto a cytosolic receptor and induces the release and denovo synthesis of macrocortin, an intracellular poly-peptide. Acting as a second messenger, macrocortininhibits phospholipase A2 activity and thereby de-creases the amount of arachidonic acid available forconversion to cyclooxygenase and lipoxygenaseproducts.28"30 Dexamethasone completely inhibitedPAF-induced vasoconstriction. BW755C, an equipo-tent inhibitor of cyclooxygenase and lipoxygenaseactivity, similarly prevented arteriolar constriction inresponse to PAF. These results implicated metabolitesof arachidonic acid as mediators of PAF-inducedactivities. To assess the role of the individual oxygen-ase pathways, animals were pretreated with indometh-acin, a nonsteroidal anti-inflammatory drug that in-


Dow

nloaded from



hibits cyclooxygenase activity.31-32 Indomethacin com-pletely abrogated the constriction of arterioles inducedby PAR The specificity of the indomethacin block wasverified by subsequent challenge with norepinephrine.These findings indicated that the vasoconstrictor effectsof PAF were mediated through the cyclooxygenasepathway. The involvement of lipoxygenase productscould not be assessed in this study because of a lack ofspecific inhibitors.

While the cyclooxygenase pathway was implicatedin PAF-induced vasoconstriction, the nature of thecyclooxygenase products responsible for the PAFactivity was unknown. Both thromboxane A2 andprostaglandin H2 are known vasoconstrictors, but it hasalso been shown that in high concentrations, vasodi-lating prostaglandins, such as prostaglandins I2 and E ,̂induce vasoconstriction, presumably by occupyingvasoconstrictor receptor sites.33 It is presently impos-sible to assess the role that individual prostaglandinsplay in mediating PAF-induced vasoconstriction, be-cause specific inhibitors for most prostaglandins havenot yet been developed. However, thromboxane syn-thetase, the enzyme responsible for conversion ofprostaglandin H2 to thromboxane A2, can be specifi-cally blocked. To investigate the role of thromboxaneA2 in PAF-induced constriction, animals were pretreat-ed with OKY-046. Pretreatment with the synthetaseinhibitor abrogated the arteriolar constriction caused byPAF, a finding that suggested that the vasoconstrictioncaused by PAF application was mediated throughthromboxane A2 production. Our finding that throm-boxane A2 is responsible for PAF-induced vasocon-striction is supported by Heffner et al,34 who found thatimidazole, another thromboxane synthetase inhibitor,reduced the PAF-induced hypertension in isolatedlungs.

Further support for the contention that thromboxaneA2 is responsible for PAF-induced arteriolar con-striction can be found in the literature. Smooth musclecells possess receptors for thromboxane A2, and thissubstance is known to contract arterial smooth musclein vitro.35 It has also been shown that thromboxane A2

can induce pulmonary hypertension in isolated lungpreparations.36

Role of receptors. The synthetic PAF receptorblocker, kadsurenone,22 inhibited the vasoconstrictorresponse to PAF. This finding implies that PAF actionsare at least in part receptor mediated.

AcknowledgmentsWe acknowledge the following generous gifts:

Kadsurenone, Merck Sharp & Dohme; BW755C,Burroughs Wellcome; and OKY-046, Ono Pharma-ceuticals.

References1. Clark PO, Hanahan DJ, Pinckard RN: Physical and chemical

properties of platelet-activating factor obtained from humanneutrophils and monocytes and rabbit neutrophils and baso-phils. Biochim Biophys Ada 1980;628:69-75

2. Camussi G, Aglietta M, Coda R, Bussolino F, Piacibello W,Tetta C: Release of platelet-activating factor and histamine. II.

The cellular origin of human PAF: Monocytes, poh/morpho-nuclear neutrophils and basophils. Immunology 1981; 42:191-199

3. Camussi G, Aglietta M, Malavasi F, Tetta A, Piacibello W,Sanavio F, Bussolino F: The release of platelet-activating factorfrom human endothelial cells in culture. J Immunol 1983;131:2397-2403

4. Pinckard RN, McManus LM, Hanahan DJ: Chemistry andbiology of acetyl glyceryl ether phosphorylcholine (platelet-activating factor), in Weissman G (ed): Advances in Inflam-mation Research, vol.4. New York, Raven Press, 1973, pp147-180

5. Gorman RR, Morton DR, Hopkins NK, Lin AH: Acetylglyceryl ether phosphorylcholine stimulates leukotriene B,,synthesis and cyclic AMP accumulation in human porymor-phonuclear leukocytes. Adv Prostaglandin Thromboxane Leu-kotriene Res 1983;12:57-63

6. Henson PM: Activation and desensitization of platelets byplatelet-activating factor (PAF) derived from IgE-sensitizedbasophils. I. Characteristics of the secretory response. J ExpMed 1976;143:937-952

7. Jouvin-Marche E, Poitevin B, Benveniste J: Platelet-activatingfactor (PAF-acether), an activator of neutrophil function.Agents Actions 1982;12:716-720

8. Henson PM, Pinckard PN: Platelet activating factor (PAF). Apossible direct mediator of anaphylaxis in the rabbit and atrigger for the vascular deposition of circulating immunecomplexes. Monogr Allergy 1977; 12:13—26

9. Henson PM, Pinckard RN: Basophil-derived platelet-activatingfactor (PAF) as an in vivo mediator of acute allergic reactions:Demonstration of specific desensitization of platelets to PAFduring IgE-induced anaphylaxis in the rabbit. J Immunol1977;119:2179-2184

10. McManus L, Pinckard N, Fitzpatrick F, O'Rourke R, CrawfordM, Hanahan DJ: Acetyl glyceryl ether phosphorylcholine.Intravascular alterations following intravenous infusion intothe baboon. Lab Invest 1981;45:303-307

11. Bessin P, Bonnet J, Apfel D, Soulard C, Desgroux L, Pelas I.Benveniste J: Acute circulatory collapse caused by platelet-activating factor (PAF-acether) in dogs. Eur J Pharm1983;86:403-413

12. McManus LM, Hanahan DJ, Demopoulos CA, Pinckard RN:Pathobiology of the intravenous infusion of acetyl glycerylether phosphorylcholine (AGEPC), a synthetic platelet-activating factor (PAF), in the rabbit. J Immunol 1980;124:2919-2924

13. Lai FM, Shepherd CA, Cerveni P, Wissner A: Hypotensive andvasodilatory activity of (±)-l-0-octadecyl-2-acetyl glyceryl-3-phosphorylcholine in the normotensive rat. Life Sci 1983;32:1159-1166

14. Levi R, Burke J, Guo Z, Hattori Y, Hoppens C, McManus L,Hanahan D, Pinckard N: Acetyl glyceryl ether phosphoryl-choline: A putative mediator of cardiac anaphylaxis in theguinea pig. Circ Res 1984;54:117-124

15. Braquet P, Vidal RF, Braquet M, Hamard H, Vargaftig BB:Involvement of leukotrienes and PAF-acether in the increasedmicrovascular permeability of the rabbit retina. Agents Actions1984; 15:82-85

16. Handley DA, Van Valen RG, Melden MK, Saunders RN:Evaluation of dose and route effects of platelet activatingfactor-induced extravasation in the guinea pig. Thromb Hae-most 1984^2:34-36

17. Pinckard RN, Halonen M, Palmer JD, Butler C, Shaw JO,Henson PM: Intravascular aggregation and pulmonary seques-tration of platelets during IgE-induced systemic anaphylaxis inthe rabbit: Abrogation of lethal anaphylactic shock by plateletdepletion. J Immunol 1977;119:2185-2193

18. O'FlahertyJT, Wykle RL, MillerCH, Lewis JC, Waite M, BassDA, McCall CE, DeChatelet LR: l-0-Alkyl-sn-glyceryl-3-phosphorylcholines: A novel class of neutrophil stimulants. AmJ Path 1981;103:70-78

19. Humphrey DM, McManus LM, Satouchi K, Hanahan D,Pinckard N: Vasoactive properties of acetyl glyceryl etherphosphorylcholine and analogues. Lab Invest 1982;46:422-427


Dow

nloaded from


Dillon el al Dose-Related Vasoconstrictor Effects of PAF 731

20. Greenblatt M, Choudari KVR, Sanders AG, Shubik P: Mam-malian microcirculation in the living animal: Methodologicconsideration. Micrvvasc Res 1969;l:420-432

21. Click RL, Gilmore JP, Joyner WL: Direct demonstration ofalterations in the microcirculation of the hamster during andfollowing renal hypertension. Circ Res 1977;41:461-467

22. Doebber TW, Wu MS, Robbins JC, Choy B, Chang MN, ShenTY: Platelet activating factor (PAF) involvement in endotoxin-induced hypotension in rats. Studies with PAF-receptor an-tagonist kadsurenone. Biochem Biophys Res Comm 1985;127:799-808

23. Bjork J, Smedegird G: Acute microvascular effects ofPAF-acether, as studied by intravital microscopy. Eur JPharmacol 1983;96:87-94

24. Findlay SR, Lichtenstein LM, Hanahan DJ, Pinckard RN:Contraction of guinea pig ileal smooth muscle by acetylgryceryl ether phosphorylcholine. Am J Physiol 1981;241:C130-C133

25. Stimler NP, Bloor CM, Hugh TE, Wykle RL, McCall CE,O'Flaherty JT: Anaphylactic actions of platelet-activatingfactor. Am J Pathol 1981;105:64-69

26. Zweifach BW: Microvascular aspects of tissue injury, inZweifach B, Grant L, McCluckey R (eds): The InflammatoryProcess. New York, Academic Press Inc, 1973, ch 1, pp 3-46

27. Spector WG, Willoughby DA: The enzyme inactivation of anadrenaline-like substance in inflammation. J Pathol Bacteriol1960;80:271-280

28. Blackwell GJ, Carnuccion R, DiRosa M, Rower FJ, ParenteL, Persico P: Macrocortin: A porypeptide causing the anti-phospholipase effect of glucocorticoids. Nature 1980;287:147-149

29. Hirato F, Schiffmann E, Venkatasubramanian K, Solomon D,

Axelrod J: A phospholipase A2 inhibitory protein in rabbitneutrophils induced by glucocorticoids. Proc Natl Acad SciUSA 1980;77:2533-2536

30. Blackwell GJ, Flower RJ: Inhibition of phospholipase. BrMedBull 1983;39:26O-264

31. Goldyne ME: Prostaglandins and other eicosanoids, in ButranG, Katzung E (eds): Basic and Clinical Pharmacology. LosAngeles, Lange Medical Publications, 1982, ch 16, pp196-204

32. Higgs GA, Vane JR: Inhibition of cyclo-oxygenase andlipoxygenase. Br Med Bull 1983;39:265-270

33. Aiken JW: Prostaglandins. J Cardiovasc Pharmacol 1984;6:S413-S420

34. Heffner JE, Shoemaker SA, Canham EM, Patel M, McMurtyIF, Morris HG, Repine JE: Acetyl glyceryl ether phosphoryl-choline-stimulated human platelets cause pulmonary hyper-tension and edema in isolated rabbit lungs. Role of thrombox-ane A2. J Clin Invest 1983;71:351-357

35. Ellis EF, Oelz 0 , Roberts LJ HI, Payne NA, Sweetman BJ,Nies AL, Oates JA: Coronary arterial smooth muscle contrac-tion by a substance released from platelets: Evidence that it isthromboxane A2. Science 1976;193:1135-1137

36. Ueno A, Tanaka K, Hirose R, Shishido M, Katori M: Possibleinvolvement of thromboxane in hypertension and bronchocon-strictor effects of LTC, and LTD4. Adv Prostaglandin Throm-boxane Uukotriene Res 1983;12:139-144

KEY WORDS • platelet-activating factor • vasoconstriction •hamster cheek pouch • microcirculation • inflammation •arachidonic acid cascade and inhibitors • thromboxane A2• leukotrienes


Dow

nloaded from


P K Dillon, A B Ritter and W N Duránmicrocirculation: dose-related relations and pathways of action.

Vasoconstrictor effects of platelet-activating factor in the hamster cheek pouch

Print ISSN: 0009-7330. Online ISSN: 1524-4571 Copyright © 1988 American Heart Association, Inc. All rights reserved.is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation Research

doi: 10.1161/01.RES.62.4.7221988;62:722-731Circ Res.

http://circres.ahajournals.org/content/62/4/722World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://circres.ahajournals.org//subscriptions/

is online at: Circulation Research Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Permissions and Rights Question and Answer about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. Further informationEditorial Office. Once the online version of the published article for which permission is being requested is

can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculation Research Requests for permissions to reproduce figures, tables, or portions of articles originally published inPermissions:


Dow

nloaded from

http://circres.ahajournals.org/content/62/4/722

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://circres.ahajournals.org//subscriptions/


vasoconstrictor effects of platelet-activating factor in the hamster

Documents