Download Conjunctival and episcleral blood vessels are permeable to

ConjunctivQl and Episcleral Blood Vessels ore
Permeable to Blood-borne Horseradish Peroxidase
Giuseppino Raviola
X
A
JL
The vessels of the conjunctival and episcleral plexuses of Macaco mulatto eye are of the continuous
type. Most of the vessels in the conjunctival plexus have the diameter of capillaries, while the vast
majority of the vessels in the episcleral plexus are venules. Both types of vessels have a simple wall,
which consists of an endothelium and a discontinuous layer of pericytes. The aim of this study was
to establish their permeability properties to blood-borne horseradish peroxidase (HRP). After intravenous injection of HRP, in 2\>l* jum chopper sections of the anterior segment of the eye examined
with the light microscope, the subconjunctival and episcleral tissues appear intensely and diffusely
stained by the reaction product. The electron microscope shows that HRP escapes from the vessels
lumen by crossing the interendothelial clefts and, in addition, a great number of pinocytotic vesicles
loaded with HRP are present on the luminal, tissue front and in the cytoplasm of the endothelial
cells. HRP, which rapidly penetrates the loose connective tissue of the region, reaches the spaces
between the cells of the conjunctival epithelium where it is finally blocked by the zonulae occludentes
that connect the most superficial epithelial cells. A slow diffusion into the compact tissue of the cornea
and of the sclera was also observed. Thus, under normal conditions, blood-borne macromolecules can
freely diffuse into the subconjunctival and episcleral loose connective tissues. On the other hand, one
can equally expect that the aqueous humor that reaches the episcleral and conjunctival blood plexuses
through the canal of Schlemm and collector channels can freely diffuse into the subconjunctival spaces
across the walls of these permeable vessels. Invest Ophthalmol Vis Sci 24:725-736, 1983
structure and permeability properties of conjunctival
and episcleral vessels in the rhesus monkey. It shows
that both conjunctival and episcleral blood vessels
possess a continuous endothelium but that they are
highly permeable to intravenously injected HRP.
The bulbar conjunctival and episcleral plexuses of
blood vessels are vascular networks lying on the surface of the most anterior part of the sclera, next to
the limbus. These plexuses, which anastomose with
each other, receive most of the aqueous humor leaving the anterior chamber and, therefore, are of special
importance for the maintenance of the normal intraocular pressure. In spite of such a crucial physiologic
role, very few studies exist on the fine structure of
these vessels and these are limited to the conjunctival
capillaries of the human limbus1 and fornix.2 Furthermore, the permeability properties of both conjunctival and episcleral vessels after intravenous injection of a physiologic electron opaque tracer such
as horseradish peroxidase (HRP) are still unknown.
The aim of this study was an analysis of the fine
u
V
Materials and Methods
Morphology of the Conjunctival and
Episcleral Vessels
Five adult Macaca mulatta monkeys of either sexes
were used. In four monkeys under general anesthesia
the eyes were enucleated, cut at the equator, and immersed for 2 hrs in a fixative fluid consisting of a
mixture of glutaraldehyde and formaldehyde in cacodylate buffer.3 In the fifth animal, 2 liters of fixative
fluid were perfused through the heart at a pressure
of about 100 mmHg. At the end of the perfusion the
head, with the eyes in situ, was left overnight in fixative at 4 C. With this procedure mechanical injuries
to the periocular structures before fixation were
avoided. Furthermore, in contrast with blood vessels,
which had an empty lumen, lymphatic vessels contained coagulated lymph and therefore could be identified easily with both the light and electron microscopes. During enucleation of the eyeball, great care
From the Department of Anatomy, Boston University School
of Medicine, Boston, Massachusetts.
Supported by USPHS Grant EY 01349. This study was conducted in part at the New England Regional Primate Research
Center, Southborough, MA, which is supported by NIH Grant RR00168 from the Division of Research Resources.
Submitted for publication April 23, 1982.
Reprint requests: Giuseppina Raviola, MD, PhD, Department
of Anatomy, Boston University School of Medicine, 80 East Concord Street, Boston, MA 02118.
0146-0404/83/0600/725/$ 1.40 © Association for Research in Vision and Ophthalmology
725
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933342/ on 05/09/2017
726
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / June 1983
was taken to preserve intact the bulbar conjunctiva.
The anterior segment of the eye was radially cut into
wedges, washed overnight in buffer, and postfixed for
2 hrs at 4 C with 1% osmium tetroxide and 1.5%
potassium ferrocyanide in distilled water.4 After dehydration with ethanol, specimens were embedded
flat in an Epon-Araldite mixture. For light microscopy, sections 1-2 ^m in thickness were stained with
toluidine blue. For electron microscopy, thin silver
sections were stained with uranyl acetate and lead
citrate and examined either in a RCA-3G or a 100
CX Jeol electron microscope.
Experiments with HRP
Three adult Macaca mulatta were anesthetized and
injected through the small saphenous vein with HRP
(Sigma, Type II, 500 mg/kg body weight) dissolved
in 5 ml phosphate buffered saline. Two different procedures were used to fix the eyeball: (1) in two monkeys the eyes were enucleated, 5, 10, 15, and 20 min
after the beginning of the injection of the tracer and
immediately immersed in thefixativefluid;(2) in the
third animal, fixation of the tissue was performed
with perfusion of the fixative fluid from the heart 10
min after the injection of HRP.
In both cases the fixative fluid was the mixture of
glutaraldehyde and paraformaldehyde previously reported for conventional electron microscopy. Subsequently, wedges of the anterior segment of the eye
were sectioned radially with a Smith-Farqhuar tissue
chopper, and sections 200 ^m in thickness were reacted for the histochemical demonstration of HRP.5
Part of these sections were dehydrated in alcohol,
cleared in xylenes, and mounted with Permount for
light microscope examination. The remaining chopper sections were osmicated, dehydrated, and embedded as previously reported for conventional electron
microscopy.
Results
Morphology of the Conjunctival and
Episcleral Vessels
Light microscopy: In radial sections of the anterior
segment of the eye, 1-2 /urn in thickness, the region
between bulbar conjunctiva and sclera has the shape
of a wedge with the apex corresponding to the corneoscleral limbus. With the light microscope it is possible to recognize, immediately beneath the conjunctival epithelium, a band of loose connective tissue,
the subconjunctival layer. Deep to it, is a layer of
dense connective tissue (Fig. 1) that corresponds to
both the attachment of the Tenon's capsule and to
the episcleral tissue. Here, it will be simply referred
Vol. 24
to as episcleral layer. Finally, deep to this layer the
sclera is seen, easily identified by the compact texture
of its connective tissue fibers.
In the subconjunctival layer both blood and lymphatic vessels are found. As a rule, the subconjunctival blood vessels are small and are particularly numerous in the region near the limbus, where they give
rise to the limbal or pericheratic plexus. They often
contact the basement membrane of the conjunctival
epithelium, and their wall consists of an endothelium
with a few surrounding pericytes. In the vicinity of
these vessels, the connective tissue fibers maintain
their irregular course and do not give rise to a distinct
adventitial layer. In sections of specimens fixed by
perfusion, numerous lymphatic vessels are seen under
the conjunctival epithelium along the entire extent
of the bulbar conjunctiva. The lymphatic vessels have
an irregular lumen and a very attenuated endothelium; the nuclei of the endothelial cells are spaced at
a greater distance from one another than in the lining
of blood vessels. Lymphatic and blood vessels often
occur very close to one another (Fig. 2).
In the underlying, dense episcleral layer, blood vessels are less numerous than in the subconjunctival
layer and their diameters are much larger. In spite of
their size, the majority of these vessels have a very
simple wall, consisting of an endothelium and a discontinous layer of pericytes; they can be easily classified as venules. Only a small proportion of the vessels of the episcleral plexus belong to the class of arterioles. Venules and arterioles can be found in close
association with one another.
Within the sclera, blood vessels are rare and have
a large diameter. Collector channels can be recognized easily when originating from Schlemm's canal.
Their endothelium is very attenuated, contains sparse
nuclei, and is intimately adherent to the surrounding,
dense connective tissue. In this respect, the morphology of their walls is very similar to that of the
scleral wall of Schlemm's canal.
Electron microscopy: The small conjunctival vessels and the venules of the episcleral plexus have identical morphology and therefore will be described together. Their wall consists of a continuous endothelium and a discontinuous layer of pericytes (Fig. 3,
4). The endothelial cells are irregular in shape, being
provided with blunt luminal protrusions and slender
basal leaflets or digitations; in this respect they are
similar to the endothelial cells of the iridial vessels
in the same animal species.6 Adjacent endothelial
cells often overlap extensively and interendothelial
clefts follow an unusually tortuous, irregular path
(Fig. 4). In most cases the interendothelial clefts are
open throughout their length (Fig. 4); only occasionally the adjacent plasma membranes approach each
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933342/ on 05/09/2017
>4,
L
No. 6
CONJUNCTIVAL AND EPISCLERAL BLOOD VESSELS / fXoviolo
727
Fig. 1. Light micrograph of the conjunctiva and episclera in an animal perfused with the fixative fluid through the heart. The plexus of
conjunctival vessels is mostly represented by small vessels located immediately beneath the conjunctival epithelium. A rather large group
of episcleral vessels is found in a deeper layer (L, lymphatic vessel X150). Fig. 2. After perfusion fixation lymphatic vessels in the conjunctiva
(LV) can be easily recognized from blood vessels (BV). They are irregular in shape, provided with a very attenuated lining and contain
coagulated lymph in the lumen (CE: conjunctival epithelium X200).
other focally at one or more points, thus occluding
the intercellular space. The cytoplasmic organelles of
the endothelial cells are unremarkable. A large number of 7 nm filaments permeate the entire cytoplasm;
these filaments are associated in bundles, which in
cross sectioned vessels appear longitudinally,
obliquely, or perpendicularly cut, indicating that they
traverse the endothelial cytoplasm in random directions. Weibel-Palade bodies are of common occurrence in the cytoplasm of the endothelial cells and
crystalloid inclusions, associated with the rough endoplasmic reticulum, are occasionally found (Fig. 4,
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933342/ on 05/09/2017
728
INVESTIGATIVE OPHTHALMOLOGY & VISUAL 5CIENCE / June 1983
Vol. 24
jurist i v a H
*-. ..
} per icyte
Fig. 3. A cu^maij ui me vuujui»,uvai JJICAUS pusscsscs a very simpie wan composea oy a continuous endothelium and a pericyte. The
endothelial cells have a rather irregular outline and cytoplasmic luminal digitations (arrows). The interendothelial clefts follow an irregular
path (END: endothelium XI 1,000),
inset): these organelles are identical to those previously reported in a variety of adult and developing
ocular blood vessels in Macaca mulatto.6"7 Within
the endothelial cells, a large number of plasmalemmal
vesicles, 60-75 nm in diameter, are present either
isolated in the cytoplasm or opening into the lumen,
tissue front, or intercellular clefts. A small number
of coated vesicles is found also.
The pericytes of these vessels do not differ from
those described elsewhere in the eye6'8"13 or in other
organs.14"17 They are ensheathed by a basal lamina
and are provided with long cytoplasmic processes that
occasionally contact the basal aspect of the endothelial cells without interposition of the basal lamina
(Fig. 4). The nucleus of the pericytes is surrounded
by a thin rim of cytoplasm, and at the edges of the
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933342/ on 05/09/2017
CONJUNCTIVE. AND EPISCLERAL BLOOD VESSEL5 / Raviolo
No. 6
lumen
•
>
-
©
•t
•V
H
Fig. 4. Part of the wall of an episcleral venule. The endothelium is attenuated and the interendothelial clefts follow a labyrinthine path
(arrow). A cytoplasmic process belonging to a pericyte is seen to contact the basal surface of the endothelial cell at the asterisk. The
arrowheads indicate numerous plasmalemmal vesicles at the adluminal surface of a pericyte whereas no plasmalemmal vesicles are present
on the opposite front of the cell. The inset shows a crystalloid inclusion occasionally found in the cytoplasm of the endothelial cells (X16,000;
inset: X 31,000).
nucleus the thickness of the cell declines abruptly.
The cytoplasm contains a variable complement of
filaments, especially concentrated in the cytoplasm
subjacent the adluminal plasmalemma (Fig. 4). Furthermore, as already noted in limbal human capillaries1 as well as in the retina,13 myocardium,14 and
iris,6 the distribution of the plasmalemmal vesicles at
the cell surface is asymmetric: they are commonly
present on the abluminal aspect of the cell and are
quite rare on the opposite cell surface. This asymmetric distribution of the plasmalemmal vesicles is
especially pronounced in the cell processes, whereas
in the region containing the nucleus, the number of
vesicles is usually the same on both fronts of the cells.
Around the pericytes a distinct adventitial layer is
absent: the bundles of collagen fibrils immediately
adjacent to the vessel walls imperceptibly merge into
the surrounding connective tissue.
Arterioles represent a very small component of the
vessel population in the episcleral tissue; they are
characterized by the presence of two to three layers
of smooth muscle cells in their wall.
Conjunctival lymphatic vessels differ from blood
vessels in their irregular lumen, attenuated endothelium, discontinuous basal lamina, and absence of
pericytes- The endothelium is continuous but the interendothelial clefts are open (Fig. 5). The cytoplasmic organelles are not different from those of the
blood capillaries and include occasional Weibel-Palade bodies (Fig. 5, inset) as it has been observed in
lymphatic vessels belonging to other regions of the
body.18
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933342/ on 05/09/2017
700
INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / June 1983
Vol. 24
blood
vessel
••##
I ymphati c
vessel
®
.4-
Fig. 5. Animal perfused with the fixative fluid through the heart. With the electron microscope a lymphatic vessel of the conjunctiva can
be easily recognized from the blood vessel for the very attenuated endothelium, irregular outline and lymph in the lumen. Occasional
lymphocytes (L) are found in these vessels. The inset shows Weibel-Palade bodies in the cytoplasm of an endothelial cells belonging to a
lymphatic vessel of the conjunctiva. (E: endothelium; P: pericyte; N: nucleus, X21,000; inset: X29.000).
Very frequently cross-sectioned bundles of unmyelinated axons ensheathed in Schwann cell processes run in close proximity of blood vessels, especially of arterioles and venules of larger diameter.
Experiments with HRP
Light microscopy: Five minutes after intravenous
injection of HRP, the subconjunctival and episcleral
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933342/ on 05/09/2017
No. 6
CONJUNCTIVAL AND EPISCLERAL BLOOD VE55ELS / Roviola
731
I
Fig. 6. Light micrograph of a chopper section of the anterior segment of the eye, 200 fim in thickness. The animal was injected
intravenously with HRP and 10 min later the eye was enucleated and immersed in the fixative fluid. Dark reaction product is present as
a diffuse stain in the episclera and has diffused at the cornea periphery. The dark color of the iris and ciliary body is mostly due to the
melanin normally present in these tissues of the eye (X50).
'V
X
tissues appear intensely and diffusely stained by the
reaction product (Fig. 6). Ten to 20 min after the
injection, the tracer had begun to invade the periphery of the cornea and to diffuse deeply into the sclera.
Electron microscopy: With the electron microscope
the tracer appears to es'cape the lumen of the blood
vessels by crossing the interendothelial clefts (Fig. 7).
Furthermore, a number of pinocytotic vesicles loaded
with reaction product is present on both the luminal
and tissue fronts of the endothelial cells as well as free
in their cytoplasm. Afterfixationby perfusion through
the heart, HRP is washed away from the lumen of
the vessels, and the number of pinocytotic vesicles
filled with HRP on the luminal front of the endothelium decreases. After this fixation procedure, the
interendothelial spaces of the vessels remain filled
with reaction product throughout their entire length
(Fig. 8). HRP, which has escaped from the vessels,
permeates the surrounding connective tissue spaces,
and it outlines, in negative contrast, connective tissue
cells and collagen fibrils. Five minutes after the injection the tracer has already reached the intercellular
spaces of the conjunctival epithelium, where it is
blocked by the zonulae occludentes that connect the
most superficial epithelial cells (Fig. 9). Only later,
does HRP diffuse into the connective tissue spaces
of the cornea and the compact tissue of the sclera.
Ten minutes after intravenous injection} HRP appears in the lumen of the lymphatic vessels of the
conjunctiva (Fig. 10).
Discussion
Morphology of the Conjunctival and Episcleral
Plexuses of Blood Vessels
In his classical description of the pericorneal blood
vessels in humans, Maggiore19 was able to identify by
meticulous dissection four layers of vessels: (1) a conjunctival network of delicate, thin vessels, that surrounds the cornea at the limbus. These vessels form
loops of variable shape and are very superficial, for
they occur immediately beneath the conjunctival epithelium. Thus, they are visible in the living eye. (2)
A subconjunctival network that is particularly prominent in the regions of the sclera intervening between
the insertions of the recti muscles; after Maggiore,
this layer corresponds to the region of Tenon's capsule. (3) An episcleral plexus, mostly composed by
venous vessels that originate from the confluence of
small venules draining the delicate conjunctival
•V
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933342/ on 05/09/2017
702
INVESTIGATIVE OPHTHALMOLOGY 6 VISUAL SCIENCE / June 1983
Vol. 24
4-
Fig. 7. Animal injected intravenously with HRP. The eye was fixed by immersion 5 min after perfusion of the tracer. From the lumen
of an episcleral venule HRP penetrates the interendothelial cleft following a tortuous path (arrow) and permeates the connective tissue
spaces around the vessel. Note, in the pericyte, the great number of plasmalemmal vesicles loaded with tracer and preferentially located
at the abluminal front of the cell (X34,500).
plexus. (4) An intrascleral network, located in proximity of Schlemm's canal. This plexus does not have
any direct communication with the canal. Schlemm's
canal is connected to the intrascleral plexus by vessels
of peculiar morphology, 20 to 30 in number, that
Maggiore termed for thefirsttime collector channels.
Thus, Schlemm's canal belongs to a vascular plexus
situated in the thickness of the limbus, mainly composed by vessels of venous type and, to a much lesser
extent, by thin arterioles and capillaries. This description provided by Maggiore more than 60 years ago
for the human is also valid for the eye of Macaca
mulatta. In the present study, however, the subconjunctival and episcleral plexuses of Maggiore have
been described together as episcleral vessels since in
radial sections of the anterior segment of the eye it
is often very difficult to distinguish the two vascular
layers as clearly separated entities.
In spite of the fact that Maggiore's observations
were essentially correct, Lee and Holze,20 prompted
by the Chambers and Zweifach description21 of the
microcirculation of the mesentery and omentum,
claimed that the vascular bed of the human conjunctiva is a composite of structural units, each sharing as a basic architectural feature a central, capillarylike arteriovenous channel, of which the true capillaries are side branches. Subsequent observations on
human eyes in vivo by Grafflin and Corddry22 did
not confirm these results: no arteriovenous units were
observed, and the vascular networks seemed to exhibit endless variations without a definite structural
pattern. Bloch,23 also in human eyes, subsequently
reported that the most superficial arterioles are situated usually deeper than their corresponding venules
and that the course of the arterioles is straighter than
that of their accompanying venules. More recently,
Fenton et al.24 observed that the feeding arterioles of
this network in the human eye have a very thin wall
and an inside diameter of 10 yum, that the majority
of the vessels in the conjunctival vasculature are either capillaries or postcapillary venules, and finally,
that all vessels with diameters above 15 ym are venules.
The present study in the rhesus monkey eye con-
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933342/ on 05/09/2017
xA-
No. 6
-V
733
CONJUNCTIVAL AND EPI5CLERAL DLOOD VESSELS / Roviolo
forms to the descriptions in humans: no arteriovenous unit can be discerned in the perilimbal plexus
and, as it is now generally accepted25 the majority of
the vessels can be classified either as capillaries or
venules.
With the electron microscope, both conjunctival
and episcleral vessels are continuous and most of
them are provided with very simple walls. In previous
descriptions on human material a few fenestrated
capillaries were reported in the conjunctiva at the
limbus1 and in the region of the inferior fornix.2 In
the present study, in spite of an accurate analysis of
the same regions, no fenestrated vessels were encountered.
Conjunctival and episcleral blood vessels are permeable to HRP: It is known that perilimbal vessels
are permeable to intravenously injected fluorescein.
Maurice26 reported that when a high concentration
offluoresceinis present in the blood stream of living
rabbits, it is possible to observe the dye entering the
cornea from the limbus. Furthermore, he described
a decline in concentration around the edge of a
stained cornea that he attributed to the return of the
dye to the blood, thus implying that fluorescein can
cross the vascular walls in either direction. Mitsui et
al.27 described the diffusion of intravenously injected
fluorescein out of the vessels in the limbus of normal
human eyes. They observed that in an early phase,
the dye had not yet leaked from the vessels and, thus,
thefluorescenceappeared as strands of cotton waste
attached to the limbus. In a successive phase, fluorescein leaked from the vessels into the surrounding
tissues. This leakage, however, was so diffuse that the
authors were unable to identify the permeable vessels.
In a final, postvascular phase, fluorescence disappeared from the blood vessels and gradually approached the center of the cornea.
Previous studies with electron opaque tracers have
yielded results of difficult interpretation. Thorium
dioxide (10 nm diameter), when introduced intravenously in rats, was seen to diffuse from the limbal
vessels,28 but this finding was attributed to the toxic
properties of Thorotrast. Subsequently, however, the
rat limbal vessels appeared impermeable to carbon
(20 nm diameter),29 and this result was taken as a
further evidence that the permeability of limbal vessels to thorium dioxide was an artifact due to the
unphysiologic properties of the tracer. The present
experiments demonstrate that HRP rapidly diffuses
out of the lumen of conjunctival and episcleral vessels; thus, these vessels are at least permeable to molecules up to the size of this tracer (3 nm radius of an
equivalent hydrodynamic sphere). Evidence that conjunctival and episcleral vessels are permeable comes
also from another source: since these vessels represent
the main route for the return of aqueous humor to
lumen
Fig. 8. Animal injected intravenously with HRP. The eye was
fixed by perfusion of the fixative fluid through the heart. The tracer
has penetrated the cleft between adjacent endothelial cells in an
episcteral venule and has diffused into the surrounding connective
tissue. No tracer is present in the vessel lumen and luminal plasmalemmal vesicles are empty (arrowheads) (X30.5OO).
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933342/ on 05/09/2017
Vol. 24
fNVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / June 1983
zonulae
udentes
•'
A.
•#
Fig. 9. Same expenmentaJ conditions as in Figure 8. HRP, after permeating the intercellular spaces of the conjunctival epithelium, is
finally blocked by the zonulae occludentes which join the lateral aspects of the most superficial epithelial cells (X24.500).
the blood stream, their permeability can be also tested
by introducing tracers into the anterior chamber.
Bill30 injected into the anterior chamber of dead
Macaca irus radioactive diodone (0.5 nm diameter)
and albumin (3.5 nm diameter). He reported that in
the episcleral tissues and in the conjunctiva the diodone space was larger than the albumin space indicating that diodone leaked out of the episcleral and
conjunctival vessels more easily than the larger albumin. Aniline blue injected into the anterior chamber of dead Macaca mulatta eyes31 readily appeared
in the episcleral veins and rapidly diffused in the surrounding tissue. Cole and Monro32 injected fluorescein-labeled dextrans (FD) of different MW into the
anterior chamber of living rabbits. After a 4'/2-min
perfusion at 15-20 mmHg, the FD 40,000 (4.4 nm
diameter) was contained in the lumen of the vessels
of the limbal and conjunctival plexuses, and at the
end of 10 min some leakage had begun into the surrounding tissues. This leakage was more rapid with
FD 3,000 MW than with the larger dextrans 40,000
and 200,000 MW. In more recent experiments fluorescein was injected into the anterior chamber of
living Macaca mulatta; the eyes were freeze-dried immediately after perfusion and sectioned for light microscopy.33M It was found that the episcleral vessels
were leaky to fluorescein and that with longer perfusion times the dye had diffused throughout the entire limbus.
In conclusion, conjunctival and episcleral blood
vessels have a continuous endothelium, but are permeable to tracers of different MW injected either into
the blood stream or into the anterior chamber. From
the behavior of these tracers it can be inferred that,
under normal conditions, blood-borne macromolecules can freely diffuse from these vessels and permeate the subconjunctival connective tissue spaces.
On the other hand, one can also expect that the
aqueous humor that reaches the episcleral and conjunctival vessels from the lumen of Schlemm's canal
-V -
•^-
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933342/ on 05/09/2017
No. 6
735
CONJUNCTIVAL AND EPISCLERAL DLOOD VESSELS / Raviola
. * I1"
4*3
Fig. 10. Ten minutes after intravenous injection of HRP into the blood stream, the tracer is found in the lumen of a conjunctiva!
lymphatic vessel. (End: endothelium, X20.000).
.1
•
»
and collector channels can freely diffuse from the
vascular lumen into the subconjunctival spaces. Furthermore, it is reasonable to suppose that a certain
amount of aqueous humor bypasses Schlemm's canal, freely diffuses across the sclera from the anterior
chamber to the episcleral tissue, andfinallyenters the
permeable vessels of this region. Thus, the composition of the subconjunctival fluid must considerably
vary as a result of changes in both the osmotic or
oncotic pressures of the plasma and the hydrodynamic conditions of the aqueous humor in the channels of the outflow system.
The fact that tracers, either injected intravenously
or into the anterior chamber, permeate the corneal
stroma, suggests that materials diffusing from the
perilimbal vessels may contribute to the nourishment
of the cornea. Furthermore, although the cornea receives most of its metabolites through its anterior and
posterior surfaces, the contribution from the limbal
vessels may be of special physiologic significance,
because macromolecules such as proteins are more
available in the subconjunctival fluid than in the
aqueous humor or tears.
Lymphatic vessels: Since fluids and macromolecules can exit and reenter the permeable episcleral
and conjunctival vessels, the significance of the lymphatic vessels located beneath the conjunctival epithelium is less clear. Possibly, they represent an
emergency system that operates only when the connective tissue spaces of the episclera are overloaded
•
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933342/ on 05/09/2017
706
INVESTIGATIVE OPHTHALMOLOGY b VISUAL SCIENCE / June 1983
with fluids. This might explain why, under normal
conditions, the drainage of the aqueous humor
through the lymphatic system is negligible.35 Lymphatic vessels, however, may be very important in
pathologic conditions, for they represent the most
direct anatomical link between the tissues of the eye
and the immunocompetent cells located in the lymph
nodes.
Key words: Macaca mulatta, conjunctival blood vessels,
episcleral blood vessels, conjunctival lymphatic vessels,
horseradish peroxidase, electron microscopy, interendothelial junctions
Acknowledgment
The expert technical assistance of Ms. Celeste Miller is
gratefully acknowledged.
References
1. Iwamoto T and Smelser GK: Electron microscope studies on
the mast cells and blood and lymphatic capillaries of the human corneal limbus. Invest Ophthalmol 4:815, 1965.
2. Tamura T: Electron microscopical observations of the capillaries of the bulbar conjunctiva in diabetic and non-diabetic
patients. Jpn J Ophthalmol 11:193, 1967.
3. Raviola G: Effects of paracentesis on the blood-aqueous barrier: An electron microscope study on Macaca mulatta using
horseradish peroxidase as a tracer. Invest Ophthalmol 13:828,
1974.
4. Karnovsky MJ: Use of ferrocyanide reduced osmium tetroxide
in electron microscopy. In Abstracts of the Eleventh Annual
Meeting of American Society of Cell Biology, New Orleans,
1971, p. 146.
5. Raviola G and Raviola E: Intercellular junctions in the ciliary
epithelium. Invest Ophthalmol Vis Sci 17:958, 1978.
6. Freddo TF and Raviola G: The homogeneous structure of
blood vessels in the vascular tree of Macaca mulatta iris. Invest
Ophthalmol Vis Sci 22:279, 1982.
7. Freddo TF, Townes-Anderson E, and Raviola G: Rod-shaped
bodies and crystalloid inclusions in ocular vascular endothelia
of adult and developing Macaca mulatta. Anat Embryol
158:121, 1980.
8. Kuwabara T and Cogan DG: Studies of retinal vascular patterns. Part I. Normal architecture. Arch Ophthalmol 64:904,
1960.
9. Kuwabara T and Cogan DG: Retinal vascular patterns. VI.
Mural cells of the retinal capillaries. Arch Ophthalmol 69:492,
1963.
10. Cogan DG, Toussaint D, and Kuwabara T: Retinal vascular
patterns. IV. Diabetic retinopathy. Arch Ophthalmol 66:366,
1961.
11. Hogan MJ and Feeney L: The ultrastructure of the retinal
vessels. II. The small vessels. J Ultrastruct Res 9:29, 1963.
12. de Oliviera F: Pericytes in diabetic retinopathy. Br J Ophthalmol 50:134, 1966.
13. Ishikawa T: Fine structure of retinal vessels in man and macaque monkey. Invest Ophthalmol 2:1, 1963.
14. Forbes MS, Rennels ML, and Nelson E: Ultrastructure of pericytes in mouse heart. Am J Anat 149:47, 1977.
Vol. 24
15. Donahue S and Pappas GD: The fine structure of capillaries
in the cerebral cortex of the rat at various stages of development. Am J Anat 108:331, 1961.
16. Cancilla PA, Baker RN, Pollock PS, and Frommes SP: The
reaction of pericytes of the central nervous system to exogenous protein. Lab Invest 26:376, 1972.
17. Movat HZ and Fernando NVP: The fine structure of the terminal vascular bed. IV. The venules and their perivascular
cells (pericytes, adventitial cells). Exp Mol Pathol 3:98, 1964.
18. Tabuchi H and Yamamoto T: Specific granules in the endothelia of blood and lymphatic vessels in the cardiac valves
of dog. Arch Histol Jpn 37:217, 1974.
19. Maggiore L: Struttura, comportamento e significato del canale
di Schlemm nell'occhio umano, in condizioni normali e patologiche. Ann Ottalmol Clin Ocul 40:317, 1917.
20. Lee RE and Holze EA: The peripheral vascular system in the
bulbar conjunctiva of young normotensive adults at rest. J
Clin Invest 29:146, 1956.
21. Chambers R and Zweifach BW: Topography and function of
the mesenteric capillary circulation. Am J Anat 75:173, 1944.
22. Grafflin AL and Corddry EG: Studies of peripheral blood vascular beds in the bulbar conjunctiva of man. Bull Johns Hop
Hosp 93:275, 1953.
23. Bloch EH: Mikroskopische Untersuchungen des zirkulierenden Blutes in der Conjunctiva bulbi beim Menschen in Gesundheit und Krankheit. Ergeb Anat Entwicklungsgesch 35:1,
1956.
24. Fenton BM, Zweifach BW, and Worthen DM: Quantitative
morphometry of conjunctival microcirculation in diabetes
mellitus. Microvasc Res 18:153, 1979.
25. Hogan MJ, Alvarado JA, and Weddell JE: Histology of the
Human Eye; An Atlas and Textbook. Philadelphia, W. B.
SaundersCo, 1971.
26. Maurice DM: The use of fluorescein in ophthalmological research. Invest Ophthalmol 6:464, 1967.
27. Mitsui Y, Matsubara M, and Kanagawa M: Fluorescence iridocorneal photography. Br J Ophthalmol 53:505, 1969.
28. Szalay J and Pappas GD: Fine structure of rat corneal vessels
in advanced stages of wound healing. I. Permeability to intravenously injected thorium dioxide. Microvasc Res 2:319,
1970.
29. Szalay J: Permeability of rat corneal vessels tofluoresceinand
carbon. Exp Eye Res 18:447, 1974.
30. Bill A: The aqueous humor drainage mechanism in the cynomolgus monkey {Macaca irus) with evidence for unconventional routes. Invest Ophthalmol 4:911, 1965.
31. Jocson VL and Sears ML: Channels of aqueous outflow and
related blood vessels. I. Macaca mulatta (Rhesus). Arch
Ophthalmol 80:104, 1968.
32. Cole DF and Monro PAG: The use of fluorescein-labelled
dextrans in investigation of aqueous humour outflow in the
rabbit. Exp Eye Res 23:571, 1976.
33. Green K, Sherman SH, Laties AM, Pederson JE, Gaasterland
DE, and MacLellan HM: Fate of anterior chamber tracers in
the living rhesus monkey eye with evidence for uveo-vortex
outflow. Trans Ophthalmol Soc U K 97:731, 1977.
34. Sherman SH, Green K, and Laties AM: The fate of anterior
chamberfluoresceinin the monkey eye. I. The anterior chamber outflow pathways. Exp Eye Res 27:159, 1978.
35. Bradbury MW and Cole DF: The role of the lymphatic system
in drainage of cerebrospinal fluid and aqueous humour. J Physiol 299:353, 1980.
X-
-w
-v
Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933342/ on 05/09/2017