Download Pathogenesis of corneal damage from pseudomonas

Volume 16
Number 1
preclude the contact between these enzymes and
membrane phospholipids with a consequent reduction in the availability of arachidonic acid for
PG synthesis and release. This hypothesis is supported by the evidence that only pharmacological
doses of corticosleroids which stabilized lysosomes also inhibited PGE release.
In addition, corticosteroids may reduce the
availability of arachidonic acid by several alternative mechanisms, such as (1) inhibition of
the release of cell membrane phospholipase, (2)
inhibition of phospholipase activity, and (3) interference with the transport of the newly formed
arachidonic acid to the microsomes.
Since CS and aspirin-like drugs act on separate
loci of PG biosynthesis, it is suggested that combined treatment with both types of drugs, in
submaximal doses, may prove beneficial in ocular
inflammation.
We are grateful to Dr. F. Kohen for supplying
us with anti-PGEi serum.
From the Department of Hormone Research,
Weizmann Institute of Science, Rehovot, Israel.
Submitted for publication July 29, 1976. Reprint
requests: Dr. U. Zor, Department of Hormone
Research, Weizmann Institute of Science, Rehovot,
Israel.
Key words: corticosteroids, prostaglandins, arachidonate, uveitis, paracentesis, inflammation, endotoxin, aqueous humor.
Reports 73
pounds on ocular prostaglandin biosynthesis,
INVEST. OPHTHALMOL. 13: 967, 1974.
8. Gryglewski, R. J., Panczenko, B., Korbut, R.,
Grodzinska, L., and Ocetkiewicz, A.: Corticosteroids inhibit prostaglandin release from
perfused mesenteric blood vessels of rabbit
and from perfused lungs of sensitized guinea
pig, Prostaglandins 10: 343, 1975.
9. Kantrowitz, F., Robinson, D. R., McGuire,
M. B., and Levine, L.: Corticosteroids inhibit prostaglandin production by rheumatoid
synovia, Nature 258: 737, 1975.
10. Floman, Y., Floman, N., and Zor, U.: Inhibition of prostaglandin E release by antiinflammatory steroids, Prostaglandins 11: 591,
1976.
11. Hong, S. C. L., and Levine, L.: Inhibition
of arachidonic acid release from cells as the
biochemical action of anti-inflammatory corticosteroids, Proc. Natl. Acad. Sci. 73: 1730,
1976.
12. Bhattacherjee, P., and Eakins, K. E.: Inhibition of ocular effects of sodium arachidonate by anti-inflammatory, compounds,
Prostaglandins 9: 175, 1975.
Pathogenesis of corneal damage from
Pseudomonas exotoxin A. BARBARA H.
IGLEWSKI, ROBERT P. BURNS, AND ILENE
K. GIPSON.
Pseudomonas aeruginosa exotoxin A was inREFERENCES
1. Ferreira, S. H., Moncada, S., and Vane, J. R.:
Prostaglandins and signs and symptoms of
inflammation, in Robinson, H. J., and Vane,
J. R., editors: Prostaglandin Synthetase Inhibitors, New York, 1974, Raven Press, p.
175.
2. Eakins, K. E., Whitelocke, R. A. F., Bennet,
A., and Martenet, A. C.: Prostaglandin-like
activity in ocular inflammation, Br. Med. J. 3:
452, 1972.
3. Bito, L. Z.: The effects of experimental
uveitis on anterior uveal prostaglandin
transport and aqueous humor composition,
INVEST. OPHTHALMOL. 13: 959, 1974.
4. Miller, J. D., Eakins, K. E., and Atwal, M.:
The release of prostaglandin E2-like activity
into aqueous humor after paracentesis and
its prevention by aspirin, INVEST. OPHTHALMOL. 12: 939, 1973.
5. Eakins, K. E.: Prostaglandins and prostaglandin synthetase inhibitors: Action in ocular
disease, in Robinson, H. J., and Vane, J. R.,
editors: Prostaglandin Synthetase Inhibitors,
New York, 1974, Raven Press, p. 343.
6. Vane, J. R.: Inhibition of prostaglandin synthesis as a mechanism of action for aspirinlike drugs, Nature New Biol. 231: 232, 1971.
7. Bhattacherjee, P., and Eakins, K. E.: A comparison of the inhibitory activity of com-
jected into rabbit corneas. Death of epithelial,
endothelial, and stromal cells resulted, and necrosis of the cornea followed. Control eyes with
exotoxin neutralized by specific antitoxin showed
minimal damage. A dose-response pattern was
evident. Antitoxin neutralization of pseudomonas
exotoxin A in corneal ulcers may have possible
therapeutic implications.
Pseudomonas aeruginosa, an opportunistic pathogen, is a common cause of severe corneal infections which progress rapidly and are very destructive.1 While endotoxin-induced damage from
bacterial cell walls is the usual mechanism of
gram-negative bacterial injury,2 a number of studies3"5 have implicated extracellular pseudomonal
proteases as the substances responsible for corneal
damage. Recently Liu0 produced from Pseudomonas aeruginosa a heat-labile protein exotoxin
(exotoxin A) which is lethal for many experimental animals and cultured mammalian cells."- 7
Exotoxin A has the same enzymatic activity as
diphtheria toxin fragment A.s Both toxins catalyze the transfer of the adenosine diphosphate
ribose (ADP-R) portion of nicotinamide adenine
dinucleotide (NAD) to mammalian elongation
factor 2 (EF-2), thereby inhibiting cellular protein synthesis.8- 9 This preliminary report de-
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933071/ on 05/14/2017
Invest. Ophthalmol. Visual Sci.
January 1977
74 Reports
Fig. 1. Central cornea or high-dose toxin-injected rabbits at 24 hours with loss ot epithelium
and endothelium (A) and only minimal remnants of keratocytes (C). Antitoxin-neutralized
exotoxin corneas show normal cellular pattern (B). (A and B x30; C xlO2.)
scribes our studies on the effect of pseudomonas
exotoxin A on rabbit eyes.
Methods, Pseudomonas aeruginosa (PA103)
was used for the production of exotoxin A.11 Exotoxin A was produced and purified as previously
described'1- s> " with the following modification.
The (NHi)-SO+ precipitated toxin was chromatographed on columns of Sephadex G200, then
DEAE-cellulose. The purified toxin had a mouse
median lethal dose (LD50) of 1 /ig per 22 grams
of Swiss Webster mice when injected intTaperitoneally and contained approximately 0.01 fig
of endotoxin per gram of protein. The toxin did
not contain detectable protease as assayed by
liquefaction of gelatin, or destruction of the enzyme activity of the toxin at 52°. s Pony antiserum
to exotoxin A was kindly provided by Liu.(i
Rabbits, albino or Dutch belted, 2 to 3 kilograms in weight, were anesthetized by intramuscular ketamine supplemented by topical proparacaine. The eyes were exposed with a lid
speculum and intracorneal injections were made
with a 30 gauge needle from a 1 c.c. tuberculin
syringe. Exotoxin A was administered in 2.5, 1.0,
and 0.1 Mg amounts, diluted to 0.1 ml. volume in
one eye, and the same weight and volume of
exotoxin, neutralized by specific antitoxin, was injected into the other eye. The needle was placed
just into the corneal stroma, approximately 0.01
ml. injected, then the needle advanced into the
resultant bleb while the rest of the 0.1 ml. dose
was injected, resulting in a milky bleb that covered
approximately two thirds of the cornea. Some of
the injected material leaked out through the injection site, but generally at least two thirds to
three fourths remained intrastromally. Within a
few hours the edema of the injection cleared.
Eyes were examined biomicroscopically daily,
then photographed and enucleated for pathologic
study at 1, 3, and 10 day intervals. Comparisons
described are between high dose (2.5 jug of toxin)
and low dose (0.1 ftg of toxin) and controls.
Results. One day after toxin injection, by biomicroscopy the entire cornea was cloudy and so
grossly thickened the iris was barely visible. There
was sometimes a purulent discharge on the eyelids, and an abscess at the injection site, and the
needle track could be seen easily. The reaction
may have been more intense in albino than the
Dutch belted rabbits but this is difficult to ascertain with the number of experiments performed
to date, since exact quantitation in small numbers
is difficult with either biomicroscopic or histopathologic preparation. Control eyes showed only
the trauma of injection at the needle track and
slight stromat opacity. Rabbit corneas injected
with low doses of exotoxin A were much less
opaque than those given the high dose, and there
was less conjunctival hyperemia and discharge.
Histopathologic examination of the whole cornea
in formalin-fixed, hematoxylin and eosin—stained
eyes of high-dose-treated rabbits showed complete absence of epithelium and endothelium and
almost total absence of keratocytes on the first
day (Fig. 1). At the corneal periphery a few cells
remained. Sometimes an abscess appeared at the
injection site, composed of rabbit polymorphonuclear leukocytes (PMN) or pseudoeosinophils,
but PMN's did not always migrate into the cornea
the first day. The low-dose exotoxin A eyes also
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933071/ on 05/14/2017
Volume 16
Number 1
Reports 75
Fig. 2. Three days after high-dose exotoxin injection, epithelium and endothelium absent and
stroma infiltrated with leukocytes. (*102.) Dead keratocyte being phagocytized by PMN, with
collagen bundles remaining intact. (xl2,000.)
showed almost total absence of corneal cells in
the center but at the periphery endothelial, epithelial, and stromal cells persisted. Acid mucopolysaccharide (AMP) staining, by a colloidal
iron technique,1" showed very marked increase in
blue staining, principally around the area where
keratocyte nuclei had been.
Electron microscopic examination of plasticembedded corneas showed de-epithelialized anterior cornea with intact subepithelial basement
membrane and collagen bundles. At the periphery
of the ulcer a few epithelial cells remained, but
these had notable lipid globules and vacuoles in the
cytoplasm and disruption of the nuclear chromatin.
Endothelium also was absent centrally, though
a few abnormal endothelial cells remained in the
peripheral cornea.
By 3 days the corneas were biomicroscopically
much more cloudy, with more purulent exudate,
although the control eyes showed only minimal
corneal haze, minimal anterior chamber protein,
and moderate hyperemia of the iris. Low doses
of toxin caused much less opacity at 3 days than
the high doses.
At 3 days, histopathologic examination of eyes
receiving high doses showed absence of endothelium and epithelium and PMN infiltration into
the stroma, with some ulceration (Fig. 2). Acid
mucopolysaccharide staining was less marked on
the third day than on the first. Central loss of
endothelium was verified by electron microscopy
but some peripheral cells showed morphologic
changes suggesting regeneration. Keratocytes appeared vacuolized and some keratocytes were
undergoing phagocytosis by PMN's. The collagen
remained intact (Fig. 2). Eyes receiving low
doses showed some intact cells through most of
the cornea.
By 10 days the high-dose exotoxin-injected
corneas were completely opaque, and often
severely ulcerated, with a hypopyon. Postinjection
cultures never showed Pseudomonas aeruginosa,
only saprophytes and nonpathogens. The control
corneas showed minimal corneal haze and moderate iris hyperemia. The eyes injected by low
doses of toxin had minimal opacity by 10 days.
In the most severely involved eyes receiving high
doses all corneal cells were absent except for an
infiltration of PMN's. Epithelial regeneration
seemed to occur in rabbits receiving low doses.
Antitoxin-treated control eyes had intense cellular
infiltration of the cornea, trabecular meshwork
(which may possibly contribute to increased intraocular pressure), and limbus.
The ulcerative process continued in the most
severely damaged eyes until 14 days, when some
eyes perforated. Animals treated with low doses
had minimal corneal opacity at 14 days, and the
control eyes were almost clear except for moderate iritis.
Discussion. It is apparent that pseudomonas
exotoxin A is a potent factor in pathogenesis of
corneal ulceration. This exotoxin kills epithelial
and endothelial cells, and presumably stromal cells
as well, in 1 day. Loss of endothelium allows the
cornea to become edematous and cloudy; loss of
epithelium and keratocytes promote later ulceration and perforation. Morphologically, these
events occur before collagen destruction and
corneal stromal ulceration.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933071/ on 05/14/2017
76 Reports
It is of interest that AMP stains are more positive in toxin-treated corneas than in controls.
Since the colloidal iron stain demonstrates nuclear
fragments as well as proteoglycans, disintegrating
nuclei may give a false impression of enhanced
proteoglycan content.
The possibility of neutralizing exotoxin A by
horse-serum antitoxin as a therapeutic approach
remains to be studied.
Invest. Ophthalmol. Visual Set.
January 1977
10. Pearse, A. G. E.: Histochemistry, Theoretical
and Applied, ed. 2, London, 1960, J. & A.
Churchill, Ltd., p. 837.
Oral vaccination and multivalent vaccine
against Pseudomonas aeruginosa keratitis. JOHN R. GERKE AND J I M S. NELSON.
We gratefully acknowledge the technical assistance of David Oldenburg.
From the Departments of Microbiology and
Immunology and of Ophthalmology, School of
Medicine, University of Oregon Health Sciences
Center, Portland, Ore. This work was supported
in part by United States Public Health Service
Grants IAI-11137 and EY-00753. Submitted for
publication July 20, 1976. Reprint requests:
Robert P. Burns, M.D., Department of Ophthalmology, University of Oregon Health Sciences
Center, 3181 S. W. Sam Jackson Park Rd., Portland, Ore. 97201.
Key words: Antitoxin, collagen, cornea, corneal
ulcer, endotoxin, exotoxin, protease, proteoglycan,
Pseudomonas aeruginosa.
REFERENCES
1. Duke-Elder, S., and Leigh, A. G.: Systems
of Ophthalmology, Vol. 8, St. Louis, 1965,
The C. V. Mosby Company, pp. 782-784.
2. Burns, R. P.: In Symposium of New Orleans
Academy of Ophthalmology: Infectious Diseases of the Conjunctiva and Cornea, St.
Louis, 1963, The C. V. Mosby Company,
p. 33.
3. Brown, S. I., Bloomfield, S. E., and Tom,
W. I.: The cornea destroying enzyme of
Pseudomonas aeruginosa, INVEST. OPHTHALMOL. 13: 174, 1974.
4. Fisher, E., Jr., and Allen, J. H.: Corneal
ulcers produced by cell-free extracts of
Pseudomonas aeniginosa, Am. J. Ophthalmol.
46: 21, 1958.
5. Kreger, A. S., and Griffin, O. K.: Physiochemical fractionation of extracellular corneadamaging proteases of Pseudomonas aeruginosa, Infect. Immunol. 9:828, 1974.
6. Liu, P. V.: Biology of Pseudomonas aeruginosa, Hosp. Pract. 11: 139, 1976.
7. Pavlovskis, O. R., Callahan, L. T., Ill, and
Pollack, M.: Pseudomonas aeruginosa exotoxin, In Schlessinger, D., editor: Microbiology 1975, Washington, D. C , 1975,
American Society for Microbiology, p. 252.
8. Vasil, M. L., Liu, P. V., and Iglewski, B. H.:
Temperature dependent inactivating factor
of Pseudomonas aeruginosa exotoxin A, Infect. Immunol. 13: 1467, 1976.
9. Iglewski, B. H., and Kabat, D.: NAD-dependent inhibition of protein synthesis by
Pseudomonas aeruginosa toxin, Proc. Natl.
Acad. Sci. 72: 2284, 1975.
Active immunization against Pseudomonas aeruginosa keratitis and systemic disease in mice was
studied. In the first series of experiments, monovalent vaccine, administered orally or intraperitoneally, protected against subsequent corneal
and intraperitoneal challenge with the homologous
strain of P. aeruginosa; however, oral administration of vaccine elicited less protection than
intraperitoneal administration. After both routes,
protection was observed at 11 and 32 days postvaccination, but it was greater at 11 days. In
the second series of experiments, multivalent
vaccine administered intraperitoneally protected
against corneal challenge with 56 to 78 percent
of 18 strains.
Pseudomonas aeruginosa keratitis following
trauma to the cornea may result in blindness.
In addition, pseudomonal septicemia and pneumonia often lead to death. Improvements in
antimicrobial therapy have reduced these threats;
however, additional means are needed.
The potential of active and passive immunization against P. aeruginosa keratitis was first shown
in 1927.x Active and passive immunotherapy based
on a multivalent vaccine have shown promise
in combating disease caused by a diversity of
antigenic types of P. aeruginosa.'- In some types
of disease, the parenteral vaccine causes severe
reactions.2 Oral vaccine offers reduced risk of
vaccine reactions.
This study was designed to determine the
answer, in part, to two questions. Does immunization with a multivalent vaccine protect against
experimental keratitis caused by diverse strains of
P. aeruginosa? Does oral immunization protect
against Pseudomonas keratitis?
Materials and methods.
P. aeruginosa cultures. The strains (Table III)
were selected for their ability to damage mouse
corneas and to represent the seven Fisher, Devlin,
and Gnabasik immunotypes. George Cole of
Parke, Davis and Co. (Detroit, Mich.) kindly
provided the immunotyping information. Four of
the strains have been reported on previously:
strains 119 and 120 were identified3 as PA7 and
PA103, and strains 186 and 187 are respectively
the type 6 and type 7 components of the heptavalent vaccine Pseudogen (Parke, Davis and Co.).
All strains were maintained in 10 percent glycerol-
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933071/ on 05/14/2017