Download Viral antibodies in normal tears.

Investigative Ophthalmology & Visual Science, Vol. 29, No. 10, October 1988
Copyright © Association for Research in Vision and Ophthalmology
Viral Antibodies in Normal Tears
P. K. Coyle* and Patrick A. Sibony*t
Viruses are a common cause of eye infection. The local mucosal response, with production of antibodies released into tears, is believed to provide an important immune defense against these agents.
However very little information exists on the viral specificity of normal tear immunoglobulins. In this
study we obtained tears, parotid saliva and serum from 40 normal subjects without eye disease.
Samples were examined by enzyme linked immunosorbent assay (ELISA) for antibodies to seven
common viruses which invade mucosa: cytomegalovirus (CMV), Epstein Barr (EBV), herpes simplex
type I (HSVI), measles, mumps, rubella and varicella zoster virus (VZV). The majority of normal
tears contained antibodies to HSVI (73%) and EBV (65%), occasionally to mumps (30%), rubella
(30%), and VZV (20%), and rarely to CMV (5%). Tear viral antibodies were mainly IgA class, but it
was not unusual to find IgG antibodies to HSVI, VZV, rubella and measles. Tear and parotid saliva
immunoglobulins from the same individual had entirely different viral reactivity. In most cases tear
viral antibodies were reflected in serum viral antibodies, although the immunoglobulin class might
differ. However, 15% of normal tears had antibodies to HSVI without detectable serum antibodies.
From this study we conclude that normal tear immunoglobulins contain antibodies to common viruses,
in particular to HSVI and EBV. These tear antibodies are mainly IgA, but can consist of IgG. Viral
antibodies in tears are independent of the antibodies present in parotid saliva, suggesting that there is
preferential homing of committed B lymphocytes to different mucosal surfaces. Finally, the finding of
tear antibodies to HSVI in the absence of detectable serum antibodies suggests that the ocular surface
may be the initial infection site for this virus in some healthy normal subjects. Invest Ophthalmol Vis
Sci 29:1552-1558,1988
tized host to eradicate virus from the infected eye.8
Moreover, there is evidence that the relative concentrations of different classes of tear HSVI antibodies
may contribute to eye infection.9 Such studies require
normative data to be able to accurately interpret their
significance. Yet, there is very limited information on
the viral specificity of normal tear immunoglobulins.
In this study we collected tears and serum as well as
an additional external secretory fluid, parotid saliva,
from a group of 40 normal subjects. Using sensitive
ELISA, we examined these samples for antibodies to
The mucosal tissues of the body provide an extensive surface that is vulnerable to a variety of common
viruses.1"3 These viruses typically colonize the mucosa and may result in local infection; some go on to
produce a viremia with systemic manifestations.
Local mucosal immunity is believed to be crucial in
determining whether these agents penetrate the mucosal barrier and infect the host. Secretory antibodies
in the external secretions appear to play a major role
in providing immunity against these viruses.3"5 Some
viruses preferentially invade ocular mucosal tissues,
causing blindness or visual impairment.6-7 The precise role that tear antibodies play in these ocular infections is not well denned. For HSVI, animal studies
have suggested that antibodies are crucial for a sensi-
seven common viruses.
Materials and Methods
Subjects
Forty normal subjects without clinical signs of eye
disease or viral infection were studied. None were on
medications which would affect tear or saliva production. Informed consent was obtained from all individuals after the nature of the procedures was fully explained. There were 17 males and 23 females aged 18
to 49 years (mean 26 years).
From the Departments of *Neurology and fOphtnalmology,
Health Sciences Center, SUNY at Stony Brook, Stony Brook, New
York.
Presented in part at the Annual Meeting of the Association for
Research in Vision and Ophthalmology, Sarasota, Florida, May
1-6, 1988.
Supported by the National Eye Institute [grant EY-05736], Bethesda, Maryland.
Submitted for publication: January 15, 1988; accepted May 19,
1988.
Reprint requests: P. K. Coyle, MD, Department of Neurology,
HSC T-12, SUNY at Stony Brook, Stony Brook, NY 11794.
Sample Collection
Three different body fluids were collected from
each subject. Reflex tears were stimulated with aro-
1552
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933367/ on 05/14/2017
No. 10
1553
TEAR ANTIBODIES / Coyle and Sibony
matic ammonia and 150 /x\ collected using glass capillary tubes gently inserted into the tear pool of the
lower lid margin, as previously reported.10 To obtain
this volume some subjects were stimulated two or
three times. Care was taken not to touch or irritate
the eye directly. Parotid saliva was stimulated with a
sour lemon ball and 500 n\ collected over 2 min directly from Stenson's duct using a modified Lashley
cup." This avoids the contamination present in a
whole saliva sample. Blood was obtained by routine
venipuncture at the time of tear and saliva sampling.
After clotting, the blood was centrifuged and the
serum collected. All samples were frozen at -20°C or
-70°C until use.
Immunoglobulin Determination
IgG, IgA and IgM were determined in samples
using ELISA as previously reported.10 Secretory IgA
was determined in tears and parotid saliva using a
modification of the IgA ELISA. Samples and all reagents were diluted in phosphate buffered saline
(PBS)-0.5% Tween to prevent nonspecific binding
and plates were washed three washes with PBSTween between each step. Affinity purified goat antihuman IgA antibodies (Tago, Inc., Burlingame, CA)
at 10 Mg/ml were used to coat microtiter wells (Costar
96-well flat bottom plates, Data Packaging Corp,
Cambridge, MA), and samples were incubated in
triplicate wells as in the IgA assay. Tears were diluted
1:10,000 and 1:50,000 for this assay. Known
amounts of purified human secretory IgA (Miles Scientific, Naperville, IL) were included in each assay to
generate a standard curve. After three washes with
PBS-Tween, 100 fi\ of mouse monoclonal antibodies
to secretory component (Accurate Chemical and Scientific Corp, Westbury, NY) diluted 1:200 was added
to each well and incubated for one hour at room
temperature. After washings, 100 n\ of goat antimouse IgG antibodies conjugated to horseradishperoxidase (HRP) (Hyclone Labs, Inc., Logan, UT) diluted 1:4000 was added to each well for 1 hr. After
washing, enzyme substrate was added and developed
as in the IgA ELISA. The optical density of test samples were averaged, read from the standard curve, and
corrected by the original dilution factor.
Virus Antibodies
Samples were examined for antibodies to the following agents: CMV, EBV, HSVI, measles, mumps,
rubella and VZV. ELISA was used to detect antibodies, as previously reported.12 In brief, viral and control antigens (Clark Labs, Jamestown, NY) were
added to alternate rows of microtiter plates and incu-
bated overnight at 4°C. Antigens were diluted 1:200,
except for HSVI (1:400) and rubella (1:100). Costar
microtiter plates were used for measles, mumps and
VZV; for the other viruses Labsystems microtiter
plates (Labsystems, Chicago, IL) were used. After
washing with PBS-Tween, 25 n\ of test samples were
added to duplicate viral and control wells for 2 hours
at room temperature. Tears were diluted 1:20, parotid saliva 1:5, and sera 1:400 and 1:700 in PBSTween. These dilutions were selected by standardizing the relative concentration of mean total immunoglobulins in each body fluid (3:1:250;
tears:saliva:serum) and the minimum volume of sample required to run all the assays. The 1:400 serum
dilution was included to assure detection of low levels
of specific antibodies in serum. Certain sera were
subsequently run at 1:100 to exclude the presence of
any specific antibodies. Known positive and negative
viral antisera were included on each plate as controls.
HRP-conjugated anti-human immunoglobulin antibodies were diluted 1:2000 in PBS-Tween and used to
screen for viral antibodies; class-specific antibodies
were measured by substituting a 1:2000 dilution of
HRP-anti-human IgG, IgA or IgM antibodies as the
enzyme conjugate. In selected cases, tears were serially diluted to titrate specific antibody levels. When
sufficient sample was available, tears which showed
HSVI-IgA activity were subsequently screened for secretory IgA directed against HSVI. Reagents were
used as outlined in the secretory IgA assay above,
except on a viral microtiter plate. In this study the
term "detectable antibodies" is defined as samples
having a mean optical density (virus wells minus
control wells) greater than the mean plus three standard deviations of at least five negative control antisera run simultaneously (commercial antisera or sera
which were consistently negative in our laboratory
for specific viral antibodies). All positive tear specimens were confirmed on two separate assays.
Results
Immunoglobulin Levels
All three body fluids showed a broad range in the
concentration of immunoglobulins (Fig. 1). In tears
secretory IgA (geometric mean level 192 tig/m\) was
present in the greatest concentration, with low levels
of IgG (mean 9.7 /xg/ml) and IgM (mean 1.7 Mg/ml)
detected in all samples. Parotid saliva showed even
lower levels of secretory IgA (mean 72 tig/m\), IgG
(mean 1.4 ng/m\), and IgM (mean 0.7 fig/ml). Serum
levels were 10 to 1500-fold higher than in tears: mean
IgG was 1,539 mg/dl, mean IgA was 211 mg/dl, and
mean IgM was 798 mg/dl.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933367/ on 05/14/2017
Vol.
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / October 1988
1554
29
TEARS
2000-p
1OO T
1000-
10 •
10.0-•
A
-8-
LO100-•
10
IgM
SIgA
IgG
PAROTID SALIVA
225-r
5.00-r
10.0-
• A.
"S"
100--
A.
'f
1.00--
V
0.10--
I
0.02
-20
0.1
IgG
IgM
SIgA
SERUM
3900-
450-
160-i
• I*
• ••
t
T
100-
1000-
A
20
60
800
IgG
100--
IgA
IgM
Fig. 1. Immunoglobulin concentrations of 40 normal subjects were measured by ELISA and expressed in fig/m\ for tears and parotid saliva,
and in mg/dl for serum. IgG and IgM were measured in all three bodyfluids,secretory IgA (SIgA) in tears and saliva, and IgA in serum. The
geometric mean value ± one standard deviation is indicated by the bar and dashed lines, respectively.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933367/ on 05/14/2017
No. 10
1555
TEAR ANTIBODIES / Coyle and Sibony
Prevalence of Viral Antibodies in Tears Compared
to Saliva and Serum
Detectable tear antibodies to HSVI and EBV were
found in 73% and 65% of normal subjects respectively (Table 1). Antibodies to mumps (30%), rubella
(30%), VZV (20%), measles (10%) and CMV (5%)
were distinctly less prevalent. Parotid saliva showed a
different profile of detectable antibodies; mumps antibodies were the most prevalent, found in 70%. As
expected, sera showed a high prevalence of detectable
antibodies to all the viruses tested; only CMV reactivity was confined to a minority of these normal adults.
Antibodies to Multiple Viruses
Nearly half of our subjects had detectable tear antibodies to three or more of the viruses examined
(Table 2), whereas only 13% of saliva samples showed
detectable antiviral activity to multiple agents. All
sera were positive for antibodies to three or more of
the viruses looked at.
Compartmentalization of Viral Antibodies
The humoral response to each virus showed a distinct pattern with regard to the appearance of antibodies in various body fluids (Table 3). EBV and
HSVI antibodies were most commonly detected in
both tears and serum. Mumps antibodies were most
commonly detected in both parotid saliva and serum.
CMV, measles, rubella and VZV antibodies were
confined to serum in the majority of subjects.
Class Specificity of the Tear Viral Antibodies
Detectable tear antibodies to EBV, HSVI, mumps
and rubella consisted largely of IgA (Table 4). With
VZV and measles however, it was more likely to find
IgG than IgA antibodies. For all viruses tested there
were certain normal subjects who had tear IgG antibodies. A few individuals even showed IgM (to
mumps, rubella, EBV and measles). Of 22 tear samples with IgA directed against HSVI, there was sufficient volume to test 12 for viral specific secretory IgA.
Five of 12 tear samples showed HSVI specific secretory IgA. Parotid saliva antibodies were exclusively
IgA for almost all the viruses looked at; the only exception was CMV antibodies which were often IgG.
Serum antibodies to all viruses were most commonly
IgG. However, IgA was also frequently noted directed
against mumps, measles, rubella and VZV.
Titration of Viral Tear Antibodies
Several tear samples with IgA viral antibodies were
serially diluted and retested. Two samples with EBV
antibodies were positive to dilutions of 1:640 and
1:1280; two samples with HSVI antibodies were positive out to 1:640 and 1:2560; and two samples with
Table 1. Detection of viral antibodies in tears
compared to saliva and serum*
Serum
Virus
Tear
antibodies
Saliva
antibodies
antibodies
CMV
EBV
HSVI
Measles
Mumps
Rubella
VZV
2 (5%)
26 (65%)
29 (72.5%)
4(10%)
12(30%)
12(30%)
8 (20%)
4(10%)
7(17.5%)
1 (2.5%)
3 (7.5%)
28 (70%)
3 (7.5%)
7(17.5%)
13(32.5%)
35 (87.5%)
23 (57.5%)
28 (70%)
34 (85%)
30 (75%)
32 (80%)
* Viral antibodies were detected using ELISA. Tears, parotid saliva and
serum were obtained simultaneously from each of 40 normal subjects. Antibodies detected combined IgG, IgA and IgM class specificity. Data are expressed as the number (%) positive of the 40 normal subjects.
mumps antibodies were positive out to 1:40 and
1:10,240. One tear sample with measles antibodies
titrated out to 1:80, and another sample with VZV
antibodies titrated out to 1:40.
Viral Antibodies Confined to Tears
In initial screening dilutions (1:20 for tears, 1:5 for
saliva, 1:400 and 700 for serum) several subjects
showed anti-viral activity against EBV, HSVI,
mumps and rubella confined exclusively to tears
(Table 3). To exclude any antibodies in their sera, we
retested serum at 1:100 dilution. Antibodies were
then detected to EBV (two subjects), HSVI (four),
mumps (one) and rubella (one). However, normal
subjects remained with tear antibodies to HSVI, EBV
and rubella in the absence of any detectable serum
titers (Table 5). These tear antibodies were not confined to IgA, but included three subjects with IgG
directed against HSVI and one with IgM directed
against rubella.
Discussion
There is increasing evidence to support a common
mucosal immune system in man.13"15 Studies both in
Table 2. Prevalence of antibodies to multiple viruses*
No. of
positive
viruses
1
6
5
4
3
2
1
0
Tears
Saliva
Serum
0
1 (2.5%)
6(15%)
10 (25%)
15(37.5%)
2 (5%)
6(15%)
0
0
2 (5%)
1 (2.5%)
2 (5%)
6(15%)
20 (50%)
9 (22.5%)
2 (5%)
9 (22.5%)
15(37.5%)
10(25%)
4(10%)
0
0
0
0
* Data are expressed as the number (%) of 40 normal subjects who had
antibodies by ELISA to the seven viruses tested. Viruses looked at included
CMV, EBV, HSVI, measles, mumps, rubella and VZV. Simultaneous samples of tears, parotid saliva, and serum were examined.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933367/ on 05/14/2017
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / October 1988
1556
Vol. 29
Table 3. Compartmentalization of viral antibodies"
Detectable antibodies
Virus
CMV
EBV
HSVI
Measles
Mumps
Rubella
VZV
Tears
_
4
10
1
3
—
Tears
&
serum
Tears
&
saliva
Saliva
serum
Serum
2
19
18
3
2
8
6
_
—
—
—
1
—
—
3
—
—
1
3
—
1
1
4
—
1
16
10
Saliva
&
2
4
9
4
23
8
19
2
Tears,
saliva & serum
_
3
1
1
8
1
2
* Antibodies to seven different viruses were examined by ELISA in 40
normal subjects. Data are expressed as the number of subjects positive for
viral antibodies according to the body fluid compartments (tears, parotid
saliva and serum) in which antibodies were detected.
animals and humans have indicated that oral feeding
or aerosol exposure of antigen will induce antigenspecific antibodies in external secretions not only of
the local gastrointestinal or respiratory tract, but also
in secretions at distant mucosal sites.16"19 This is apparently due to homing of committed B cells in a
controlled migration pattern that is poorly understood.20 The local presence of antigen at a mucosal
site results in greater clonal expansion of specific antibodies than at antigen-free, distal sites.13 This system
helps explain why tears or parotid saliva might well
contain antibodies to some of the viruses we studied,
since all are mainly respiratory agents normally
spread by aerosol droplets or saliva.2 Some of these
viruses are even capable of directly infecting ocular
tissues, including CMV, EBV, HSVI, measles, rubella
and VZV.6-7'21"23 External secretions can harbor infectious agents; a variety of viruses have been isolated
from tears or saliva, often in the setting of subclinical
infection.24"27 Viruses in secretions would be expected to bind to specific antibody, forming immune
complexes and resulting in lower free antibody levels.
Although this theoretically could result in a false neg-
ative assay for viral antibody, this would only occur
very early in the immune response when antigen predominated over antibody. It is unlikely that this was a
significant factor in the present study.
The role of mucosal immunity in preventing viral
infections has been strongly suggested in a number of
studies.3"5 The mucosal defense system involves nonspecific mechanisms, such as the anatomic barrier
offered by intact cells; mucosal colonization by nonpathogenic flora; the presence of a gelatinous mucin
coating which interferes with organism attachment;
and normal components of the neutral pH external
secretions which have antimicrobial activity such as
lactoferrin, lysozyme, lactate and long-chain fatty
acids.28 Specific defense mechanisms include both
cell-mediated and humoral mucosal immune reactions. However, the best correlate of effective immunity to many viruses appears to be the production of
local antibodies by plasma cells within the conjunctiva and lacrimal gland, with their subsequent release
into secretions.
Surprisingly little is known about the antiviral
specificity of normal secretions. Most studies on tear
viral antibodies have focused on patients with ocular
infections. For example secretory IgA, IgG and IgA
tear antibodies to HSVI have been reported in patients with ocular herpetic infections.29"31 The controls included subjects with a variety of ocular disorders; only two were normal. Tear antibodies after
oral vaccination with influenza have been reported in
five healthy controls.32 The paucity of information on
normals has been due in part to the lack of sensitive
assays capable of analyzing microliter sample volumes. In this study we used ELISA, which is known
to be a sensitive and specific technique33 to detect tear
viral antibodies.
The prevalence rates of detectable serum antibodies to the seven viruses are consistent with previous
studies,2'34'35 confirming the validity of our assay.
Table 4. Class specificity of tear viral antibodies*
Virus
CMV
EBV
HSVI
Measles
Mumps
Rubella
VZV
No. antibody
positive
2t
26
29
4
12
12
8
IgG
IgA
IgM
1
2
18
3
_
26
22
2
9
6
5
_
3
—
1
5
4
—
2
6
7
* Tear samples from 40 normal subjects which had tested positive for
anti-viral immunoglobulin by ELISA were then examined by ELISA for
class-specific IgG, IgA or IgM viral reactivity. Data are expressed as the
number of subjects positive for viral antibodies.
t One tear sample with antibodies to CMV could not be tested for class
specificity.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933367/ on 05/14/2017
No. 10
1557
TEAR ANTIDODIES / Coyle and Sibony
Using only 5 n\ of tears per viral assay, we found tear
antibodies to all seven viruses. The relatively low dilution used to screen for antibodies made it unlikely
that positive tear specimens were missed. Viral reactivity was not a nonspecific backround binding, since
positive samples showed activity even when diluted
several hundred to thousand-fold. Viral antibodies in
tears and parotid saliva differed among individuals,
suggesting that these antibodies were not simply the
result of programmed B cell migration to all mucosal
compartments. Rather, it suggests that there was selection of B cells migrating to different mucosal sites,
or that local exposure to the virus differed at these
sites.
While the detection of antibodies in tears was generally mirrored by serum, there were several subjects
with detectable viral antibodies in tears not detectable
in serum. There were also a number of examples of
different classes of viral antibodies in tears and
serum. Thus the detection of tear antibodies cannot
be explained solely by simple transudation from
serum. It is possible that a selective enrichment of
committed B cells occurs within the ocular mucosa.
Normal tears showed a high prevalence of antibodies
to HSVI and EBV. In fact, HSVI antibodies were
more frequently detected in tears than in sera in this
normal population. Six subjects had tear HSVI antibodies without detectable serum antibodies. This is
an interesting and unexplained observation. It is
known that primary HSV infection can involve ocular surfaces as well as perioral or more rarely genital
areas.35 Virus can become latent in the trigeminal36 or
superior cervical ganglia,37 and with reactivation can
appear in tears.38'39 Viral transmission is probably via
infected secretions.35 Yet none of these individuals
had any history or signs of ocular infection. Could the
eye serve as a portal of entry for HSVI in certain
individuals, thus generating a local immune reaction
in the absence of a systemic reaction? This would
require subclinical viral infection, which certainly
occurs with HSVI.35 Is there preferential organ-selective homing of HSVI-committed B cells to ocular
tissues, perhaps based on lymphocyte recognition of
specific venule homing receptors,40'41 and does this
serve to protect the eye from clinical disease? We also
found that 42% of HSVI-specific tear IgA consisted at
least in part of secretory IgA. This has been touted as
a marker of active herpes infection,29 but we could
not confirm this.
Normal tears showed a high prevalence of antibodies to EBV. This is surprising, especially when compared to the rather low prevalence in saliva, since
EBV is believed to persist for life in salivary glands as
well as B lymphocytes.42 EBV can infect the eye,23
and has been implicated in autoimmune tear deficiency states.43 Perhaps EBV-committed B cells also
Table 5. Class specificity of viral antibodies
confined to tears*
Virus
No. antibody
positive
EBV
HSVI
Rubella
2
6 .
2
IgG
IgA
IgM
3.
2
4
1
1
* Several of the 40 normal subjects showed anti-viral antibodies detected
by ELISA in tears only, in the absence of serum or saliva antibodies. Sera
from these subjects were retested to exclude any detectable levels of antibodies (1:100 dilution) and remained negative. The class specificity of the tearspecific viral antibodies in this group was then examined by ELISA. Data are
reported as the number positive for IgG, IgA or IgM viral antibodies.
home to ocular tissue as a protective mechanism. Alternatively lacrimal or conjunctival tissue might act
as a reservoir for this virus.
We observed that antiviral activity in normal tears
is not limited to IgA, the major immunoglobulin in
secretions.44 There are experimental data to suggest
that the interplay of different classes of mucosal antibodies can contribute to virus-induced mucosal disease.1'9 In our normal subjects IgG antibodies were
frequent, particularly to HSVI, and tear IgM antibodies were noted in a few cases to mumps, rubella and
EBV. Most of the subjects with tear IgM antibodies
showed evidence in serum of recent viral infection.
In summary, this study has shown that a variety of
specific anti-viral antibodies are detectable in normal
tears and that these antibodies do not necessarily correlate with serum or parotid saliva antibodies in the
same individual. Tear anti-viral antibodies are
mainly of the IgA class, but can consist of IgG and in
the setting of recent systemic viral infection even
IgM. Antibodies to HSVI were particularly common
in tears, were frequently IgG, and in 15% of individuals were confined to tears in the absence of a systemic humoral response. The significance of an antiviral humoral response confined to tears, and the relationship between different classes of tear viral
antibodies, remains to be determined. ELISA now
offers the ability to probe the ocular immune response to viral infection. Future studies on selected
patient populations should help clarify the role of tear
antibodies in determining susceptibility to viral eye
invasion.
Key words: tears, viral antibodies, saliva, herpes simplex
virus, mucosal immune system
References
1. Dhar R and Ogra PL: Local immune responses. Br Med Bull
41:28, 1985.
2. Johnson RT: Viral Infections of the Nervous System. New
York, Raven Press, 1982, pp. 40-41, pp. 129-150.
3. Tomasi TB and Grey HM: Structure and function of immunoglobulin A. Prog Allergy 16:81, 1972.
4. Ogra PL and Karzon DT: Formation and function of polio-
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933367/ on 05/14/2017
1558
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
INVESTIGATIVE OPHTHALMOLOGY 6 VISUAL SCIENCE / Ocrober 1988
virus antibody in different tissues. Prog Med Virol 13:156,
1981.
Ogra PL, Kerr-Grant D, Umana G, Dzierba J, and Weintraub
D: Antibody response in serum and nasopharynx after naturally acquired and vaccine—induced infection with rubella
virus. N Engl J Med 285:1333, 1971.
Report of the Corneal Diseases Panel, Vision Research: A National Plan, 1983-1987. No. 84-2472, US Department of
Health and Human Services, p. 2.
Monnickendam MA and Darougar S: Editorial: Post measles
blindness. Br J Ophthalmol 71:325, 1987.
Lausch RN, Monteiro C, KJeinschrodt WR, and Oakes JE:
Superiority of antibody versus delayed hypersensitivity in
clearance of HSV-I from eye. Invest Ophthalmol Vis Sci
28:565, 1987.
Centifanto YM and Kaufman HE: Secretory immunoglobulin
A and herpes keratitis. Infect Immun 2:778, 1970.
Coyle PK and Sibony PA: Tear immunoglobulins measured by
ELISA. Invest Ophthalmol Vis Sci 27:622, 1986.
Mandel ID and Wotman S: The salivary secretions in health
and disease. Oral Sci Rev 8:25, 1976.
Coyle PK and Sibony PA: Viral Specificity of multiple sclerosis
tear immunoglobulins. J Neuroimmunol 14:197, 1987.
Gregory RL and Allansmith MR: Naturally occurring IgA antibodies to ocular and oral microorganisms in tears, saliva and
colustrum: Evidence for a common mucosal immune system
and local immune response. Exp Eye Res 43:739, 1986.
Mestecky J: The common mucosal immune system and current strategies for induction of immune responses in external
secretions. J Clin Immunol 7:265, 1987.
Tomasi TB and Plaut AG: Humoral aspects of mucosal immunity. In Advances in Host Defense Mechanisms, Gallin JI and
Fauci AS, editors. New York, Raven Press, 1985, pp. 31-61.
Czerkinsky C, Prince SJ, Michalek SM, Jackson S, Russell
MW, Moldoveanu Z, McGhee JR, and Mestecky J: IgA antibody-producing cells in peripheral blood after antigen ingestion: Evidence for a common mucosal immune system in
humans. Proc Natl Acad Sci USA 84:2449, 1987.
Jertborn M, Svennerholm A-M, and Holmgren J: Saliva,
breast milk and serum antibody responses as indirect measures
of intestinal immunity after oral cholera vaccination or natural
disease. J Clin Microbiol 24:203, 1986.
Mestecky J, McGhee JR, Arnold RR, Michalek SM, Prince SJ,
and Babb JL: Selective induction of an immune response in
human external secretions by ingestion of bacterial antigen. J
Clin Invest 61:731, 1978.
Weisz-Carrington P, Grimes SR and Lamm ME: Gut-associated lymphoid tissue as source of an IgA immune response in
respiratory tissues after oral immunization and intrabronchial
challenge. Cell Immunol 106:132, 1987.
Slade HB and Schwartz SA: Mucosal immunity: The immunology of breast milk. J Allergy Clin Immunol 80:346, 1987.
Aaberg TM, Cesarz TJ, and Rytel M W: Correlation of virology
and clinical course of cytomegalic retinitis. Am J Ophthalmol
74:407, 1972.
Cooper LZ: Congenital rubella in the United States. In Infections of Fetus and the Newborn Infant, Vol. 3, Krugman S and
Gershen AA, editors. New York, Alen R. Liss, 1975, pp 1-22.
Wong KW, D'Amico DJ, Hedges TR, Soong HK, Schooley
RT, and Kenyon KR. Ocular involvement associated with
chronic Epstein-Barr virus disease. Arch Ophthalmol 105:788,
1987.
Fujikawa LS, Salahuddin SZ, Palestine AG, Masur H, Nussenblatt RB, and Gallo RC: Isolation of HTLV-III from the
tears of a patient with AIDS. Lancet 2:529, 1985.
Groopman JE, Salahuddin SZ, Sarngadharan MG, Markham
PD, Gonda M, Sliski A, and Gallo RC: HTLV-III in saliva of
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
Vol. 29
people with AIDS-related complex and healthy homosexual
men at risk for AIDS. Science 226:447, 1984.
Hawkins EC, Johnson L, Pederson NC, and Winston S: Use of
tears for diagnosis of feline leukemia virus infection. J Am Vet
MedAssoc 188:1031, 1986.
Langford MP, Barber JC, Sklar VEF, Clark SW, Patriarca PA,
Onarato IM, Yin-Murphy M, and Stanton GJ: Virus-specific,
early appearing neutralizing activity and interferon in tears of
patients with acute hemorrhagic conjunctivitis. Curr Eye Res
4:233, 1985.
Smolin G: The role of tears in the prevention of infections. Int
Ophthalmol Clin 27:25, 1987.
Fox PD, Khaw PT, McBride BW, McGill JI, and Ward KA:
Tear and serum antibody levels in ocular herpetic infection:
Diagnostic precision of secretory IgA. Br J Ophthalmol 70:584,
1986.
McBride BW and Ward KA: Herpes simplex-specific IgG subclass response in herpetic keratitis. J Med Virol 21:179, 1987.
Pedersen B, Moller Andersen S, Klauber A, Ottovay E, Prause
JU, Zhong C, and Norrild B: Secretory IgA specific for herpes
simplex virus in lacrimal fluid from patients with herpes keratitis—a possible diagnostic parameter. Br J Ophthalmol
66:648, 1982.
Bergmann K-CH, Waldman RH, Tischner H, and Pohl W-D:
Antibody in tears, saliva and nasal secretions following oral
immunization of humans with inactivated influenza virus
vaccine. Int Arch Allergy Appl Immunol 80:107, 1986.
Fuccillo DA, Shekarchi IC, and Sever JL: Rapid viral diagnosis. In Manual of Clinical Laboratory Immunology, Rose
NR, Friedman H, and Fahey JL, editors. Washington DC,
American Society for Microbiology, 1986, pp. 489-496.
Baum SG and Litman N: Mumps virus. In Principles and
Practice of Infectious Disease, Mandell GL, Douglas RB, and
Bennett JE, editors. New York, John Wiley, 1985, pp.
871-877.
Hirsch MS: Herpes simplex virus. In Principles and Practice of
Infectious Disease, Mandell GL, Douglas RG, and Bennett JE,
editors. New York, John Wiley, 1985, pp. 945-952.
Baringer JR and Swoveland P: Recovery of herpes-simplex
virus from human trigeminal ganglions. N Engl J Med
288:648, 1973.
Warren KG, Brown SM, Wroblewska Z, Gilden D, Koprowski
H, and Subak-Sharpe J: Isolation of latent herpes simplex virus
from the superior cervical and vagus ganglions of human
beings. N Engl J Med 298:1068, 1978.
Stroop WG and Schaefer DC: Severity of experimentally reactivated herpetic eye disease is related to the neurovirulence of
the latent virus. Invest Ophthalmol Vis Sci 28:229, 1987.
Yamamoto Y and Hill JM: HSV-I recovery from ocular tissues
after viral inoculation into the superior cervical ganglion. Invest Ophthalmol Vis Sci 27:1447, 1986.
Pals ST, Kraal G, Horst E, Degroot A, Scheper RJ, and Meijer
CJLM: Human lymphocyte-high endothelial venule interaction: Organ-selective binding of T and B lymphocyte populations to high endothelium. J Immunol 137:760, 1986.
St John TB, Gallatin WM, Siegelman M, Fried V, Smith H,
and Weissman IL: Expression-linked cloning of a putative
lymph node homing receptor cDNA: Ubiquitin is the reactive
species. Science 231:845, 1986.
Straus SE: EB or not EB—that is the question. JAMA
257:2335, 1987.
Pflugfelder SC, Wilhelmus KR, Osato MS, Matoba AY, and
Font RL: The autoimmune nature of aqueous tear deficiency.
Ophthalmology 93:1513, 1986.
Tomasi TB, Tan EM, Solomon A, and Prendergast RA: Characteristics of an immune system common to certain external
secretions. J Exp Med 121:101, 1965.
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933367/ on 05/14/2017