Download Retinal S-antigen epitopes in vertebrate and invertebrate

1016
INVESTIGATIVE OPHTHALMOLOGY b VISUAL SCIENCE / July 1985
after age 30 and most pronounced after age 50), and
into the layer of rods and cones (only after age 40,
most common after age 50). Accompanying the loss
of nuclei in the ONL, they noted shrunken and
deformed rods and cones, and a diminution in the
numbers of photoreceptors. They also stated, however,
that nuclear displacement was rare in the center of
the fovea. Thus, although receptor cells are increasingly lost with age, no conclusive evidence is available
whether or not foveal cones are significantly affected
before the age of 50. The disks in the outer segments
of the photoreceptors which contain the photopigment
are known to be subject to a continuous process of
renewal.8 The efficacy of this mechanism might be
the reason why we find that the cones remain in fine
shape up to the age of about 50 yr.
Key words: cones, pigment density, age effects, pigment
regeneration
From the Royal Netherlands Eye Hospital,* Dondersstraat 65,
3972 JE Utrecht, and the Institute for Perception TNO,t Kampweg
5, 3769 DE Soesterberg, The Netherlands. Submitted for publication:
September 24, 1984. Reprint requests: Dirk van Norren, PhD,
Vol. 26
Royal Netherlands Eye Hospital, Dondersstraat 65, 3972 JE Utrecht,
The Netherlands.
References
1. Kilbride DE, Hutman LP, Read JS, and Fishman M: The
aging human eye and cone pigment density difference in the
fovea. ARVO Abstracts. Invest Ophthalmol Vis Sci 25(Suppl):
198, 1984.
2. Van Norren D and van der Kraats J: A continuously recording
retinal densitometer. Vision Res 21:897, 1981.
3. Smith VC, Pokorny J, and van Norren D: Densitometric
measure of human cone photopigment kinetics. Vision Res
23:517, 1983.
4. Weale RA: Senile changes in visual acuity. Trans Ophthalmol
Soc UK 95:36, 1975.
5. Kjlbride PE, Read JS, Fishman GA, and Fishman M: Determination of human cone pigment density difference spectra in
spatially resolved regions of the fovea. Vision Res 23:1341,
1983.
6. Baker HD and Kuyk TK: In vivo densitometry of cone
pigments after repeated complete bleaching. In The effects of
constant light on visual processes, Williams TP and Baker BN,
editors. New York, Plenum Press, 1980, pp. 347-353.
7. Gartner S and Henkind P: Aging and degeneration of the
human macula. 1. Outer nuclear layer and photoreceptors. Br
J Ophthalmol 65:23, 1981.
8. Young RW: Biogenesis and renewal of visual cell outer segment
membranes. Exp Eye Res 18:215, 1974.
Retinal 5-Antigen Epitopes in Vertebrate and Invertebrate Photoreceptors
Mossoud Mirshohi, Cloude Doucheix, Germoine Collenor, Brigitte Thillaye, and Jean-Pierre Faure
Monoclonal antibodies specific for the retinal S-antigen
were obtained by hybridization of spleen cells from a
BALB/c mouse immunized with bovine S-antigen and NS1 myeloma cells. Five selected antibodies specifically labeled
the photoreceptor cells of the retina by immunofluorescence.
Whereas antibody S9E2 only reacted with bovine S-antigen,
the other antibodies showed interspecies cross-reactivity.
They were used for the characterization of specific epitopes
of S-antigen in photoreceptors from a wide range of species
representative of various classes of vertebrates and invertebrates. The presence of S-antigen in distant species (vertebrates, Amphioxus, nemerteans, annelids, molluscs) indicates a high phylogenetic stability and suggests an important
role for this protein in photoreceptor function. Invest
Ophthalmol Vis Sci 26:1016-1021, 1985
The retinal "S-antigen" is a specific component of
photoreceptor cells that has been isolated and purified
from the retina of several mammals. 1 " 3 The immunologic properties of this protein include organ specificity, interspecies cross-reactivity and immunopathogenicity. Most work with S-antigen has dealt
with its ability to induce experimental autoimmune
uveoretinitis (EAU) in laboratory animals. This an-
Downloaded From: http://iovs.arvojournals.org/ on 06/16/2017
tigen is also involved in ocular autoimmune disease
in man. 45 EAU and circulating antibodies are produced after immunization with xenogenic or allogenic
S-antigen and even with autologous retina.6 These
antibodies have been used to localize S-antigen in the
retinal photoreceptors of guinea pigs by immunofluorescence.1 Cross reactivity between S-antigen from
various mammals has been demonstrated by immunodiffusion,2 enzyme immunoassay (ELISA),7 immunofluorescence1 and pathogenic activity.2 The use
of monoclonal antibodies to bovine S-antigen in
ELISA studies showed the presence of two types of
epitopes in the protein. Some epitopes were specific
to bovine S-antigen, while others were common to Santigen from various mammals. These nonspeciesspecific epitopes were recognized in the retinas of
other classes of vertebrates by immunofluorescence.8
In this article, we present an immunofluorescent
analysis of the distribution of S-antigen epitopes in
the photoreceptors of selected species representative
of various classes of vertebrates and invertebrates.
Materials and Methods. Details of hybridization
and of the specificities of the antibodies analyzed by
No. 7
ELISA have been reported elsewhere.8 Briefly, monoclonal antibodies were obtained after fusion of NS-1
myeloma cells and spleen cells from a BALB/c mouse
immunized with purified bovine S-antigen. Thirty
anti-bovine S-antigen hybridomas were detected by
ELISA. ELISA was performed by using microELISA
plates coated with purified bovine (or other mammalian) S-antigen (1 fig/m\ in 0.05 M carbonate
buffer, pH 9.6, at 4°C overnight).7 Hybridoma supernatants and peroxidase-labeled goat antimouse IgG
(EY laboratories; San Mateo, CA) were diluted in
phosphate-buffered saline (PBS) containing 0.05%
Tween 20 and 0.5% bovine serum albumin. Enzymatic
activity was detected by orthophenylenediamine 0.04%
in 0.1 M citrate buffer, pH 5, with 0.01% H 2 O 2 .
Out of 12 cloned hybridomas, three were specific
for the bovine S-antigen and the others cross-reacted
with S-antigen from various mammals. Five antibodies
were selected for this study (Table 1). They were
strongly reactive with S-antigen and showed no reactivity with other proteins or tissues in several preliminary tests, including ELISA with various proteins
and immunofluorescence on various tissues.8
For the immunohistologic study, all tissues were
fixed in Bouin solution for 18 hr, followed by standard
paraffin embedding. Five-/um sections were immediately prepared and kept at —20°C. Dissection of the
tissues varied from one species to another depending
on the size of the animal. Sections included either a
part of the retina and neighboring tissues (bovine),
or the whole eye (other vertebrates); and eventually,
a large part or the totality of the animal (Amphioxus,
invertebrates). The test was performed after elimination of paraffin by a 15-min incubation in toluene
and 5 min in ethanol followed by two washings in
PBS. The slides were covered with monoclonal antibodies as undiluted culture supernatants, then with
FITC-labeled goat antimouse IgG antibody (Nordic;
Tilburg, Netherlands), diluted 1/40. All washings
were made in PBS.
A rat antiserum was used as positive control for
the presence of S-antigen. This was a pool of 10 sera
obtained after four injections of 50 /ng of bovine
S-antigen in complete Freund's adjuvant. This serum
only revealed the S-antigen precipitation line in immunodiffusion against crude retinal extracts. Unrelated monoclonal antibodies, antifibrinogen9 and antidifferentiation10 antigens prepared in the laboratory
of one of us (CB) were used as negative controls.
Results. Testing of a wide range of tissues demonstrated that the five antibodies were specific for the
photoreceptor cells. In vertebrate retinas, antibodies
S8D8, S7D6, S2D2, S6H8 labeled different parts of
the visual cells: the reaction was strongly positive in
the outer segment, but even more in the inner
Downloaded From: http://iovs.arvojournals.org/ on 06/16/2017
1017
Reporrs
Table 1. Properties of 5 monoclonal antibodies to
retinal S antigen
Reactivity in ELISA with
purified S-antigen
Antibody
S9E2
S8D8
S7D6
S2D2
S6H8
Isotype
IgG
IgG
IgG
IgG
IgG
Cattle
Other mammals
2a
2a
1
2b
2a
* Man, swine, guinea pig, mouse.
segment; there was always a weaker labeling of the
cell body without staining of the nucleus; the outer
plexiform layer, containing the axonal and synaptic
parts of the cell, was also strongly reactive (Fig. 1).
The pattern of labeling was different with S9E2,
which is specific for bovine S-antigen and stained
predominantly the perinuclear area and less intensively the inner and outer segments.
In all mammals of this study, both cones and rods
bound monoclonal or polyclonal antibodies. In the
chicken, not all photoreceptor cells were labeled, the
highest percentage of positive cells was in the visual
axis, whereas the number of positive cells decreased
in the periphery of the retina. In reptiles, all photoreceptor cells were positive (viper, lizard), or, in the
turtle retina, all photoreceptors located in the visual
axis were positive, whereas progressive decrease of
intensity was observed towards the periphery of the
retina. Urodeles and frog showed a different pattern
of labeling with monoclonal antibodies between cones
and rods in that cones were positive and rods were
negative, whereas in Xenopus all photoreceptors were
labeled. However, in every amphibian, the polyclonal
serum against S-antigen stained both rods and cones.
In fishes, all photoreceptor cells were positively stained.
The distribution of S-antigen epitopes in the various
species is shown in Table 2. The epitope S9E2 was
the most restricted since it was present only in bovine
photoreceptors. The epitopes recognized by antibodies
S8D8, S7D6, S2D2, and S6H8 were widely expressed
and were found as a group in all vertebrate retinas,
with the exception of the frog, where only S7D6 gave
a positive staining. The four epitopes were also present
in the protochordate Amphioxus, in the nemertean
Lineus and the annelid Nereis (Fig. 1). Three epitopes
were found in the mollusc Pecten and two in the
mollusc Helix. None of the antibodies stained the
photoreceptors in the compound eyes of three arthropods, or in the simple eyes of a planarian and a
starfish. In protochordates, molluscs, and annelids we
found species which were positive with several antibodies as well as species that were negative. The
polyclonal serum yielded similar results as the mono-
1018
INVESTIGATIVE OPHTHALMOLOGY b VISUAL SCIENCE / July 1985
Vol. 26
'If
1
Mi
Downloaded From: http://iovs.arvojournals.org/ on 06/16/2017
Fig. 1. Immunofluorescent staining of photoreceptor cells by monoclonal antibody S6H8.
In vertebrates, man (A) or turtle Pseudemys
(B), the labeling is limited to the outer and
inner segment, the perinuclear and synaptic
parts of visual cells. In Amphioxus (C), the
antibody strongly reacted with the Hesse cells,
photoreceptors scattered along the neural tube
and capped by crescent-shaped pigment cells.
In the annelid Nereis (D), the brightly labeled
photoreceptor structures form a continuous
layer located in the inner part of the ocellus
wall. The label can also be observed on the
other side of the pigmented layer.
1019
Reports
No. 7
Table 2. Immunofluorescent reactivity of 5 monoclonal antibodies and a rat polyclonal antiserum (AS)
to S-antigen on photoreceptors from various species
S9E2
VERTEBRATES
Mammals
Man
Cat
Ox
Rat, mouse, guinea pig, rabbit
Birds
Gallus
Reptiles
Lizard
Snake (Vipera)
Chelonia (Pseudemys)
Amphibians
Anurans (Xenopus)
(Rand)
Urodeles (Pleurodeles, Ambystoma)
Teleostei
(Salmo, Carassius)
Selachians
(Scylliorhinus, Torpedo)
PROTOCHORDATES
Amphioxus
Ascidia (Ciona)
INVERTEBRATES
Echinodermata
Asterias (Marthasterias)
Nemertea
(Lineus)
Platyhelminthes
Planaria (Dugesia)
Annelida
Nereis
Hinido
Mollusca
Lamellibranchia {Pecten)
Gastropodia (Helix)
Cephalopodia (Loligo)
Arthropoda
Insects (Sarcophaga, Aeschna)
Arachnida (Scodra)
S8D8
S7D6
S2D2
S6H8
AS
NT
NT
NT: not tested.
clonal antibodies; in all species in which at least one
monoclonal antibody was positive, there was staining
with the polyclonal rat serum; the other species were
negative. However, the staining of photoreceptors was
less intense and the background staining was heavier
with the polyclonal serum. The unrelated monoclonal
antibodies used as controls did not give any staining
of photoreceptor cells.
Discussion. Our investigation shows that in addition
to the strong organ specificity and autoantigenicity
characteristic of the retinal S-antigen, there exists a
large interspecies cross-reactivity. Such properties are
shared by only a limited number of proteins in the
body and allow an immunologic approach to the
taxonomic distribution of the epitopes of homologous
proteins using monoclonal antibodies. Compared to
the study of protein homology by amino acid sequence
Downloaded From: http://iovs.arvojournals.org/ on 06/16/2017
analysis, the use of monoclonal antibodies presents a
major difference: they may detect conformational
epitopes; ie, molecular arrangements not necessarily
linked to an amino acid sequence but consisting of
the juxtaposition of amino acids, carbohydrates, or
both, which are brought together by the folding of
the protein, but which are not due to their proximity
in the sequence. Such conformational epitopes may
be closely related to the function of the molecule.
Their identification by monoclonal antibodies could
be the basis for a simple method suitable for analyzing
phylogenetic relationships between species.
The five monoclonal antibodies used in this study
can be considered highly specific for S-antigen. They
do not react with other proteins in extracts of the
retina or other tissues in ELISA, and they only stain
photoreceptor cells in sections of vertebrate eyes in
1020
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / July 1985
immunofluorescence.8 The rat hyperimmune serum,
though directed at purified S-antigen,3 is probably
less specific and gave a higher background on sections.
The interspecies distribution of the S-antigen epitopes recognized by the five monoclonal antibodies
indicates that these epitopes are different, since they
are not found together in all positive species. Only
S6H8 and S2D2 were always found together in this
study. However, in a recent study performed on
human retinoblastoma, we have observed different
patterns for these two antibodies (data not shown),
suggesting that they also recognize different epitopes.
With the immune serum and four monoclonal
antibodies, we found in all vertebrates the same
localization of labeling, previously described by
Wacker et al1 in guinea pig retina using guinea pig
serum against bovine S-antigen. The pattern outlined
the entire photoreceptor cell except the nucleus. The
reason for the difference of labeling between S9E2,
which stained predominantly the perinuclear area,
and the other antibodies is not known. It is hypothesized that the epitope S9E2 is present on the native
molecule but disappears or is masked during further
processing of the molecule towards its site of action.
S-antigen epitopes were found in all vertebrates
and in some species of protochordates, nemerteans,
annelids, and molluscs. They were not detected in
other groups of invertebrates, such as arthropods.
Since only a few species were investigated in the
various invertebrate phyla, this study does not allow
a definitive conclusion on the absence of the Santigen in those groups. Therefore, we can only state
that S-antigen is present in species belonging to both
protostomia (that include annelids and molluscs) and
deuterostomia (echinoderms and all chordates). Considering the theoretical age of divergence between
protostomia and deuterostomia, the four epitopes
S8D8, S7D6, S2D2, and S6H8 should have appeared
at the early Cambrian era or even earlier, as invertebrate phyla evolved during the preCambrian era and
were well differentiated by the earliest paleozoic period
(ie, Cambrian). This indicates a high phylogenetic
stability suggesting that the structure of S-antigen has
been well suited to its unknown function, very early
in evolution.
It may be of interest to consider the possibility of
a relationship between the distribution of S-antigen
and the ciliary or rhabdomeric type of photoreceptors.''
Indeed, among the most perfect eyes, the photoreceptors of vertebrates are ciliary and positive for Santigen; whereas those of arthropods and cephalopods
are rhabdomeric and negative for S-antigen. However,
in protochordates and most invertebrates, the presence
of S-antigen epitopes are not clearly related to the
existence of ciliary structures in photoreceptors. It
Downloaded From: http://iovs.arvojournals.org/ on 06/16/2017
Vol. 26
should be pointed out that the dichotomy between
the two types of photoreceptors is not strict and the
presence of S-antigen epitopes could be linked to
some functional difference in photoreceptors rather
than anatomic differences.
Despite major variations in the architecture of
photoreceptor organs and the structure of photoreceptor cells between vertebrate and invertebrate phyla,
they are directed and adapted for a common function:
light captation, and visual function for the most
elaborated photoreceptor organs. It is, therefore, not
surprising that some common molecular mechanisms
are shared by these morphologically different structures as it is the case for the visual pigments. It would
be likely that other proteins involved in photoreceptor
function, eg, in the amplification of the light stimulus,
are common to different photoreceptor cells. The
localization and wide interspecies distribution of Santigen suggest that it could belong to such molecules.
Until now only S-antigen extracted from mammals
(man, cattle, swine, guinea pig) has been shown to
induce EAU in laboratory animals. It will be of
interest to examine the ability of the S-antigen from
more distant species to induce disease in rat or guinea
pig models to determine if the immunopathogenic
property is linked to transevolutionary antigenic determinants.
Key words: retina, retinal S-antigen, monoclonal antibodies,
interspecies cross-reactivity, evolution, immunofluorescence
Acknowledgments. The purification of S-antigen was carried out by Mrs. C. Dorey. The authors acknowledge A.
Bernadou, Y. de Kozak, J. Y. Perrot, J. Sebag, P. Krief,
and J. Soria for their helpful comments and J. Bierne,
M. L. Celerier, J. Dorchen, J. C. Lacroix, J. Reperant, J.
Taxi, G. Vernet, the Station Biologique de Roscoff, the
laboratoire Maritime de Concarneau, and the Laboratoire
Maritime de Banyuls for providing animal specimens.
From the Laboratoire d'Immunopathologie de 1'Oeil, CNRS ER
227, INSERM U 86, Universite de Paris VI, Hotel-Dieu, Paris, the
Unite INSERM U 253, Villejuif, the Equipe de Recherche Biologie
du Developpement, Universite de Paris VII, Paris, France. Presented
at the ARVO meeting, Sarasota, Florida, May 1984. Submitted for
publication: July 2, 1984. Reprint requests: Jean-Pierre Faure, MD,
PhD, Unite de Recherche d'Ophtalmologie, Hotel-Dieu, 1 Parvis
Notre-Dame, 75181 Paris 04, France.
References
1. Wacker WB, Donoso LA, Kalsow CM, Yankeelov JA, and
Organisciak DT: Experimental allergic uveitis: isolation, characterization and localization of a soluble uveitopathogenic
antigen from bovine retina. J Immunol 14:1949, 1977.
2. Dorey C and Faure JP: Isolement et caracterisation partielle
d'un antigene retinien responsable de ruveo-retinite autoimmune experimentale. Ann Immunol (Inst Pasteur) 128c:229,
1977.
1021
Reports
No. 7
3. Dorey C, Cozette J, and Faure JP: A simple and rapid method
for isolation of retinal S antigen. Ophthalmic Res 14:249,
1982.
4. Nussenblatt RB, Gery I, Kuwabara T, de Monasterio FM, and
Wacker WB: The role of the retinal S-antigen in primate
uveitis. In Immunology of the Eye, Workshop II, Helmsen RJ,
Suran A, Gery I, Nussenblatt RB, editors. Washington, DC,
Information Retrieval, 1981, pp. 49-65.
5. Faure JP and de Kozak Y: Cellular and humoral reaction to
retinal antigen; specific suppression of experimental uveoretinitis.
In Immunology of the Eye, Workshop II, Helmsen RJ, Suran
A, Gery I, and Nussenblatt RB, editors. Washington, DC,
Information Retrieval, 1981, pp. 33-48.
6. Faure JP: Autoimmunity and the retina. Curr Topics Eye Res
2:215, 1980.
7. Tuyen VV, Faure JP, Thillaye B, de Kozak Y, and Fortier B:
Antibody determination by ELISA in rats with retinal S
antigen-induced uvoretinitis. Curr Eye Res 2:7, 1982.
8. Faure JP, Mirshahi M, Dorey C, Thillaye B, de Kozak Y, and
Boucheix C: Production and specificity of monoclonal antibodies
to retinal S antigen. Curr Eye Res 3:867, 1984.
9. Boucheix C, Perrot JY, Mirshahi M, Giannoni F, Billard M,
Bernadou A, and Rosenfeld C: A new set of monoclonal
antibodies against acute lymphoblastic leukemia. Leukemia
Res, in press.
10. Soria J, Soria C, Boucheix C, Mirshahi M, Perrot JY, Bernadou
A, Samama M, and Rosenfeld C: Immunochemical differentiation of fibrinogen, fragment D or E and crosslinked fibrin
degradation products using monoclonal antibodies. In Fibrinogen: Structuren, Functional Aspects, Metabolism, Haverkate
E, Henschen A, Nieuwenhuizen W, and Straub, editors. Walter
de Gruyter and Co, 1983, pp. 227-233.
11. Eakin RM: Structure of invertebrate photoreceptors. In Handbook of Sensory Physiology, VII/1, Photochemistry of Vision,
Dartnall HJA, editor. Berlin, New York, Springer-Verlag, 1972,
pp. 625-684.
Reduction of Body Swoy by Stimuli Imoged within
o Corticol Scotomo: A Cose Study
Jane E. Raymond* and Herschel W. Leibowirz
The reduction of body sway by visual stimulation was
equally effective for stimuli imaged within a cortical scotoma
or in the mirror image position in the normal visual field.
The results are consistent with the concept of distinct visual
orientation and discrimination modes of processing visual
information, which suggests that spatial orientation functions
do not necessarily involve awareness. Invest Ophthalmol
Vis Sci 26:1021-1024, 1985
Using ablation techniques, Schneider1 demonstrated that visual discrimination and visually-guided
spatial orientation can be selectively dissociated in
the hamster. His finding that removal of either the
visual cortex or the superior colliculus disrupted
discrimination or orientation functions, respectively,
suggested the existence of two reasonably separate
and parallel visual systems. For humans, Held has
proposed the term "two modes of processing" to
differentiate object recognition, ie, "focal" vision,
from spatial orientation, ie, "ambient" vision.2 Subsequent research has suggested that the focal mode,
which addresses the question of "what," is cortically
based and well-represented in consciousness. Alternately, the ambient mode, which concerns the question of "where," is thought to be mediated reflexively
with little or no conscious concomitant. 3
Evidence for the dissociation of discrimination and
orientation functions in humans has been found in
the study of cortically blind individuals. Observations
of visual functions in patients with unilateral cortical
Downloaded From: http://iovs.arvojournals.org/ on 06/16/2017
lesions has shown that while pattern vision is probably
absent in the scotomatous field,4 some residual visual
functions (eg, flicker perception, motion perception,
or the appearance of eye movements toward unseen
objects in the blind hemifield) may remain. Although
earlier views of cortically blind patients conservatively
held that any reported residual vision resulted from
surviving cortical tissue,5 more recent evidence suggests that some visual functions persist and may be
mediated by subcortical pathways. Perenin and Jeannerod4 examined the capacity of hemianopic patients
to point to a briefly flashed light presented in their
blind field. By comparing patients with pre- and postgeniculate lesions, they were able to demonstrate that
elimination of cortical inputs only (ie, postgeniculate
lesions) did not impair pointing, whereas removal of
both cortical and subcortical pathways did. Similarly,
accurate pointing to unseen objects by cortically blind
patients was also reported by Bridgeman and Staggs.6
In a study involving patients with unilateral surgical
removal of one hemisphere, it was reported that
patients were able to make saccades towards objects
placed in their hemianopic field,7 although the accuracy of such eye movements is quite poor.8
In the studies described above, the cortically blind
observer is asked to make a conscious judgement
regarding the position of unseen objects in space.
Since residual visual functions in cortically blind
individuals are most likely to be ambient and, therefore, primarily reflexive, their existence may be more