Specific Binding of Peanut Lectin to a Class

of Retinal Photoreceptor Cells

A Species Comparison

Janet C. Blanks*!}: and Lincoln V. Johnsonf

Although lectins have been used to study surface oligosaccharides of photoreceptor cells in intact
retinas and dissociated retinal cells, the specificity of lectin binding to cones versus rods in a variety
of species has not been been examined closely. The authors previously found that application of fluorescein
isothiocyanate (FITC)-conjugated peanut agglutinin (PNA), a lectin with high affinity for galactose-
galactosamine disaccharide residues, to cryostat sections of unfixed mouse retina results in staining
that is confined to synaptic regions and a subpopulation of photoreceptor cells. To further investigate
the possibility that PNA binding is specific for cone photoreceptors, the authors extended their studies
to include the duplex retinas of fish, rabbit, monkey, and human in addition to the cone-dominant retina
of the chick. These studies show that PNA binding is specific for cone inner and outer segments and
also is likely to be associated with the large synaptic pedicles of cone photoreceptor cells. In addition,
the authors compared PNA binding with that of Ricinus communis agglutinin I (RCA), another lectin
that preferentially binds terminal D-galactose moieties. While RCA does bind to cones in the species
examined, it also binds to a lesser extent to rod photoreceptor inner segments. The pattern of binding
of RCA in other regions of the retina differs markedly from that of PNA. Significantly, RCA serves
as a specific marker for retinal vasculature in the human, monkey, and mouse. These results suggest
that certain PNA-binding macromolecules may be important in defining the molecular and cellular
specificity of cone photoreceptor cells and that PNA may provide a means for the isolation of cones
and cone-specific molecules. RCA may prove to be of value in monitoring vascular changes associated
with normal development and pathologic conditions. Invest Ophthalmol Vis Sci 25:546-557, 1984

In our previous studies of the binding of a battery may exist. In fact, 3H-fucose has been shown to be
of carbohydrate-specific lectins during mouse retinal incorporated into a glycoprotein that is a component
development,1 it was observed that peanut agglutinin of cone but not rod outer segments.2'3
(PNA) from Arachis hypogaea showed selective binding To further investigate the basis of this cellular het-
to discrete patches within the outer synaptic layer and erogeneity and its possible relationship to differences
to a subpopulation of photoreceptor cells believed to between rods and cones, we extended our PNA binding
be cones. This pattern of PNA binding suggested het- studies to include retinas from a variety of species that
erogeneity in cellular carbohydrate composition within differ in their ratios of cone and rod photoreceptor
the photoreceptor layer. The cone-specific binding of cells. In addition, we compared PNA binding with that
PNA suggests that significant differences in glycocon- of another galactose-specific lectin, Ricinus communis
jugate composition of rod and cone photoreceptor cells agglutinin I (RCA). We have employed a new embed-
ding method using polyacrylamide (modified from ref-
From the Departments of Ophthalmology* and Anatomy and erence 4; Johnson and Blanks, manuscript submitted)
Cell Biology,t University of Southern California School of Medicine, to obtain thinner sections and tissue morphology su-
and Estelle Doheny Eye Foundation,:): Los Angeles, California. perior to that obtainable with fresh frozen cryostat
Supported in part by National Institutes of Health research grants sectioning techniques.
EYO-4741 (LVJ) and EYO-3042 (JCB) and Core Center grant EYO-
3040 to the Doheny Eye Foundation. JCB and LVJ are the recipients
of Research Career Development Awards from the National Eye Materials and Methods
Institute (JCB) and the National Institute for Child Health and Hu-
man Development (LVJ). Lectins and Hapten Sugars
Submitted for publication: September 2, 1982.
Reprint requests: Janet C. Blanks, PhD, Estelle Doheny Eye Foun- Fluorescein-isothiocyanate conjugated (FITC) pea-
dation, 1355 San Pablo Street, Los Angeles, CA 90033. nut agglutinin (PNA) and Ricinus communis agglutinin

546

Downloaded from iovs.arvojournals.org on 10/12/2020

 No. 5 FITC-PNA AND RCA BINDING TO PHOTORECEPTOR CELLS / Blanks and Johnson 547

I (RCA) were obtained from E. Y. Laboratories (San Retinas were rinsed overnight in PBS and then in-
Mateo, CA) and Vector Laboratories (Burlingame, cubated in the presence of 200 Mg/ml FITC-PNA in
CA). The hapten sugar D (H-)-galactose was obtained PBS containing 1 mg/ml BSA for 30 min at room
from Sigma (St. Louis, MO). Hapten-specific inhibition temperature. The whole retina then wasrinsedin three
of PNA or RCA binding was evaluated by preincu- changes of PBS over a 90-min period. Four to six small
bation of the FITC-lectin (100 ng/m\) in 10 mM so- peripheral incisions were made to allow the retina to
dium phosphate buffered 0.85% saline (PBS, pH 7.2- be flat-mounted on a microscope slide. The flattened
7.4) containing 25-50 mM D-galactose. retina was covered with a mounting medium of PBS
containing 1 mg/ml BSA and 1 mg/ml p-phenylene-
Histochemical Procedures diamine. A coverslip wasfloatedon the mounting me-
dium and medium was slowly withdrawn so the cov-
Experimental animals were treated in conformity erslip contacted but did not crush the retina. Obser-
with the ARVO Resolution on the Use of Animals in vations were made by epifluorescence microscopy using
Research. They were killed by cervical dislocation an Olympus BH-S microscope (New Hyde Park, NY)
(mice, C57BL/6J), decapitation (fish, Carassius au- equipped with narrow band blue excitation filters.
ratus; chickens, Gallus gallus), or an overdose of phe-
nobarbital (rabbits, Oryctolagus cuniculus; monkeys,
Macaca fasicularis). A total of 50 mouse, 20 chicken, Results
10 rabbit, 7 fish, 4 monkey, and 3 human eyes were In all retinas, except that of the cone-dominant chick
examined in this study. The eyes were enucleated, cor- eye, heterogeneity of PNA and RCA binding within
neas removed and the eyecups immersed in 4% para- the photoreceptor layers was observed. Lectin binding
formaldehyde in PBS for 1 hr (unless noted otherwise). was not detectable in the nuclear layers of the retina
The human eyes were fixed longer than the other tis- but, rather, was specific for the synaptic layers and the
sues. One eye wasfixed3 days in 2% paraformaldehyde inner segment (IS) and outer segment (OS) regions of
and 2% glutaraldehyde in 0.1 M phosphate buffer; the certain photoreceptor cells. More detailed descriptions
other two were fixed in 4% paraformaldehyde for 24 of FITC-PNA and FITC-RCA binding patterns in the
hr. One of the human eyes came from an enucleation individual species examined are as follows:
procedure for anterior segment melanoma and the
other two eyes came from eye bank donors; the retinas
Human
did not appear pathologic. After fixation, the eyecups
were rinsed for several hours in PBS and embedded In the paramacular region of the human eyes, the
in polyacrylamide according to our modification of cone inner segments, and, to some extent, the outer
the procedure of Hausen and Dreyer4 (Johnson and segments showed PNA binding (Fig. 1A). Rods, ap-
Blanks, manuscript submitted). parent as negative images (r, Fig. 1A), did not show
PNA binding. PNA binding to the outer synaptic layer
Staining Procedure (OSL) was not observed but lectin binding in OSL
typically is not detectable in tissues subjected to pro-
The staining procedure with FITC-conjugated lectins
longed fixation, such as the human tissue had been.
was based on that of Hatten et al.5 Staining solutions Tangential sections of the human retina in the pho-
were prepared by dilution of lectin to 100-200 Mg/ml toreceptor layer showed rings of fluorescence corre-
with PBS containing 500 Mg/ml bovine serum albumin sponding to cone inner segments cut in cross-section
(PBS-BSA) followed by centrifugation. In brief, sections (arrowheads, Fig. IB). In some areas the negative im-
were preincubated for 10 min in PBS-BSA followed ages of rods were detected (r, Fig. IB).
by application of 25-50 ^1 of lectin solution and in-
cubation in a humid chamber for 15-30 min. Sections FITC-RCA binding to human retina showed a dif-
were then rinsed in PBS and mounted in PBS-BSA ferent pattern of binding compared with that of PNA.
containing 1 mg/ml p-phenylenediamine to inhibit Photoreceptor outer segments of both rods and cones
quenching of the fluorochrome.6 Photographs were were fluorescent (Fig. 1C), but it also was possible to
identify the large inner segments of the cones (single
made on a Zeiss Photomicroscope III (New York, NY)
arrowheads, Fig. 1C). A band of fluorescence repre-
equipped for epifluorescence.
senting the outer limiting membrane (OLM) was ev-
ident. Spots of intensefluorescencewere visible in the
Whole Mounts IS-OS junctional region (double arrowheads, Fig. 1C).
Whole-mount preparations of rabbit retina were ob- Patchy areas in the OSL and the entire ILM were
tained by dissection of the retina from eyecups fixed fluorescent. Retinal blood vessels (BL) were intensely
for 2 hr in 4% paraformaldehyde as described above. fluorescent.

Downloaded from iovs.arvojournals.org on 10/12/2020

 548 INVESTIGATIVE OPHTHALMOLOGY 6 VISUAL SCIENCE / Moy 1984 Vol. 25

OLM
OSL

ISL

ILM

Fig. 1. Light micrograph showing retinal binding of FITC-PNA

(A) and (B) and FITC-RCA (C) in human. A, A retinal section from
the paramacular region of the human retina exposed to FITC-PNA
after prolonged fixation in 2% glutaraldehyde and 1% paraformal-
dehyde, simultaneously illuminated for fluorescence and phase-con-
trast. The cone photoreceptors (c) are fluorescent while the negative
images of nonfluorescent rod photoreceptors <r) are apparent. The
cone IS {single arrowheads) and the cone OS (double arrowheads)
are visible directly above the OLM. The OSL is virtually unlabeled,
while the ISL and ILM are faintly fluorescent (X360). B, A tangential
section through the OS and IS region of a human retina incubated
in FITC-PNA. Profiles, believed to be cone (c) IS because of their
size, show either a fluorescent ring, suggesting PNA binding to cone
plasma membrane or extracellular matrix. The negative images of
rod (r) photoreceptors are sometimes visible and virtually unstained
in comparison to the cones (X375). C, Fluorescence micrograph of
the paramacular region after incubation with FITC-RCA shows
patchy fluorescence at the OS-IS junction (double arrowheads). Hazy
fluorescence is visible in the OS region. The IS of rods scarcely bind
FITC-RCA, but the IS of cones (C) are outlined. Patchy fluorescence
is present in the OSL. Blood vessels (BL) are intensely fluorescent.
The ILM also is fluorescent (X370).

Monkey eyes or fixed by immersion for 1 hr in 4% paraform-

aldehyde after enucleation. Application of FITC-PNA
These animals were perfused with the same dialde- resulted in intensefluorescenceonly in the IS and OS
hyde fixative described above for one of the human of cones (Figs. 2A-C). Often the pattern of PNA bind-

Downloaded from iovs.arvojournals.org on 10/12/2020

 No. 5 FITC-PNA AND RCA BINDING TO PHOTORECEPTOR CELLS / Blanks and Johnson 549

ing in the OSL appeared patchy (Fig. 2A). The inner In contrast, FITC-RCA stained the region of the
synaptic layer (ISL) was barely fluorescent with PNA. OLM and photoreceptor ISs and the ILM, in addition
The pattern of binding of FITC-RCA was quite dif- to a diffuse staining in the OSL (Fig. 6B). As reported
ferent in the monkey retina (Fig. 3) since both rod previously,1 retinal blood vessels stained intensely
and cone photoreceptor outer segments were lightly with RCA.
stained with this galactose-specific lectin. A higher
magnification of the photoreceptor region (Fig. 3C) Chicken
shows that the staining pattern with FITC-RCA ap-
peared to "cap" the region of the cone ellipsoid. Diffuse The chicken retina showed the most intense staining
fluorescence is visible on the rod outer segments, being in the photoreceptor layer with FITC-PNA (Fig. 7A);
so intense at the region dividing the outer and inner virtually all photoreceptor cells (both IS and OS) ap-
segments that spots of fluorescence were seen (double peared to bind this lectin. Faint staining was observed
arrowheads, Fig. 3C). Fluorescence also was observed in the OSL and several bands with varying fluorescent
in the OSL with FITC-RCA (Fig. 3C). Blood vessels intensities were present in the ISL (arrowheads, Fig.
(BL, Fig. 3A) were intensely stained with RCA. 7A). These bands were not seen in the other animals,
except when FITC-PNA was injected intraocularly in
Rabbit the mouse (unpublished observations). The nerve fiber
layer (NFL, Fig. 7) also was intensely fluorescent.
Intense binding of PNA was observed in the cone FITC-RCA stained fewer photoreceptor outer seg-
IS (single arrowhead) and OS (double arrowhead) re- ments (double arrowheads, Fig. 7C) and inner segments
gions, in addition to the OSL (Fig. 4A). There also (single arrowheads, Fig. 7C) in addition to the OSL
appeared to be slight binding of FITC-PNA to the cone (Fig. 7C). The staining in the OSL appeared beaded.
cell bodies and their axonal extensions (curved arrow, The NFL also showed binding to FITC-RCA.
Fig. 4A), leading to the stained patches in the OSL As a control for the specificity of lectin binding,
(arrowheads in OSL, Fig. 4A). sections of adult chick retina were exposed to FITC-
In contrast, FITC-RCA intensely stained the inner PNA or FITC-RCA preincubated with 50 raM D-ga-
segments of apparently all photoreceptor cells in the lactose. Incubation in the presence of this hapten sugar
rabbit retina (IS, Fig. 4B). Cone—but apparently not dramatically decreased PNA (Fig. 7B) and RCA (Fig.
rod—outer segments also were labeled with FITC-RCA 7D) binding. Preincubation with the nonhapten sugar
(double arrowheads, Fig. 4B). The OSL, ISL, and the glucose had no effect on the pattern of fluorescent
inner limiting membrane (ILM) also were intensely staining by either lectin (data not shown). Similar con-
stained with FITC-RCA. The regions surrounding the trols were performed on the other tissues used in this
nuclei of the photoreceptor cells stained with FITC- study; similar results were obtained (data not shown).
RCA (ONL, Fig. 4B).
In an attempt to further localize FITC-PNA binding, Fish
a whole mount of the rabbit retina was prepared (Fig.
5). Since the retina was dissected away from the retinal In the fish, PNA showed binding to cone inner and
pigment epithelium, it was possible to see the tips of outer segments (Fig. 8A). Higher magnification shows
the outer segments—a view that was not possible in intense PNA binding to the region connecting the OS
sectioned tissue. The tips of the cone outer segments and IS (insert, Fig. 8 A). The rest of the retina was only
were intensely fluorescent, as was the plasma mem- faintly fluorescent. This species posed the largest prob-
brane of the cone inner and outer segments. The optic lem in obtaining optimum fixation for ease of cryostat
disk area also stained with FITC-PNA (arrow, Fig. 5). sectioning and consistent lectin binding results.
The negative images of unlabeled rod photoreceptor FITC-RCA showed intense binding in the OSL, ISL,
cells (r, insert, Fig. 5) are adjacent to fluorescently and ILM (Fig. 8B). The rest of the retina was only
labeled cones (c, insert, Fig. 5). faintly fluorescent.

Mouse Discussion
PNA staining again was observed to be selective for This paper details a comparative study of peanut
a subpopulation of photoreceptor cells. PNA binding agglutinin and Ricinus communis agglutinin, two lec-
was intense in cone IS and OS (Fig. 6A); punctate tins with binding affinities for galactose-containing oli-
staining was observed in the OSL. The ISL displayed gosaccharides, to retinal sections from a variety of ver-
uniform diffuse staining. tebrate species. In general, the method involves fixing

Downloaded from iovs.arvojournals.org on 10/12/2020

 550 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / May 1984 Vol. 25

Fig. 2. Light micrographs showing retinal binding of FITC-PNA in monkey. A, Fluorescence micrograph of the macular region with intensely
fluorescent cone OS and cone IS (C). Punctate fluorescence in OSL may be due to cone pedicles. Faint fluorescence is present in the ISL
(X375). B, Phase micrograph of same section as in A. Cone IS (C); PE, pigment epithelium; OLM, outer limiting membrane; ONL, outer
nuclear layer; INL, inner nuclear layer; GCL, ganglion cell layer (X375). C, A retinal section from the paramacular region of a monkey retina
simultaneously illuminated for fluorescence and phase-contrast. The cone IS (C, single arrowhead) are visible directly above the OLM. Several
cone OS are evident (double arrowheads). The negative images of rods (R) are visible between the intensely fluorescent cones (X500).

Downloaded from iovs.arvojournals.org on 10/12/2020

 No. 5 FITC-PNA AND RCA DINDING TO PHOTORECEPTOR CELLS / DIonks ond Johnson 551

Fig. 3. Binding of FITC-RCA to peripheral region of monkey retina. A, Fluorescence micrograph showing cone IS and OS stained by FITC-
RCA. Intensely fluorescent regions are evident at what appears to be the rod IS-OS junction. Patchy fluorescence in OSL may be due to cone
pedicles. Blood vessels (BL) stain intensely with RCA (X430). B, Phase contrast micrograph of same area of monkey retina shown in A. Cone
IS (arrowhead, C) can be clearly differentiated from adjacent rod photoreceptor cells. Arrows depict the same blood vessels seen in A (X430).
C, Higher magnification of outer retina showing FITC-RCA staining of cone IS (C). Rod OS are more fluorescent than the IS, Intense
fluorescent areas {double arrowheads) are present at the OS-IS junction of the rods. Intense, somewhat patchy fluorescence is present in the
OSL (X740).

Downloaded from iovs.arvojournals.org on 10/12/2020

 552 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / May 1984 Vol. 25

Fig. 4. Fluorescence micrographs showing binding of F1TC-FNA (A) and Fl 1 (J-KCJA (B) to rabbit retina. A, Cone US (double arrowheads),
cone IS (single arrowheads) and patches within the OSL (arrowheads) stain intensely with FITC-PNA. The cell bodies of the cones are lightly
fluorescent as are, in some cases, their narrow axonal extensions (curved arrow) to the OSL (X600). B, FITC-RCA diffusely stains the region
of the OLM and photoreceptor IS. Cone IS and OS (double arrowheads) also are stained with RCA, as are the ILM, ISL, and OSL. In addition,
the paranuclear regions of photoreceptor cells in the ONL are stained by RCA (X475).

Downloaded from iovs.arvojournals.org on 10/12/2020

 No. 5 FITC-PNA AND RCA DINDING TO PHOTORECEPTOft CELLS / Blanks and Johnson 553

Fig. 5. Whole-mount preparation of rabbit retina stained with FITC-PNA, The tips of cone OS are intensely stained as well as the rest of
the cone OS membrane. The cones appear to radiate from the optic disc (arrow) (X210). Insert: Tangential view of whole mount showing
the photoreceptors with rods (r) unstained while the plasma membranes of cones appear asfluorescentrings(c), and the fluorescent tips of
the cones as bright dots (X425).

tissue in 4% paraformaldehyde for a brief time, axonal extensions leading to the OSL were lightly flu-
embedding in acrylamide, followed by freezing and orescent (Fig. 4A). In addition, PNA binding in the
cryostat sectioning, and,finally,application of the lectin inner synaptic layer of the chick retina exhibits a strat-
to the tissue sections. The tissue is fixed prior to ap- ified pattern. To a lesser extent, the retinal vasculature
plication of the lectin to preserve tissue morphology and the inner limiting membrane also show PNA
and to prevent lectin-induced rearrangement of binding binding specificity.
sites. In addition, this treatment allows lectins access Our results with FITC-PNA do not agree with the
to intracellular and extracellular components; the ob- binding in isolated photoreceptor preparations in frog.
served staining patterns, thus, are likely to be the result Bridges and Fong7 found that FITC-PNA bound to
of binding to cytoplasmic as well as cell membrane- the OS and not to the IS of freshly isolated ROS prep-
associated and extracellular matrix materials. The re- arations, some of which had the IS attached, and
sults obtained confirm and extend our previous ob- Bridges8 reported similar results using isolated prep-
servations on the binding of these two lectins to frozen, arations of frog retinas. The latter paper also dem-
unfixed mouse retina.1 onstrates FITC-PNA binding to accessory cones in frog.
PNA is shown to bind specifically to photoreceptor It seems probable that these discrepancies are due to
cells morphologically identifiable as cones in a variety species differences. It is interesting to note that PNA
of species, including human, monkey, rabbit, mouse, has been shown to bind preferentially to photoreceptor
chicken, andfish.In these species, PNA binds intensely cell precursors in the eye disc of ommatidia in the
to cone inner and outer segments and often shows a Drosophila eye.9
patchy distribution of binding in the outer synaptic Although the binding of RCA is similar to PNA
layer. When present, the size of the PNA-binding binding in retina, it is not identical. The major dif-
patches in the OSL is comparable with that of cone ference is that PNA binds cones exclusively in all spe-
pedicles. The staining in the OSL appears to be espe- cies studied while RCA binds rods to some extent in
cially sensitive to prolonged fixation. In some cases, addition to cones. In the primates studied, FITC-RCA
the cone photoreceptor cell bodies and their narrow showed patchy fluorescence of the outer segments,

Downloaded from iovs.arvojournals.org on 10/12/2020

 554 INVESTIGATIVE OPHTHALMOLOGY b VISUAL SCIENCE / May 1984 Vol. 25

Fig. 6. Fluorescence micrographs of mouse retina showing binding of FITC-PNA (A) and FITC-RCA (B). A, Intense fluorescence of selected
IS (single arrowhead) and OS (double arrowheads). There is a punctate pattern of heavy fluorescence in the OSL, while bright fluorescence
is apparent in the ISL. (X510). B( The OLM-IS region and the OSL are barely fluorescent. Blood vessels (BL) are brightly fluorescent. The
ILM is also fluorescent (X55O).

being more intense at the region connecting outer and 6B). In chick, FITC-RCA binding produced a beaded
inner segments and much weaker in the inner seg- appearance offluorescencein the OSL and appeared
ments. These fluorescent patches were reported first to stain some photoreceptors (Fig. 7B). However, their
by Uehara et al10 on sections offixedparaffin embedded identification remains unclear. And finally, in the fish
monkey retina. It is possible that these patches (see retina FITC-RCA did not bind to either the OS or IS
Figs. 1C and 3C) correspond to the region of the con- region but only to the synaptic layers and the ILM
necting cilium in rods. (Fig. 8B). We conclude that there is considerable species
Our results in human and monkey with RCA con- variation in RCA binding to retina.
firm earlier studies using ferritin-RCA by Nir and Our results of RCA binding to blood vessels confirms
Hall," Bridges,8 Bridges and Fong7 in the frog, and our earlier observation in the mouse retina.1 Although
by Uehara et al10 in monkey in which it was shown this study was confined to the retina, Raedler et al13
that RCA binds preferentially to rod outer segment have reported that RCA (referred to as RCL in their
(ROS) plasma membranes. McLaughlin and Wood'2 study) binds to the vasculature of the central nervous
also showed that horseradish peroxidase conjugated system in general. Other researchers14 have used an-
RCA binds to ROS in the rat. In rabbit, RCA appeared other lectin, Ulex europaeus (UEA), as a marker for
to bind to the OLM, the IS region and to a few pho- vascular endothelium in human tissues such as kidney,
toreceptor outer segments believed to be cones (Fig. liver, pancreas, and cerebellum.
4B). In mouse, the OLM and to a lesser extent the IS The differences in photoreceptor binding of these
region were faintly fluorescent with FITC-RCA (Fig. two galactose-binding lectins probably is best explained

Downloaded from iovs.arvojournals.org on 10/12/2020

 No. 5 FITC-PNA AND RCA BINDING TO PHOTORECEPTOR CELLS / Blanks and Johnson 555

Fig. 7. Fluorescent micrographs of chicken retina showing binding by FITC-PNA (A), FITC-RCA (C), and controls for PNA binding (B),
and RCA binding (D). A, Chicken retina fixed by cardiac perfusion with 4% paraformaldehyde. FITC-PNA binds intensely to cone IS (single
arrowhead) and to OS (double arrowhead) presumed to be cone OS. Cardiac perfusion often decreases OSL staining. A banding of fluorescence
is visible in the ISL (arrowheads, ISL). Ganglion cell axons in the NFL are intensely fluorescent (X500). B, Control for PNA binding: Section
of chick retina adjacent to that in A incubated in the presence of 50 mM D-galactose shows reduced levels of binding of PNA (X510). C, IS
(single arrowheads) and OS (double arrowheads) thought to belong to cones are intensely fluorescent with FITC-RCA. Patchy fluorescence is
present in the OSL. The NFL is also fluorescent (X540). D, Control for RCA binding: Section of chick retina adjacent to that in C incubated
in the presence of 50 mM D-galactose shows reduced levels of binding of RCA (X490).

by the fact that PNA has its highest affinity for the residues.16 Our observation that PNA binds to cones
disaccharide sequence D-Gal-/3(1 —» 3)D-Gal NAc, 15 preferentially, while RCA binds to some extent to both
while RCA has a specificity for /^-linked D-galactose rods and cones suggests that oligosaccharides present

Downloaded from iovs.arvojournals.org on 10/12/2020

 556 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / Moy 1984 Vol. 25

Fig. 8. Binding of FITC-PNA (A) and FITC-RCA

(B) to a goldfish retina fixed by cardiac perfusion with
4% paraformaldehyde. A, Fluorescent micrograph
showing connecting region of cone OS intensely flu-
orescent due to FITC-PNA binding; lighter binding
of ellipsoid region of the IS (X480). Insert: There ap-
pears to be a ring of fluorescence (arrows) separating
the cone OS and IS (X 1,010). B, Fluorescent micro-
graph showing intense fluorescence of the OSL, ISL
and ILM with FITC-RCA. There appears to be more
fluorescence in the region of the ISL directly above
the GCL layer (X430).

on rod photoreceptor IS and OS lack the disaccharide The preferential binding of FITC-PNA to cones in
linkage noted above or the linkage is inaccessible to a variety of species suggests a major biochemical dif-
lectins in sectioned material. ference between rods and cones that may be related

Downloaded from iovs.arvojournals.org on 10/12/2020

 No. 5 FITC-PNA AND RCA BINDING TO PHOTORECEPTOR CELLS / Blanks and Johnson 557

to their different visual pigments. The marked ap- 5. Hatten ME, Schachner M, and Sidman RL: Histochemical char-
pearance of PNA binding in the species presented here acterization of lectin binding in mouse cerebellum. Neuroscience
4:921, 1979.
suggests that it may provide a means of separating 6. Johnson GD and Nogueira Araujo GM: A simple method of
cone from rod photoreceptors. Electron microscopic reducing the fading of immunofluorescence during microscopy.
investigations are underway to further characterize J Immunol Methods 43:349, 1981.
PNA-positive photoreceptor cells and specific regions 7. Bridges CDB and Fong SL: Different receptors for distribution
of the synaptic layers, as are biochemical character- of peanut and ricin agglutinins between inner and outer segments
of rod cells. Nature 282:513, 1979.
izations of PNA-binding molecules in the retina.
8. Bridges CDB: Lectin receptors of rods and cones. Visualization
by fluorescent label. Invest Ophthalmol Vis Sci 20:8, 1981.
Key words: peanut agglutinin (PNA), Ricinus communis ag- 9. Fristrom DK and Fristrom JW: Cell surface binding sites for
glutinin I (RCA), retina, photoreceptors, cones peanut agglutinin in the differentiating eye disc of Drosophila.
Dev Biol 92:418, 1982.
10. Uehara F, Sameshima M, Muramatsu T, and Ohba N: Local-
Acknowledgments ization of fluorescence-labeled lectin binding sites on photore-
The authors wish to thank Ms. Maria Estevez, Mrs. Joyce ceptor cells of the monkey retina. Exp Eye Res 36:113, 1983.
Tink, Mr. Robert Zink, and Mrs. Lois Samuels for their 11. Nir I and Hall MO: Ultrastructural localization of lectin binding
technical assistance. Miss A. Dawson gave helpful suggestions sites on the surface of retinal photoreceptors and pigment epi-
in the preparation of the manuscript. The animals employed thelium. Exp Eye Res 29:181, 1979.
in these studies were maintained in facilities fully accredited 12. McLaughlin BJ and Wood JG: The localization of lectin binding
by the American Association for Laboratory Animal Science. sites on photoreceptor outer segments and pigment epithelium
of dystrophic retinas. Invest Ophthalmol Vis Sci 19:728, 1980.
13. Raedler E, Raedler A, and Feldhaus S: Varying expressions of
References lectin receptors within embryonic cell layers of murine cerebral
cortex. Anat Embryol 162:21, 1981.
1. Blanks JC and Johnson LV: Selective lectin binding of the de- 14. Holthofer H, Virtanen I, Kariniemi A-L, Hormia M, Linder E,
veloping mouse retina. J Comp Neurol, 221:31, 1983. and Miettinen A: Ulex europaeus I lectin as a marker for vascular
2. Bunt AH: Fine structure and radioautography of rabbit pho- endothelium in human tissues. Lab Invest 47:60, 1982.
toreceptor cells. Invest Ophthalmol Vis Sci 17:90, 1978. 15. Lotan R, Skutelsky E, Danon D, and Sharon N: The purification,
3. Bunt AH and Klock IB: Comparative study of 3H-fucose in- composition, and specificity of the anti-T lectin from peanut
corporation into vertebrate photoreceptor outer segments. Vision (Arachis hypogaea). J Biol Chem 250:8518, 1975.
Res 20:739, 1980. 16. Debray H, Decout D, Strecker G, Spik G, and Montreuil J:
4. Hausen P and Dreyer C: The use of polyacrylamide as an Specificity of twelve lectins towards oligosaccharides and gly-
embedding medium for immunohistochemical studies of em- copeptides related to N-glycosylproteins. Eur J Biochem 117:41,
bryonic tissues. Stain Technol 56:287, 1981. 1981.

Downloaded from iovs.arvojournals.org on 10/12/2020