Download Lipid mediators of tumor necrosis factor-alpha-induced uveitis.

Investigative Ophthalmology & Visual Science, Vol. 32, No. 8, July 1991
Copyright © Association for Research in Vision and Ophthalmology
Lipid Mediators of Tumor Necrosis
Facfor-a-lnduced Uveitis
Lloyd N. Fleisher, Jenny B. Ferrell, Mariberh G. Smith, and M. Christine McGahan
The authors tested the hypothesis that platelet-activating factor (PAF) and cyclooxygenase metabolites of arachidonic acid mediate the ocular inflammatory response to intravitreally injected tumor
necrosis factor-a (TNFa). Rabbits were treated with the PAF receptor antagonist SRI 63-441, the
cyclooxygenase inhibitors indomethacin and naproxen, or SRI 63-441 and indomethacin. At 24 hr after
intravitreal injection of TNF (20,000 U), the severity of inflammation was assessed based on iridal
hypermia, aqueous humor leukocyte number and aqueous humor protein, immunoreactive-prostaglandin E (I-PGE), and leukotriene B4 (LTB4) concentrations. Although all of the treatments significantly
reduced the severity of anterior uveitis, SRI 63-441 plus indomethacin was the most effective, indomethacin and naproxen were intermediately effective, and SRI 63-441 was the least effective. The
results of this study are consistent with an important role for cyclooxygenase metabolites of arachidonic acid in the inflammatory response to TNFa, particularly with respect to dilation of iridal blood
vessels. Although naproxen was nearly as effective as indomethacin in reducing aqueous humor I-PGE
levels, indomethacin conferred significantly more protection to the blood- aqueous barrier as shown by
lower protein levels in the aqueous humor. Thus, although cyclooxygenase inhibition may partially
explain the protection afforded the blood-aqueous barrier by these nonsteroidal anti-inflammatory
agents, the data suggest that indomethacin may also exert anti-inflammatory effects that are independent of cyclooxygenase inhibition. Furthermore, the data are consistent with TNFa releasing PAF in
the eye and PAF acting both directly, by increasing vascular permeability, and indirectly, by promoting
release of cyclooxygenase and lipoxygenase metabolites of arachidonic acid. Invest Ophthalmol Vis
Sci 32:2393-2399,1991
Tumor necrosis factor-a (TNFa) is a cytokine that
is released by monocytes/macrophages in response to
bacterial lipopolysaccharide and other stimuli.12 In
nonocular tissues, TNFa is a pluripotent inflammatory mediator because of its ability to enhance neutrophil and monocyte chemotaxis,3'4 neutrophil-endothelial cell adherence,5'6 endothelial cell permeability,7 endothelial cell procoagulant properties,8 and
neutrophil and monocyte/macrophage degranulation, respiratory burst, and cytotoxicity.9"12 In the eye,
effects of TNFa are just being denned. TNFa produces uveitis when it is injected into the vitreal
chamber of the rabbit eye, as characterized by hyper-
From the Department of Anatomy, Physiological Sciences, and
Radiology, North Carolina State University, College of Veterinary
Medicine, Raleigh, North Carolina.
Supported by grant #EY-08688 from the National Institutes of
Health and a grant from the State of North Carolina.
Presented in part at the meeting of the Association for Research
in Vision and Ophthalmology, Sarasota, Florida, 1990.
Submitted for publication: September 10, 1990; accepted March
10, 1991.
Reprint requests: Lloyd N. Fleisher, PhD, Department of Anatomy, Physiological Sciences, and Radiology, North Carolina State
University, College of Veterinary Medicine, Raleigh, NC 27606.
emia of iridal blood vessels, disruption of blood-ocular barriers, infiltration of leukocytes, and elevation of
prostaglandin (PG) levels in aqueous humor.1314
TNFa stimulates the production of platelet-activating
factor (PAF)1516 and cyclooxygenase1718 and lipoxygenase19'20 metabolites of arachidonic acid (eicosanoids) in nonocular tissues. We hypothesized that the
ocular inflammatory effects of intravitreally injected
TNFa might be due to the release of PAF and eicosanoids. To test this hypothesis, uveitis was induced
in rabbit eyes by intravitreal injection of TNFa. The
effects of the cyclooxygenase inhibitors, indomethacin and naproxen, and the PAF receptor antagonist,
SRI 63-441, on the severity of inflammation were assessed by biomicroscopic, cellular, and biochemical
criteria.
Materials and Methods
Induction of Inflammation and Drug Treatments
All experiments were conducted in accordance
with the ARVO Resolution on the Use of Animals in
Research. Male 2-kg New Zealand white rabbits were
anesthetized with intramuscular injections of xylazine (12 mg/kg) and ketamine (60 mg/kg). Human
recombinant TNFa (0.1 ml; 20,000 U; 1 Mg protein;
2393
Downloaded From: http://iovs.arvojournals.org/ on 06/12/2017
2394
INVESTIGATIVE OPHTHALMOLOGY b VISUAL SCIENCE / July 1991
Genzyme Corp., Boston, MA) was injected into the
right eye; the left eye received an equal volume of
sterile vehicle that consisted of phosphate-buffered saline that contained 0.1 % low-endotoxin bovine serum
albumin (Fraction V; Sigma Chemical Co., St. Louis,
MO). Injections were made through a 30-g needle that
was attached to a syringe, approximately 3 mm posterior to the limbus. Animals were divided into five
treatment groups: group I animals received injections
of drug vehicles. The schedule and route of administration of vehicle injections corresponded to those
given to the four groups of drug-treated animals
(groups II-V). Group II animals received intraperitoneal injections of the cyclooxygenase inhibitor indomethacin (10 mg/kg) at 0.5 hr before TNFa and 8 hr
after TNFa injection. Indomethacin was delivered as
the trihydrate after reaction with equimolar amounts
of sodium carbonate. Group III animals received intraperitoneal injections of the cyclooxygenase inhibitor naproxen sodium (50 mg/kg) 0.5 hr before TNFa
and 7, 14, and 23 hr after TNFa injection. Group IV
animals received intravenous injections of the PAF
receptor antagonist SRI 63-441 (10 and 20 mg/kg) 0.5
hr before and 2, 4, and 6 hr after TNFa injection.
Group V animals were treated with indomethacin
and SRI 63-441 as described for groups II and IV. SRI
63-441 was initially given at a dose of 20 mg/kg.21"23
Because this dose produced hemolysis in three of six
animals, it was subsequently reduced to 10 mg/kg.
SRI 63-441-induced hemolysis has also been reported
in equine erythrocytes.2425 The results reported are
the combined means of four animals that received the
lower dose and three (in which hemolysis did not occur) that received the higher dose of SRI 63-441. The
same dose of indomethacin was used as that reported
to reduce PGE2 levels in aqueous humor after intravitreal injection of bacterial lipopolysaccharide.26
However, we found that two injections over 24 hr
reduced the PGE2 levels in aqueous humor after intravitreal injection of lipopolysaccharide (aqueous humor PGE2 levels: indomethacin treatment, 0.18
± 0.31 ng/ml, n = 3; indomethacin vehicle treatment,
mean: 20.17, range = 12.10 - 28.23, n = 2) as effectively as the three doses given in the previous study.
To determine the proper dose regimen for naproxen,
a dose-response experiment was performed (Table 1).
Assessment of Inflammation and Removal
of Ocular Fluids
Twenty-four hours after TNFa injection, animals
were examined with a slit lamp for intensity of iridal
hyperemia and were rated on a scale of 0-3, with 3
being the most intense response.27 Animals were then
Downloaded From: http://iovs.arvojournals.org/ on 06/12/2017
Vol. 32
Table 1. Effect of dosage regimen of naproxen
sodium on I-PGE levels in aqueous 24 hr after
intravitreal injection of E. coli lipopolysaccharide.
Dose
(mg/kg)
Injections/24 hr
0
2-4
20
2
40
2
50
2
50
4
Aqueous humor I-PGE
(ng/ml)
6.55
19.21
9.98
7.89
1.79
9.70
5.69
13.62
0.99
0.90
Data represent individual values for each animal (n = 2).
killed by exposure to an atmosphere of 100% CO2,
their eyes were removed, and the aqueous and vitreous humors were aspirated into heparin-treated syringes. Aliquots of aqueous humor (5-10 n\) were removed for determination of protein and total leukocyte cell number. Aqueous humor (100 fi\) was then
centrifuged in a Micro-centrifuge (Model 235c; Allied
Fisher Scientific, Atlanta, GA) for 1 min, and the supernatant was aspirated and transferred to polypropylene centrifuge tubes. Aqueous humor supernatants
were then subjected to organic extraction as described
below.
An inflammatory index was designed to facilitate
the comparison of the effects of each drug treatment
on the severity of the inflammatory response. For
each inflammatory parameter, the drug treatment
that produced the greatest effect was assigned a rating
of 4, and the treatment that produced the least effect
was assigned a rating of 1. Drug effects on vitreous
protein concentration were not included in the computation of the inflammatory index because this concentration was not affected by intravitreal injection of
TNFa (see Results section).
The inflammatory index was designed to provide a
simple way to visualize the effects of several drug
treatments on multiple inflammatory parameters. It
is meant to complement the actual results that are
summarized in Table 3, not substitute for them.
Protein Concentration and Leukocyte Number in
Aqueous Humor
Protein concentration in aqueous humor was determined with the use of the Lowry method28 that was
modified for reduced volumes. Total leukocyte number in aqueous humor was determined with the use of
a hemacytometer.
TUMOR NECROSIS FACTOIVa-INDUCED UVEITIS / Fleisher er ol
No. 8
2395
Effects of Cyclooxygenase Inhibitors and a PAF
Receptor Antagonist on TNFa-Induced Uveitis
Extraction of Aqueous Humor and
Radioimmunoassay (RIA) of Eicosanoids
Aqueous humor supernatant was acidified to pH
3.5 with 5 jul H3PO4 (0.49 M), extracted into six volumes of ethyl acetate (twice), dried under vacuum in a
Speed-Vac Concentrator (Savant, Hicksville, NY),
and reconstituted in 500 /x\ methanol. The efficiency
of extraction, based on recovery of [3H]-PGE2 and
[3H]-LTB4, was 63.8% and 74.2%, respectively. Aliquots of the methanol reconstitute were dried under
vacuum, reconstituted in 300 n\ of a modified Krebs
—Henseleit buffer (pH = 7.4), and PGE and LTB4
were measured by RIA. Details of the PGE RIA have
been described.14 The antiserum used in this assay
does not distinguish between PGE, and PGE2. Therefore, PG levels are referred to as immunoreactive
PGE (I-PGE). LTB4 levels were measured with the use
of a highly specific commercially available RIA kit
(Amersham, Inc., Arlington, IL).
Data Analysis
Effects of drug treatment on TNFa-induced uveitis
were evaluated using one-way ANOVA. Multiple
means were compared with the use of Tukey's Omega
test. Differences between TNFa-treated and vehicletreated eyes were determined with a paired student
t-test. Means were considered significantly different at
P < 0.05. Data are expressed as mean ± SEM.
Results
Effects of TNFa in Animals That Received
Drug Vehicles
No inflammation was seen in any eyes injected
with TNFa vehicle. In group I animals, comparison
of TNFa-injected eyes with the paired vehicle-injected eyes showed significant uveitis in the anterior
segment based on hyperemia of iridal blood vessels
and increases in aqueous humor leukocyte number
and I-PGE and LTB4 concentrations (Table 2). However, based on vitreous humor protein concentration,
the posterior segment blood-ocular barrier was not
affected at 24 hr post-TNFa injection. These results
are consistent with those of a previous report.14
Twenty-four hours after intravitreal injection of
TNFa, hyperemia of iridal blood vessels, disruption
of the blood-aqueous barrier, leukocyte infiltration
into aqueous humor and accumulation of I-PGE and
LTB4 in aqueous humor, were significantly reduced
by treatment with the cyclooxygenase inhibitors, indomethacin and naproxen, or with the PAF receptor
antagonist, SRI 63-441 (Table 3). However, differences were seen in the effectiveness of the drug treatments on each of the five inflammatory indices that
were measured. With the possible exception of LTB4
concentration in aqueous humor (see below), TNFainduced uveitis was more effectively diminished by
cyclooxygenase inhibition than by PAF receptor antagonism. Furthermore, aqueous humor I-PGE concentration and leukocyte number were more effectively reduced by PAF receptor antagonism plus cyclooxygenase inhibition than by cyclooxygenase
inhibition or PAF receptor antagonism alone. Hyperemia of iridal blood vessels was especially sensitive to
either of the cyclooxygenase inhibitors, whereas integrity of the blood-aqueous barrier, as indicated by protein concentration in aqueous humor, was equally
protected by indomethacin or indomethacin plus SRI
63-441, but less protected by naproxen or SRI 63-441
alone. With respect to the 5-lipoxygenase metabolite,
LTB4, differences in the effects of the four drug treatments were less pronounced. Indomethacin or SRI
63-441 reduced aqueous humor LTB4 concentrations
to levels lower than those seen after treatment with
naproxen. However, the effect of treatment with SRI
63-441 plus indomethacin was not significantly different from that of indomethacin, SRI 63-441, or naproxen.
In Table 4, the relative effectiveness of drug treatments is expressed in terms of the inflammatory index
(see Materials and Methods for details). Based on this
index, indomethacin or indomethacin plus SRI 63441 most effectively reduced the inflammatory response to intravitreally injected TNFa. Treatment
with the PAF receptor antagonist or with naproxen
afforded the least protection.
Table 2. Effects of intravitreal injections of TNFa and TNFa vehicle in animals that receive systemic treatment
with drug vehicles
Intravitreal
injection
Iridal
hyperemia
Aqueous cells
(\d)
Aqueous protein
(vg/nO
TNFa
TNFa vehicle
2.1*
0.0
1882 ± 160*
0.0
11.1 ± 1.4*
0.9 ±0.1
1
Significantly different from TNFa vehicle (P < 0.05).
Downloaded From: http://iovs.arvojournals.org/ on 06/12/2017
Vitreous protein
(ng/nl)
0.8 ±0.1
0.6 ±0.1
Mean ± SEM (n = 10-12).
Aqueous I-PGE
(ng/ml)
Aqueous LTB4
(ng/ml)
3.78 ± 0.62*
0.66 ±0.12
0.154 ±0.030*
0.050 ±0.012
2396
Vol. 32
INVESTIGATIVE OPHTHALMOLOGY 6 VISUAL SCIENCE / July 1991
Table 3. Effects of cyclo-oxygenase inhibition and PAF receptor antagonism on TNFa-induced uveitis
Treatment
TNF a
TNF a Indomethacin
TNF a Naproxen
TNF a SRI 63-441
TNF a Indomethacin
SRI 63-441
Aqueous cells
(id)
Aqueous protein
(vg/id)
Aqueous I-PGE
(ng/ml)
Aqueous LTB4
(ng/ml)
1882 ± 160
(11)
715±315*§"
(11)
477 ± 139*§"
(6)
1235±317*tt"
(7)
251 ± 85*f*§
(9)
11.1 ± 1.4
(12)
2.1 ±0.2*t§
3.78 ± 0.62
(12)
0.61 ± 0.15*§N
(11)
0.95 ± 0.16*§N
(6)
1.63 ± 0.3 l*tt"
(7)
0.23 ± 0.1 O*tt§
(9)
0.154 ±0.030
(10)
0.043 ±0.011**
(11)
0.077 ± 0.012*t§
(6)
0.042 ± 0.010**
(7)
0.068 ± 0.012*
(9)
Iridal
hyperemia
2.1
(12)
0.1 *§
(11)
0.3*§
(6)
1.2*tt"
(7)
0.1 *§
(9)
(ID
6.7 ± 0.8*t"
(6)
7.9 ± 1.7*t"
(7)
2.1 ±0.5*J§
(9)
§ Significantly different from treatment 4 (P < 0.05).
Significantly different from treatment 5 (P < 0.05).
Data represent mean ± SEM for the numbers of animals in parentheses.
* Significantly different from treatment 1 (P < 0.05).
t Significantly different from treatment 2 (P < 0.05).
i Significantly different from treatment 3 (P < 0.05).
11
Discussion
The results of this study support key roles for eicosanoids and PAF in the anterior uveitis seen 24 hr
after intravitreal injection of human recombinant
TNFa. Dilation of iridal blood vessels, leakage of
plasma proteins into aqueous humor, and infiltration
of leukocytes into aqueous humor were all reduced by
the cyclooxygenase inhibitors, indomethacin and naproxen, or the PAF receptor antagonist, SRI 63-441.
Uveitis that was induced by intravitreal injection of
interleukin-1, a cytokine with many properties similar to TNFa, was also dependent on eicosanoid and
PAF production.29 SRI 63-441 (tetrahydrofuranylmethoxyphosphinyloxy-ethylquinolinium) is a selective PAF receptor antagonist that prevents or reduces
many of the hemodynamic and biochemical changes
that are induced by infusion of PAF or endotoxin in
many species.21"2330 Based on the results shown in Table 3 and in the inflammatory index (Table 4), SRI
63-441 plus indomethacin most effectively reduced
TNFa-induced uveitis. This finding highlighted the
ability of this drug combination to reduce aqueous
cell number and I-PGE levels. Indomethacin was
more effective than naproxen as an inhibitor of
TNFa-induced uveitis (Tables 3 and 4). For example,
indomethacin caused a greater reduction in aqueous
humor protein and LTB4 levels than naproxen. Although the reason for the superior efficacy of indomethacin is not known, it may not be due to differen-
tial cyclooxygenase inhibition because these agents
did not differ significantly in this respect. However,
aqueous humor I-PGE levels that follow indomethacin, although not statistically different from those that
follow naproxen, were approximately 36% lower.
This decrease may contribute to the superior efficacy
of indomethacin over naproxen.
Although inflammation was assessed 24 hr after intravitreal injection of TNFa, the PAF receptor antagonist was only administered for the first 6 hr after
TNFa injection. Because of the relatively short duration of action of SRI 6 3-441,30 it is unlikely that there
was significant PAF receptor antagonist activity 24 hr
after induction of uveitis. This suggests that PAF release is a relatively early event in the uveitic response
to TNFa. Furthermore, it raises the possibility that a
PAF receptor antagonist with a longer duration of action might more dramatically reduce TNFa-induced
uveitis.
The vascular changes and disruption of the bloodaqueous barrier that occur during uveitis are believed
to be caused by the combined effects of iridal vasodilation, increased vascular permeability, and alterations in the nonpigmented epithelium of the ciliary
processes.3132 Inhibition of cyclooxygenase activity
would reduce production of vasodilatory PGs, such as
PGE2, and cause less dilation of iridal blood vessels.
As shown in Table 3, iridal hyperemia was most effectively reduced by indomethacin or naproxen. Iridal
hyperemia was also reduced by PAF receptor antago-
Table 4. Relative effectiveness of drug treatments on the inflammatory response to intravitreally injected TNFa
Inflammatory index
Drug vehicle
Indomethacin
Naproxen
SRI 63-441
Indomethacin
SRI 63-441
0.0
3.2*f
2.4*
2.0*
3.6*f
* Significantly different from drug vehicle treatment (P < 0.05).
t Significantly different from naproxen and SRI 63-441 treatments (P
< 0.05).
Downloaded From: http://iovs.arvojournals.org/ on 06/12/2017
Data represent the mean rank for five inflammatory parameters.
No. 8
TUMOR NECROSIS FACTOR-a-INDUCED UVEITIS / Fleisher er ol
nism, but to a lesser extent than with indomethacin or
naproxen. Thisfindinghas several implications. First,
it is consistent with TNFa-induced release of PAF in
the eye, a process that has been shown in nonocujari5,i6,2o a n ( j o c u i a r tissues.33"37 Second, because PAF
is generally not a vasodilator,38*39 it is not likely to
directly induce iridal hyperemia. However, PAF
could act indirectly by releasing vasodilatory PGs. Indeed, PAF-induced release of cyclooxygenase metabolites has been shown in nonocular24-4041 tissues and in
rabbit iris.42 Third, because SRI 63-441 only partially
reduced aqueous humor I-PGE levels, compared with
indomethacin and naproxen, PAF-induced PG release can account for only part of the total PG production that is induced by TNFa. The remainder of PG
release is presumably caused by a direct action of
TNFa on ocular tissues and/or on leukocytes. This
dual mode of action for TNFa-induced stimulation of
cyclooxygenase metabolite production is consistent
with the effects of drug treatments on I-PGE concentration in aqueous humor (Table 3). Compared with
group I, the blockade of PAF receptors with SRI 63441 reduced aqueous humor I-PGE levels by 57%,
whereas inhibition of cyclooxygenase activity with indomethacin or naproxen reduced these levels by 84%
and 75%, respectively. However, PAF receptor blockade plus cyclooxygenase inhibition (indomethacin
plus SRI 63-441) reduced I-PGE levels by 94%.
PAF potently alters vascular permeability as shown
by its ability to increase cutaneous43'44 and conjunctival3545 vascular permeability. Furthermore, PAF has
been implicated in the increased uveal permeability
that accompanies laser irradiation of rabbit iris37 and
the increased vascular permeability that accompanies
endotoxin-induced uveitis and paracentesis in rabbits.36 The decrease in aqueous humor protein concentration after treatment with SRI 63-441 may be
due to antagonism of the effects of PAF on vascular
permeability. However, antagonism of PAF receptors
with SRI 63-441 did not reduce aqueous humor protein concentration as effectively as indomethacin. Indomethacin-induced decreases in iridal vasodilation
that are secondary to the inhibition of vasodilatory
PG production would contribute to reductions in
aqueous humor protein concentration. However, the
findings suggest that processes that are independent of
cyclooxygenase inhibition may also be involved in indomethacin-induced preservation of the bloodaqueous barrier. Despite similar reductions in
aqueous humor I-PGE concentrations, naproxen was
less effective than indomethacin in reducing the
aqueous humor protein concentration. Furthermore,
we have reported that although another cyclooxygenase inhibitor, flunixin meglumine, was as effective as
indomethacin in reducing aqueous humor PGE2 lev-
Downloaded From: http://iovs.arvojournals.org/ on 06/12/2017
2397
els during endotoxin-induced uveitis, only indomethacin significantly reduced iridal hyperemia and
aqueous humor protein concentration.26
The reduction in aqueous humor LTB4 levels after
PAF receptor antagonism is consistent with the blockade of PAF-induced release of 5-lipoxygenase metabolites of arachidonic acid.4041-46 PAF has also been implicated in the stimulation of 5- and 12-hydroxyeicosatetraenoic acid release in a rabbit model of corneal
inflammation.34 Because LTB4 levels in aqueous humor from TNFa-vehicle injected eyes (Table 2) were
not different from those in SRI 63-441-treated animals (group IV from Table 3; student t-test, P > 0.3),
TNFa-induced stimulation of LTB4 release may be
secondary to the release of PAF. The decrease in
aqueous humor LTB4 levels that is induced by the
cyclooxygenase inhibitors is not likely due to a direct
effect on the lipoxygenase pathway, but may be a reflection of the reduction in ocular inflammation.
Also, nonsteroidal anti-inflammatory drugs, of which
indomethacin and naproxen are representatives, inhibited neutrophil functions through effects on membrane function that is independent of PG formation.47
Leukocyte number in aqueous humor was most effectively reduced by a combination of cyclooxygenase
inhibition plus PAF receptor antagonism; cyclooxygenase inhibition was intermediately effective; PAF
receptor antagonism was the least effective (see Table
3). We have shown that the leukocyte response 24 hr
after TNFa injection was largely monocytic (67%);
neutrophils represented approximately 24% of the
leukocytes that were present at this time.14 Although
LTB4 is chemotactic for neutrophils,4849 the chemotactic factors for monocytes during uveitis are poorly
understood. Therefore, it was not surprising that the
blockade of PAF receptors, which was accompanied
by maximal inhibition of LTB4 production, did not
reduce leukocyte infiltration more dramatically. Because cyclooxygenase metabolites are not believed to
possess significant chemotactic properties,5051 the association between cyclooxygenase inhibition and reduced leukocyte number in aqueous humor is more
difficult to explain. This could be an indirect effect
that is secondary to the anti-inflammatory action of
these agents.
Further investigation is required to elucidate the
mechanisms by which eicosanoids and PAF mediate
TNFa-induced uveitis. Based on the results shown, it
is clear that these potent autacoids are important mediators of the ocular inflammatory response to intravitreally injected TNFa.
Key words: tumor necrosis factor, platelet-activating factor,
eicosanoids, SRI 63-441, uveitis
2098
INVESTIGATIVE OPHTHALMOLOGY b VISUAL SCIENCE / July 1991
Acknowledgments
The authors thank Dean A. Handley, PhD, Platelet Department, Sandoz Research Institute, East Hanover, New
Jersey, for the SRI 63-441.
References
1. Nedwin GE, Svedersky LP, Bringman TS, Palladino MA, Jr,
and Goeddel DV: Effect of interleukin-2, interferon-gamma,
and mitogens on the production of tumor necrosis factors-alpha and beta. J Immunol 135:2492, 1985.
2. Beutler BA and Cerami A: Cachectin: More than a tumor necrosis factor. N Engl J Med 316:379, 1987.
3. Ming WJ, Bersani L, and Mantovani A: Tumor necrosis factor
is chemotactic for monocytes and polymorphonuclear leukocytes. J Immunol 138:1469, 1987.
4. Rampart M, de Smet W, Fiers W, and Herman AG: Inflammatory properties of recombinant tumor necrosis factor in rabbit
skin in vivo. J Exp Med 169:2227, 1989.
5. Gamble JR, Harlan JM, Klebanoff SJ, and Vadas MA: Stimulation of the adherence of neutrophils to umbilical vein endothelium by human recombinant tumor necrosis factor. Proc Natl
Acad Sci U S A 82:8667, 1985.
6. Pober JS, Gimbrone MA, Jr, Lapierre LA, Mendrick DL, Fiers
W, Rothlein R, and Springer TA: Overlapping patterns of activation of human endothelial cells by interleukin 1, tumor necrosis factor, and immune interferon. J Immunol 137:1893,
1986.
7. Horvath CJ, Ferro TJ, Jesmok G, and Malik AB: Recombinant
tumor necrosis factor increases pulmonary vascular permeability independent of neutrophils. Proc Natl Acad Sci U S A
85:9219, 1988.
8. Bevilacqua MP, Pober JS, Majeau GR, Fiers W, Cotran RS,
and Gimbrone MA, Jr: Recombinant tumor necrosis factor
induces procoagulant activity in cultured human vascular endothelium: Characterization and comparison with the actions
of interleukin 1. Proc Natl Acad Sci U S A 83:4533, 1986.
9. Philip R and Epstein LB: Tumour necrosis factor as immunomodulator and mediator of monocyte cytotoxicity induced by
itself, g-interferon and interleukin-1. Nature 323:86, 1986.
10. Larrick JW, Graham D, Toy K, Lin LS, Senyk G, and Fendly
BM: Recombinant tumor necrosis factor causes activation of
human granulocytes. Blood 69:640, 1987.
11. Nathan CF: Neutrophil activation on biological surfaces. Massive secretion of hydrogen peroxide in response to products of
macrophages and lymphocytes. J Clin Invest 80:1550, 1987.
12. Warren JS, Kunkel SL, Cunningham TW, Johnson KJ, and
Ward PA: Macrophage-derived cytokines amplify immune
complex-triggered superoxide anion responses by rat alveolar
macrophages. Am J Pathol 130:489, 1988.
13. Rosenbaum JT, Howes EL, Jr, Rubin RM, and Samples JR:
Ocular inflammatory effect of intravitreally-injected tumor necrosis factor. Am J Pathol 133:47, 1988.
14. Fleisher LN, Ferrell JB, and McGahan MC: Ocular inflammatory effects of intravitreally injected tumor necrosis factor-alpha and endotoxin. Inflammation 14:325, 1990.
15. Camussi G, Bussolino F, Salvidio G, and Baglioni C: Tumor
necrosis factor/cachectin stimulates peritoneal macrophages,
polymorphonuclear neutrophils, and vascular endothelial cells
to synthesize and release platelet-activating factor. J Exp Med
166:1390, 1987.
16. Valone FH and Epstein LB: Biphasic platelet-activating factor
synthesis by human monocytes stimulated with IL-l-b, tumor
necrosis factor, or IFN-g1. J Immunol 141:3945, 1988.
17. Dayer JM, Beutler B, and Cerami A: Cachectin/tumor necrosis
Downloaded From: http://iovs.arvojournals.org/ on 06/12/2017
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
Vol. 32
factor stimulates collagenase and prostaglandin E2 production
by human synovial cells and dermal fibroblasts. J Exp Med
162:2163, 1985.
Kawakami M, Ishibashi S, Ogawa H, Murase T, Takaku F, and
Shibata S: Cachectin/TNF as well as interleukin-1 induces
prostacyclin synthesis in cultured vascular endothelial cells.
Biochem Biophys Res Commun 141:482, 1986.
Huber M, Beutler B, and Keppler,D: Tumor necrosis factor a
stimulates leukotriene production in vivo. Eur J Immunol
18:2085, 1988.
Camussi G, Tetta C, Bussolino F, and Baglioni C: Tumor necrosis factor stimulates human neutrophils to release leukotriene B4 and platelet-activating factor. Induction of phospholipase A2 and acetyl-CoA:l-alkyl-sn-glycero-3-phosphocholine
O2-acetyltransferase activity and inhibition by antiproteinase.
Eur J Biochem 182:661, 1989.
Chang S-W, Feddersen CO, Henson PM, and Voelkel NF:
Platelet-activating factor mediates hemodynamic changes and
lung injury in endotoxin-treated rats. J Clin Invest 79:1498,
1987.
Handley DA, Tomesch JC, and Saunders RN: Inhibition of
PAF-induced systemic responses in the rat, guinea pig, dog and
primate by the receptor antagonist SRI 63-441. Thromb Haemost 56:40, 1986.
Sessler CN, Glauser FL, Davis D, and Fowler AA, III: Effects of
platelet-activating factor antagonist SRI 63-441 on endotoxemia in sheep. J Appl Physiol 65:2624, 1988.
Morris DD and Moore JN: Equine peritoneal macrophage production of thromboxane and prostacyclin in response to platelet activating factor and its receptor antagonist SRI 63-441.
Circ Shock 28:149, 1989.
Carrick JB, Moore JN, and Morris DD: Inhibition of PAF-induced aggregation of equine platelets by a PAF-receptor antagonist (SDZ 63-441). In Platelet Activating Factor in Endotoxin
and Immune Disorders, Handley DA, Saunders RN, Houlihan
WJ, and Tomesch JC, editors. New York, Marcel Dekker,
1990, pp. 77-89.
Fleisher LN: Effects of inhibitors of arachidonic acid metabolism on endotoxin-induced ocular inflammation. Curr Eye Res
7:321, 1988.
Hogan MJ, Kimura SJ, and Thygeson P: Signs and symptoms
of uveitis. I. Anterior uveitis. Am J Ophthalmol 47:155, 1959.
Lowry OH, Rosebrough NJ, Farr AL, and Randall RJ: Protein
measurement with the Folin phenol reagent. J Biol Chem,
193:265, 1951.
Rubin MR and Rosenbaum JT: A platelet-activating factor
antagonist inhibits interleukin-1-induced inflammation. Biochem Biophys Res Commun 154:429, 1988.
Handley DA, Van Valen RG, Tomesch JC, Melden MK, Jaffey JM, Ballard FH, and Saunders RN: Biological properties of
the antagonist SRI 63-441 in the PAF and endotoxin models of
hypotension in the rat and dog. Immunopharmacol 13:125,
1987.
Cole DF: The site of breakdown of the blood-aqueous barrier
under the influence of vaso-dilator drugs. Exp Eye Res 19:591,
1974.
Unger WG: Changes in the anterior uveal vessels and epithelium in the inflamed rabbit eye. Bibl Anat 18:278, 1978.
Bussolino F, Gremo F, Tetta C, Pescarmona GP, and Camussi
G: Production of platelet-activating factor by chick retina. J
Biol Chem 261:16502, 1986.
Bazan HEP, Reddy STK, Woodland JM, and Bazan NG: The
accumulation of platelet activating factor in the injured cornea
may be interrelated with the synthesis of lipoxygenase products. Biochem Biophys Res Commun 149:915, 1987.
Gautheron PD, Coulbault L, and Sugrue MF: A study of PAF-
•No. 8
36.
37.
38.
39.
40.
41.
42.
43.
TUMOR NECROSIS FACTOR-a-INDUCED UVEITIS / Fleisher er ol
induced ocular inflammation in the rat and inhibition by the
PAF antagonist, L-652,731. J Pharm Pharmacol 39:857, 1987.
Rubin RM, Samples JR, and Rosenbaum JT: Prostaglandinindependent inhibition of ocular vascular permeability by a
platelet-activating factor antagonist. Arch Ophthalmol
106:1116, 1988.
Verbey NL, Van Delft JL, Van Haeringen NJ, and Braquet P:
Platelet-activating factor and laser trauma of the iris. Invest
Ophthalmol Vis Sci 30:1101, 1989.
Braquet P, Touqui L, Shen TY, and Vargaftig BB: Perspectives
in platelet-activating factor research. Pharmacol Rev 39:97,
1987.
Piper PP and Stewart AG: Antagonism of vasoconstriction induced by platelet-activating factor in guinea-pig perfused
hearts by selective platelet-activating factor receptor antagonists. BrJ Pharmacol 90:771, 1987.
Tanizawa H and Tai HH: Synergism between chemotactic
peptide and platelet-activating factor in stimulating thromboxane B2 and leukotriene B4 biosynthesis in human neutrophils.
Biochem Pharmacol 38:2559, 1989.
Stewart AG, Dubbin PN, Harris T, and Dusting GJ: Plateletactivating factor may act as a second messenger in the release of
icosanoids and superoxide anions from leukocytes and endothelial cells. Proc Natl Acad Sci U S A 87:3215, 1990.
Yousufzai SYK and Abdel-Latif AA: Effects of platelet-activating factor on the release of arachidonic acid and prostaglandins
by rabbit iris smooth muscle. Biochem J 228:697, 1985.
Wedmore CV and Williams TJ: Platelet-activating factor
(PAF), a secretory product of polymorphonuclear leucocytes,
increases vascular permeability in rabbit skin. Br J Pharmacol
74:916P, 1981.
Downloaded From: http://iovs.arvojournals.org/ on 06/12/2017
2399
44. Humphrey DM, McManus LM, Satouchi K, Hanahan DJ, and
Pinckard RN: Vasoactive properties of acetyl glyceryl ether
phosphorylcholine and analogues. Lab Invest 46:422, 1982.
45. Stock EL, Roth SI, Kim ED, Walsh MK, and Thamman R:
The effect of platelet-activating factor (PAF), histamine, and
ethanol on vascular permeability of the guinea pig conjunctiva.
Invest Ophthalmol Vis Sci 31:987, 1990.
46. Lin AH, Morton DR, and Gorman RR: Acetyl glyceryl ether
phosphorylcholine stimulates leukotriene B4 synthesis in human polymorphonuclear leukocytes. J Clin Invest 70:1058,
1982.
47. Abramson SB, Cherksey B, Gude D, Leszczynska-Piziak J,
Philips MR, Blau L, and Weissmann G: Nonsteroidal antiinflammatory drugs exert differential effects on neutrophil function and plasma membrane viscosity. Studies in human neutrophils and liposomes. Inflammation 14:11, 1990.
48. Ford-Hutchinson AW, Bray MA, Doig MV, Shipley ME, and
Smith MJH: Leukotriene B: a potent chemokinetic and aggregating substance released from polymorphonuclear leukocytes.
Nature 286:264, 1980.
49. Bhattacherjee P, Hammond B, Salmon JA, Stepney R, and
Eakins KE: Chemotactic response to some arachidonic acid
lipoxygenase products in the rabbit eye. Eur J Pharmacol
73:21, 1981.
50. Bito LZ, Nichols RR, and Baroody RA: A comparison of the
miotic and inflammatory effects of biologically active polypeptides and prostaglandin E2 on the rabbit eye. Exp Eye Res
34:325, 1982.
51. Kulkarni PS and Srinivasan BD: The effect of intravitreal and
topical prostaglandins on intraocular inflammation. Invest
Ophthalmol Vis Sci 23:383, 1982.