A Hapten Model of Topically-Induced Ocular
Anaphylaxis in the Rat
Stefan D. Trocme,*§ Srefono Donini,*§ Marie C. Trocme,t Neal P. Darney,*§
Kurt J. Dloch4|| and Mathea R. Allansmirh*§
A hapten (DNP) model of topically induced ocular anaphylaxis has been developed. Rats immunized
with DNP-Ascaris were skin-tested with DNP-bovine serum albumin (DNP-BSA) and Evans blue and
challenged topically with varying amounts of di-DNP-lysine. The degree of clinical conjunctival edema
was assessed, and eye tissues were evaluated histologically. Clinical conjunctival edema and histologic
mast cell degranulation increased with higher concentration of di-DNP-lysine. In general, rats with
positive skin tests showed more clinical conjunctival edema and more mast cell degranulation than those
with negative skin tests. Three other groups of rats with positive skin tests to the DNP-BSA were
injected intravenously with I25I-BSA and challenged topically with di-DNP-lysine. Retention of 125IBSA in ocular adnexa and in globes was higher in di-DNP-lysine- than in PBS-challenged eyes. The
hapten model simulates the ocular component of human hay fever in that ocular anaphylaxis is induced
in immunized rats by topical challenge with antigen alone. Invest Ophthalmol Vis Sci 28:264-269,1987
have also been studied in guinea pig models by Dwyer
et al,7 Kathami et al,8'9 and Woodward et al.10
In the current study, ocular anaphylaxis was induced
in rats without DTT, by topical application of di-DNPlysine, a compound of relatively low molecular weight,
to eyes of rats immunized with DNP-Ascaris. Previous
studies have demonstrated that crude Ascaris suum
extract is a potent immunogen" 1 2 and that the dinitrophenylated derivative of the extract (DNP-Ascaris)
evokes production of DNP-specific IgE antibodies in
the rat.13 Ocular anaphylaxis was assessed by clinical
observation, evaluation for histologic mast cell degranulation, and determination of 125I-BSA retention
in ocular tissues.
We demonstrated previously that ocular tissues participate in systemic anaphylaxis elicited by intravenous
challenge with antigen in immunized rats.1 Ocular
anaphylaxis has also been elicited by local injection of
antigen into eye tissues of immunized rats2 and by local
injection of anti-IgE in unimmunized rats.3 Since allergic reactions in the human conjunctiva are presumably induced by topically acquired antigen, the ocular
component of human hay fever was more closely simulated by models in which rats were topically challenged with antigen after immunization with egg
albumin 4 or infection with Nippostrongylus brasiliensis.5 However, the rat conjunctiva appears to be relatively impermeable to topically administered antigens
of even moderate molecular weight, such as egg albumin and worm extract. In the above-mentioned topical
models, 45 ocular anaphylaxis was elicited only by application of antigen to eyes topically pretreated with
dithiothreitol (DTT), a mucolytic agent. Iso et al6 encountered similar problems in inducing ocular anaphylaxis by topical application of egg albumin to passively immunized rat eyes, but succeeded in inducing
ocular anaphylaxis by intravenous administration of
egg albumin. Various aspects of ocular anaphylaxis
Materials and Methods
Antigens
Ascaris suum extracts were prepared according to
Strejan and Campbell's method.12 Dinitrophenyl
(DNP) groups were added to Ascaris extracts by the
method of Eisen et al.14 Dinitrophenylated bovine
serum albumin (BSA) was prepared by a similar procedure. Crystalline N,N'-di-2,4-DNP-L-lysine (Sigma
Chemical; St. Louis, MO), MW = 478.4, was dissolved
in phosphate-buffered saline (PBS), pH 7.6, immediately before use. Because of poor solubility, concentrations greater than 1.00 mg/ml were not tested.
From the Eye Research Institute of Retina Foundation,* Forsyth
Dental Center,! Departments of Medicine^ and Ophthalmology,§
Harvard Medical School, and the Clinical Immunology and Allergy
Units, || Massachusetts General Hospital, Boston, Massachusetts.
Supported by grants EY-02099-09 and EY-07074 from the National Institutes of Health, Bethesda, Maryland.
Submitted for publication: February 24, 1986.
Reprints not available.
Immunization
Male Sprague-Dawley rats weighing approximately
250 g were injected intraperitoneally with 1 ml of a
264
Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017
No. 2
265
TOPICALLY-INDUCED OCULAR ANAPHYLAXI5 IN RATS / Trocme er QI.
Table 1. Characteristics of rat groups
Group
1
2
3
4
5
6
7
8
Number
of rats
Immunized
Challenge level
ofdi-DNP-lysine
(mg/ml)
yes
yes
0.01
0.10
yes
yes
yes
.00
.00
.00
.00
.00
.00
9
tl
14
4
5
9
3
5
yes
no
no
suspension containing DNP-Ascaris (200 jig) and alum
(20 mg).
All procedures using rats conformed to the ARVO
Resolution on the Use of Animals in Research.
Active Cutaneous Anaphylaxis (Skin Test)
On day 13, immunized rats were tested with intradermal injection (100 M0 of DNP-BSA solution (100,
10, 1 /ug/ml, diluted in PBS), followed immediately by
intravenous injection of 1 ml 1% Evans blue dye solution. Reactions were scored 30 min later by measuring the diameter of the blue skin reactions; lesions
with a diameter greater than 4 mm were considered
positive. The reactions were scored on a scale of 0 (no
injection site positive) to 3+ (all DNP-BSA injection
sites positive). Di-DNP-lysine did not elicit positive skin
tests in immunized rats at the concentrations used for
conjunctival challenge (1.00, 0.10, and 0.01 mg/ml).
Histologic Evaluation of Eye Tissues
On day 14, 34 immunized rats were assigned to one
of three groups (1, 2, and 3) for topical conjunctival
challenge with different concentrations (1.00, 0.10, and
0.01 mg/ml) ofdi-DNP-lysine (Table 1). All groups
included both skin-test negative (0) and positive (1 + 3+) rats. Skin-test results were evenly distributed in
the three groups. Prior to challenge, all rats were lightly
anesthetized with ether and given 35 mg/kg pentobarbital intraperitoneally. Rats were challenged by topical
application of 10 jul di-DNP-lysine to the experimental
eye. One hour after challenge, the rats were assessed
clinically and killed by exsanguination from the neck
arteries. Orbits were exenterated and tissues processed
as previously described;1 one micrometer of Eponembedded sections were stained with alkaline Giemsa.1
Mast cells were counted in 30 subepithelial fields, 10
fields in each of three tissue sections, obtained at least
15 jum apart. Mast cells, identified as previously described,1 were classified as granulated, containing all
purple granules or fewer than four pink granules, or
degranulated, containing four or more pink granules.
Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017
Skin-test
range
0-3+
0-3+
0-3+
3+
3+
3+
no test
no test
Interval between
antigen challenge
and death (min)
60
60
60
15
30
60
60
60
Retention of Radioiodinated Bovine Serum Albumin
(125I-BSA) in Eye Tissues
Retention of 125I-BSA in ocular tissues was assessed
using a modified Byars and Ferraresi method. 15 BSA
solutions were labeled with I25I according to the method
of Fraker and Speek.16
On day 14, 18 immunized rats scored as 3+ on the
skin test were subdivided into three groups (4, 5, and
6) (Table 1) and injected intravenously with 35 mg/kg
pentobarbital and 3 X 106 CPM 125I-BSA in 1 ml saline.
All animals were subsequently challenged with 10 p\
di-DNP-lysine (1.00 mg/ml) applied to one eye and 10
/il PBS to the fellow eye. Rats were assessed clinically
and killed 15 min (group 4), 30 min (group 5), and 60
min (group 6) after challenge. Blood samples were collected by cardiac puncture, and the animals were then
perfused with 60 ml PBS injected into the left ventricle.
The inferior vena cava was divided below the diaphragm to allow intravascular fluid to escape. After
perfusion, orbits were exenterated, and globes (including bulbar conjunctivas) were separated from the adnexa. All tissues were weighed on a Mettler H 20 balance (Mettler, Highstown, NJ). The amount of radioactivity present in tissues and serum was determined
with a gamma counter. Tissue counts in each rat were
Fig. 1. Rat undergoing ocular anaphylaxis at 1 hr (clinical reaction
of 3+) in the right eye. The left eye received PBS for control.
266
E
d
Vol. 28
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / February 1987
3.0T
3.0x
2.5 •
2.5-•
2.0
d
B
2.0-
e
1.5- •
1.0-•
1.5- •
••
1.0- •
0.5-
0.5-
0.00.01 mg/ral
NS
0.1 mo/ml
p<0.01
0.0+v,,,»,*
0.01
1.0 mg/ml
fflg/ral
0.1 mo/ml
1.0 mg/ml
p<0.01
NS
Concentration of dl-DNP-lyelne
Concentration of dl-DNP-lyalne
IOOT
IOOT
90-'
70-0
60-
40--
•
30-20-10--
' p*< 61'fl i'""
" p"< 6761"
p < o!& i"
o- •v//jf/&//za
0.01 ing/ml
1.0 fl/
p<0.05
0.1 ng/ml
Concentration of dl-DNP-lyalne
Concentration of dl-DNP-lyaine
Wfflfa Dl-ONP-lysine
I
I
PBS
Fig. 2. Di-DNP-lysine dose response in rats with positive and negative skin tests. Unshaded, PBS-treated eyes; shaded, di-DNP-lysine-treated
(antigen-challenged) eyes. Each point represents a value obtained from an individual rat; the height of the bar corresponds to the median value.
Clinical edema in skin-test positive rats (A) and skin-test negative rats (B); mast cell degranulation in skin-test positive rats (C) and skin-test
negative rats (D).
adjusted first to tissue weight and then to serum counts
(tissue CPM/tissue weight (mg)/serum CPM/^1 serum).
scale of 0 (normal conjunctiva) to 3 (maximal observed
edema).
Clinical Evaluation
Toxicity Test
Conjunctival edema of the lower eyelid was assessed
under a dissecting microscope (magnification X 3) at
different time points after challenge and rated on a
Two groups of unimmunized rats (groups 7 and 8)
served as a control for toxicity. Rats in group 7 were
challenged topically with 10 yX di-DNP-lysine (1.00 mg/
Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017
267
TOPICALLY-INDUCED OCULAR ANAPHYLAXIS IN RATS / Trocme er ol.
No. 2
B
3-
o-20
1
2
Clinical assessment
40
60
80
X mast cell degranulation
100--
300- •
60-
40"
x loo- •
20"
id" 3
1
2
Clinical assessment
1
2
Clinical assessment
Fig. 3. Correlations between skin-test rating and clinical edema (A), between skin-test rating and percentage of mast cell degranulation (B),
between % mast cell degranulation and clinical assessment (C), and between 125I-BSA retention and clinical assessment (D), in eyes challenged
with di-DNP-lysine (1 mg/ml).
ml) and killed at 60 min. Eye tissues were processed
to assess histologic mast cell degranulation. Rats in
group 8 were challenged topically with 10 /A di-DNPlysine (1.00 mg/ml) immediately after intravenous injection of 125I-BSA. The rats were assessed clinically
and killed 60 min after challenge. Eye tissues were
evaluated for I25I-BSA retention.
Statistical Analysis
The amounts of radioactivity in eye tissues and the
cell counts were compared by a one-tailed MannWhitney U test.17 Rank correlations (Spearman's rho)17
were calculated using the RS/1 software (BBN Research
Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017
Systems, Cambridge, MA). A probability of 0.05 or
less was considered significant.
Results
Histologic and Clinical Evaluation
All rats in groups 1, 2, and 3 were assessed 60 min
after challenge; clinical and histological findings were
compared.
Skin-test positive rats had significant clinical edema
(Fig. 1) following challenge with di-DNP-lysine at 1.00
(group 3; P < 0.01) and 0.10 mg/ml (group 2; P< 0.01)
levels, but not at the 0.01 mg/ml level (Fig. 2A). Skin-
Vol. 28
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / February 1987
268
skin-test negative animals a significant increase in mast
cell degranulation was observed in eyes challenged with
di-DNP-lysine at 1.00 mg/ml (group 3; P < 0.05) but
not at 0.10 (group 2) or 0.01 mg/ml (group 1) levels
(Fig. 2D).
Skin-test rating was significantly correlated with
clinical assessment (rho = 0.74; P < 0.01) (Fig. 3A)
and with mast cell degranulation (rho = 0.90; P < 0.01)
(Fig. 3B). Mast cell degranulation was also significantly
correlated with clinical assessment (rho = 0.65; P
< 0.05) (Fig. 3C).
Retention of 125I-BSA in Eye Tissues
15 Min
p<0.05
30 min
p<0.01
Time after challenge
60 min
p<0.01
Skin-test positive rats (all 3+) had significantly increased 125I-BSA accumulation in the adnexa of eyes
topically challenged with DNP-bis-lysine at 15 min
(group 4; P < 0.05), 30 min (group 5; P < 0.01), and
60 min (group 6; P < 0.01) after challenge (Fig. 4A).
The 125I-BSA retention in adnexa did not differ significantly among the 15, 30, and 60 min time points.
Significant retention of I25I-BSA was also found in globe
specimens (globe and bulbar conjunctiva) obtained
from antigen-challenged eyes at 30 (P < 0.01) and 60
min (P < 0.01), but not at 15 min (Fig. 4B). Furthermore, conjunctival edema was significantly increased
at 30 (P < 0.01) and 60 min (P < 0.01) (Fig. 5). Clinical
edema and 125I-BSA retention were significantly cor-
3.0T
2.5--
2.0-"
15 min
NS
60 min
p<0.01
30 min
p<0.01
Time after challenge
Oi-DNP-lysine
I
I PBS
Fig. 4. Retention of I25I-BSA after topical challenge with di-DNPlysine in skin-test positive rats. A, The I25I-BSA retention in adnexa
obtained from challenged eyes; B, in specimens of globe and bulbar
conjunctiva.
0.5--
o.o--
test negative animals in these groups did not have statistically significant clinical edema (Fig. 2B).
Eye tissues in skin-test positive rats showed a
significant increase in mast cell degranulation in
eyes challenged with the 0.01 (group 3; P < 0.01),
0.10 (group 2; P < 0.01), and 1.00 mg/ml (group 1; P
< 0.01) concentrations of di-DNP-lysine (Fig. 2C). In
Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017
15 min
p<0.05
30 min
p<0.01
Time after challenge
60 min
p<0.01
Di-DNP-lysine
PBS
Fig. 5. Clinical assessment at different time points after topical
challenge with di-DNP-lysine.
No. 2
TOPICALLY-INDUCED OCULAR ANAPHYLAXIS IN RATS / Trocme er ol.
269
related (rho = 0.68; P < 0.05) in groups 4, 5, and 6
combined (Fig. 3D).
tival mast cell degranulation may be present in patients
with itchy eyes that appear normal upon examination.
Toxicity Test
Key words: DNP, hapten model, ocular anaphylaxis, skin
test
Challenged eyes in group 7 had no clinical or histologic signs of anaphylaxis. Rats in group 8 showed
no evidence of clinical edema or increased 125I-BSA
uptake in di-DNP-lysine challenged eyes.
Discussion
Contrary to previous topical models, 45 which required treatment with DTT prior to topical challenge,
topically applied antigen (di-DNP-lysine), in the present hapten model, appears to penetrate the conjunctival
epithelium without DTT pretreatment. The ability to
elicit ocular anaphylaxis by application of a substance
that does not require pretreatment with DTT makes
possible pharmacologic and immunologic modulation
studies on ocular anaphylaxis without the need to control for possible effects of DTT on topically applied
drugs or on tear immunoglobulins. By skin-testing the
rats prior to ocular challenge, it was possible to assess
each rat's state of immunity to hapten. A positive skin
test (1 +-3+) provided a fairly good indication that the
eye would undergo some degree of anaphylaxis after
topical di-DNP-lysine challenge. Conversely, a negative
skin test (0) predicted a low grade or no anaphylactic
reaction after challenge.
In addition to clinical and histologic evaluations,
ocular anaphylaxis in the hapten model was assessed
by measuring retention of 125I-BSA in ocular tissues.
Retention of 125I-BSA in ocular adnexa is positively
correlated with clinical edema rating and appears to
offer a more precise and objective means of assessing
conjunctival edema than does clinical observation. The
clinical observation that antigen-challenged eyes
showed some swelling of the bulbar conjunctiva was
corroborated by a significant increase in I25I-BSA retention in globe specimens (globe and bulbar conjunctiva) obtained from antigen-challenged eyes at 30 and
60 min. However, although the adnexa had a significantly increased 125I-BSA retention at 15 min, the retention in globe specimens at that time point was not
significantly elevated. This lack of significance may be
explained by the small number of rats in the 15-min
group and the lower level of I25I-BSA accumulation
per mg tissue in globe specimens than in adnexa.
Like our previous topical rat models,4-5 this hapten
model demonstrated significant levels of mast cell degranulation in the absence of clinical edema in some
rats. This information may have clinical implications
since it suggests that ocular allergy caused by conjunc-
Downloaded From: http://iovs.arvojournals.org/ on 06/18/2017
References
1. Allansmith MR, Baird RS, and Bloch KJ: Degranulation of ocular
mast cells in rats undergoing systemic anaphylaxis. Invest
Ophthalmol Vis Sci 19:1521, 1980.
2. Allansmith MR, Bloch KJ, Baird RS, and Sinclair K: Ocular
anaphylaxis: Induction by local injection of antigen. Immunology
44:623, 1981.
3. Allansmith MR, Baird RS, and Bloch KJ: Anaphylaxis of ocular
adnexa induced by injection of anti-IgE antibody. Exp Eye Res
40:797, 1985.
4. Trocme SD, Trocme MC, Bloch KJ, and Allansmith MR: Topically induced ocular anaphylaxis in rats immunized with egg
albumin. Ophthalmic Res 18:68, 1986.
5. Trocme SD, Baird RS, Bloch KJ, and Allansmith MR: Worm
antigen-induced ocular anaphylaxis in rats infected with Nippostrongylus brasiliensis. Exp Eye Res 42:219, 1986.
6. Iso T, Nakajima N, Suda H, Yamauchi H, and Uda K: Passive
anaphylaxis in rat conjunctiva and topical effects of antiallergic
agents: hypersensitivity in conjunctiva and drug efficacy.
Ophthalmic Res 12:9, 1980.
7. Dwyer RS, Turk JL, and Darougar S: Immediate hypersensitivity
in the guinea pig conjunctiva. I. Characterisation of the IgE and
IgGl antibodies involved. Int Arch Allergy Appl Immunol 46:
910, 1974.
8. Kathami M, Donnelly JJ, John T, and Rockey JH: Vernal conjunctivitis. Model studies in guinea pigs immunized topically
with fluoresceinyl ovalbumin. Arch Ophthalmol 102:1683, 1984.
9. Kathami M, Donnelly JJ, and Rockey JH: Induction and downregulation of conjunctival type-I hypersensitivity reactions in
guinea pigs sensitized topically with fluoresceinyl ovalbumin.
Ophthalmic Res 17:139, 1985.
10. Woodward DF, Ledgard SE, and Nieves AL: Conjunctival immediate hypersensitivity: re-evaluation of histamine involvement
in the vasopermeability response. Invest Ophthalmol Vis Sci 27:
57, 1986.
11. Strejan GH, Hussain R, and Bradbury S: Regulatory role of the
immunogen in reagin formation. In Mechanisms in Allergy: Reagin-Mediated Hypersensitivity, Goodfriend L, Sehon A, and
Orange R, editors. New York, Marcel Dekker, 1973, pp. 33-42.
12. Strejan GH and Campbell DH: Hypersensitivity to Ascaris antigens. I. Skin-sensitizing activity of serum fractions from guinea
pigs sensitized to crude extracts. J Immunol 98:893, 1967.
13. Strejan GH and Marsh DG: Hapten-carrier relationships in the
production of rat homocytotropic antibodies. J Immunol 107:
306, 1971.
14. Eisen HN, Belman S, and Carsten ME: The reaction of 2,4dinitrobenzenesulfonic acid with free amino groups of proteins.
J Am Chem Soc 75:4583, 1953.
15. Byars NE and Ferraresi RW: Intestinal anaphylaxis in the rat as
a model of food allergy. Clin Exp Immunol 24:352, 1976.
16. Fraker PJ and Speck JC Jr: Protein and cell membrane iodinations with a sparingly soluble chloramide, 1,3,4,6-tetrachloro3a,6a-diphrenylglycoluril. Biochem Biophys Res Commun 80:
849, 1978.
17. Hollander M and Wolfe DA: Nonparametric Statistical Methods.
New York, Wiley, 1973, p. 71.