Abstract
purpose. To evaluate the effects of ganglion cell depletion from experimental
glaucoma on the relative metabolic activities of neurons in the
cytochrome oxidase blobs of V1 cortex in the macaque visual system.
methods. Monocular experimental glaucoma was induced in adult monkeys
(Macaca mulatta and Macaca fascicularis) by laser
application to the trabecular meshwork, increasing the intraocular
pressure. After other experiments, the primary visual cortices were
analyzed for functional excitation from surviving ganglion cells, as
indicated by cytochrome oxidase histochemistry.
results. Cytochrome oxidase reactivity was uniformly reduced in blobs with input
from the glaucomatous eye in a manner consistent with loss of known
afferent inputs. The average size of glaucomatous blobs in layers 2 and
3 of V1 cortex was reduced by half.
conclusions. Experimental glaucoma in monkeys reduces retinal input to the central
nervous system, thereby reducing the metabolic drive to downstream
targets, as indicated by the reduction in the size of cytochrome
oxidase blobs in layers 2 and 3 of V1 cortex. The pattern of cytochrome
oxidase loss within the blob was uniform, suggesting that all sources
of afferent input to the blobs were affected by experimental
glaucoma.
In a previous article we described the downstream reduction
in metabolic activity in the lateral geniculate nuclei (LGN) and in the
magnocellular (M) and parvocellular (P) input divisions of the layer 4C
to V1 cortex after experimental glaucoma.
1 It was
concluded that as experimental glaucoma depleted the retina of ganglion
cells, there was a corresponding metabolic reduction in both M-cell and
P-cell pathways, with a significantly greater reduction in the P-cell
rather than the M-cell pathway in layer 4C of V1 cortex. This article
continues the presentation of the results of those experiments by
describing the concurrent change in neuronal metabolism in the
cytochrome oxidase (CO) blobs of the superficial layers 2 and 3 of the
primate V1 cortex, the target of differential input from the two
parallel pathways into layer 4C. The density and distribution of
cytochrome oxidase reactivity (COR) is used as an indicator of the
level of afferent visual signal from retinal ganglion
cells
2 to the blobs in layers 2 and 3.
CO is an essential mitochondrial membrane protein for brain oxidative
metabolism, catalyzing the last step in the formation of adenosine
triphosphate (ATP), the energy source for neuronal
function.
2 3 4 The brain tissue content of CO is
inhomogeneously distributed
4 5 6 and varies with the
metabolic demand attendant with neuronal activation.
7 8 9 10 11 12 In the afferent visual system of primates, histochemistry has shown CO
reactivity to decrease monotonically with the duration of sensory
visual deprivation by enucleation or by tetrodotoxin (TTX)
blockade.
13 14 Therefore, it is reasonable to expect CO
reactivity to be at covariance with the level of activation of neurons
in the monkey afferent visual system, including the blobs of
superficial layers 2 and 3.
CO-rich blobs are oft described, and now well-known, metabolic features
of the superficial layers 2 and 3 of primary visual cortex. In most
primates, CO blobs are distributed in periodic rows directly above the
center of the eye-dominance columns (ODC) of input layer 4C and are
known to derive a major afferent input from the two laminar
subdivisions, layer 4Cα and 4Cβ, directly beneath. In turn, the
input to these two subdivisions comes from the two major ganglion cell
projection pathways, with the M-cell pathway (carrying primarily
retinal information gathered by the large parasol retinal ganglion
cells) projecting into layer 4Cα, and the P-cell pathway (carrying
retinal information gathered by the small- to medium-sized retinal
midget ganglion cells) projecting into layer 4Cβ. In addition to
these cortical afferents, the CO blobs get a direct thalamic input from
the K-cell laminae of the lateral geniculate nucleus (carrying retinal
information gathered by small bistratified ganglion cells). Although
both layer 4C subdivisions send projections to superficial layers 2 and
3, the CO blobs are thought to receive a major input from layer
4Cα.
15 Whereas the surrounding interblob areas get input
from both divisions of layer 4C, the input from 4Cβ contributes far
more synaptic contacts.
16 Because these studies have shown
that there are multiple afferent inputs to CO blobs, and because we and
Vickers et al.
17 have shown that experimental glaucoma
reduces CO staining in the macaque afferent visual system and we have
shown a differential effect of experimental glaucoma on two of the
primary afferent inputs to the blobs,
1 we next measured
changes in the size and metabolism of the CO blobs in these same
monkeys.
Five adult monkeys (three
Macaca mulatta and two
Macaca fascicularis) were used in accordance with Office of
Protection from Research Risks Public Health Service (OPRR-PHS) Policy
on Humane Care and Use of Laboratory Animals (revised 1986), the ARVO
Statement for the Use of Animals in Ophthalmic and Vision Research, and
all research protocols had the approval of the institutional animal
care and use committee (IACUC). The intraocular pressure (IOP) of the
right eye of each subject was elevated by laser application to the
trabecular meshwork using the procedures described by Quigley and
Holman.
18 Subsequently, IOPs were measured in the
anesthetized subjects, typically at weekly intervals for a period of
approximately a year, using a handheld applanation tonometer. The
animals had clear ocular media throughout. The details of the
treatments, behavioral testing, electrophysiology, and metabolic
changes in the lateral geniculate nucleus (LGN) and in layer 4C of V1
cortex have been referenced and presented earlier.
1 All
monkeys had severe experimental glaucoma-related visual damage as
indicated by the Humphrey visual field map (HVFM; Humphrey Instruments,
San Leandro, CA; see Reference
1 , Tables 1 and 2, for
background of three of the subjects:
M. mulatta OHT9, OHT18,
and OHT21).
Figure 1 is presented to indicate the level of glaucomatous damage in the five
monkeys whose visual brains were analyzed in this article. The figure
shows the HVFM for four monkeys.
Figure 1A is a map (OHT19) typical of
the threshold for detecting a visual defect, whereas the HVFMs of
Figs 1B 1C and 1D are from monkeys used in the present study, and show
severe defects from experimental glaucoma (OHT18, OHT9, and OHT21,
respectively). Two other animals (
M. fascicularis AL540 and
AL203) were treated similarly, but systematic treatment and behavioral
data were not collected on them. The visual field defects shown in
Figures 1B 1C and 1D are indicative of advanced glaucomatous damage
and a severe loss of retinal ganglion cells. Although counts of the
loss in ganglion cells in these experimental animals are in preparation
and incomplete, it is clear from our comparable counts in other animals
treated similarly (e.g., see Reference
19 Fig. 2 ) that all animals analyzed in this report had advanced glaucomatous
retinal damage.
Measurements of CO blobs were made in six visual cortices, four
ipsilateral and two contralateral to the glaucomatous experimental eye.
The measurements were made at a cortical locus corresponding to the
upper temporal retinal representation of the hemifield, approximately
7° to 10° from the fovea; a retinal representation generally
showing severe but not total loss of visual sensitivity.
At the end of the behavioral or electrophysiological study, the
animals were killed by overdose (100 mg/kg) of pentobarbital sodium
(Nembutal; Abbott, Abbott Park, IL), and exsanguinated with 2 l of
saline followed by 2 l of a 2% paraformaldehyde–0.5%
glutaraldehyde fixative in phosphate buffer (pH 7.4). The brain and
optic nerves were removed and (in most cases) the visual cortices were
dissected and gently flattened on a glass slide. The tissue was
refrigerated overnight in the fixative before beginning a sucrose
dehydration gradient of 10%, 20%, and 30%. The visual cortices were
embedded in aluminum foil cups filled with TissueTek (Miles, Elkhart,
IN) and frozen by lowering into an acetone bath cooled by liquid
nitrogen. Tangential sections of 30-μm thickness were collected and
stained for the histochemical localization of CO according to the
protocol of Wong–Riley.
2 The CO-stained sections were
mounted on gelatinized slides, dehydrated, and coverslipped (Permount;
Fisher Scientific, Fairlawn, NJ).
The CO-stained sections were homogeneously back illuminated, and
a digital image was captured at a 1200-pixel resolution with a camera
(DCM1; Polaroid, Cambridge, MA), which was linear with optical density,
with an R = 0.99. The density of the COR was scaled 0
to 255, where 0 = opacity and 255 = the incident light.
Measurements were taken before any filtering or contrast enhancement.
COR = I − T, where I was the incident light (nominally, a
value of 255), and T was the light transmitted through the tissue
containing the CO-reaction product. The value of COR from a blob
connected with the glaucomatous experimental eye
(CORG) was always compared in the same section
with the value of COR measured in the directly adjacent blob connected
with the normal companion eye (CORN). The ratio,
CORG/CORN, constituted the
primary data from which mean values and variances were calculated. The
mean COR and SD of a closed contour, drawn by eye to best outline the
blob, were recorded for each blob and compared with the values obtained
in the companion blob with primary input from the opposite eye. Thus,
the ratio of CORG/CORN was
formed. A minimum of 10 such ratios was then averaged for each tissue
location. Most often, the ratio
CORG/CORN was expressed as
a percentage reduction of COR, relative to that in the companion site
that had input from the normal eye.
For the relative size measurement, the numbers of pixels contained
within the closed contour of the tracing of the CO blob image was
compared. It is recognized that there is no generally accepted
objective criterion for drawing the CO blob boundaries, because there
is a characteristic decreasing COR gradient from the blob center toward
the edge.
14 Therefore, we present a statistical comparison
of the subjective sizes of pairs of CO blobs, collected by a technician
naive to the purposes of the study, and the results evaluated by the
paired
t-test.
The brain section image was processed using the University of Texas at
San Antonio Image Tool 2.0 image analysis software. With the
reticulated CO pattern characteristic of layer 4A as a reference, the
perimeters of the CO blobs in layer 3 were traced by eye and the
numbers of pixels within the enclosed perimeter taken as an index of
the relative area of blobs with input from the normal and the
glaucomatous eye. The average area and relative COR were measured for
pairs of CO blobs, one from the normal left (ODC), paired with the
adjacent CO blob from the ODC of the glaucomatous right eye. A minimum
of 10 pairs was measured from the V1 cortices of each of five of the
experimental monkeys. The primary data were imported into a spreadsheet
(QuattroPro; Corel, Ottawa, Canada) for computation and graphing,
whereas statistical comparisons were made using the paired t-test (SigmaStat; Jandel Scientific, San Rafael, CA).
To describe the distribution of relative COR within the blob, pairs of
normal and companion glaucomatous blobs were scanned along a line
orthogonal to the course of the eye dominance columns. Smoothing the
resulting noisy curves was achieved by subjecting the data to a 3-point
rolling average to generate a 48-point profile for 10 pairs of blobs.
To present a different and enhanced view of the density, size, and
distribution of COR within the blobs, the following operations were
performed on selected blob fields. First, the density range of the
image was inverted so that the low-intensity blob COR density signal,
represented by the dark COR blobs, became high-intensity values. Next,
the pixel values in the holes representing the blood vessels were
replaced by the average value of the immediately surrounding pixels. A
surface plot was then made so that the pixel intensity was coded both
by height of the surface and by color. An azimuth and elevation was
chosen to optimize inspection of differences between rows of blobs. In
practice, low-pass filtering was often performed to accentuate the
major features over high-frequency variations.
To fill and replace the blood vessel holes, two different protocols
were used with approximately equal success. In the first protocol, the
holes were selected in the image management software (Photoshop; Adobe,
San Jose, CA) by selecting that range of intensities. A mask was
created corresponding to the holes. The remainder of the image without
the holes was then blurred by several passes of a median filter. The
calculation facility of the software program was then used to replace
the masked area corresponding to the holes with corresponding portions
of the blurred image. This replaced the white holes with a local
average of the surrounding area. In the second protocol, a custom“
zippering” routine was implemented in a statistical analysis
software program (The MatLab; MathWorks, Natick, MA). By this means, an
upper threshold was imposed that distinguished the high-intensity holes
from the darker tissue. The routine then replaced pixels exceeding
this threshold with the value of their nearest neighbor not exceeding
the threshold. By this mechanism, the surrounding regions “flooded”
the holes with their values. This procedure was more robust in dealing
with a wide variety of tissues but sometimes replaced the holes with
values somewhat lighter than the near background, probably because of
light scattering near the holes.
The effects of experimental glaucoma on the neural metabolism
within the CO blobs of V1 layers 2 and 3 are illustrated in the two
panels of
Figure 2 . The top panel is a view of a tangential section
showing a view of the rows of CO blobs that lie directly above a
comparable field of ODC stripes in layer 4C (lower panel). Parallel
rows of encircled CO blobs, running top to bottom, are illustrative of
the tracings made by eye for the estimation of the relative area of the
CO blobs having input from the normal eye (large arrows, N) and from
the glaucomatous eye (small arrow, G) indicated in the lower panel. The
CO blobs associated with input from the normal left eye are
characteristically large and interconnected by a band of COR of a
higher density than background. The adjacent row of CO blobs is
associated with afferent input from the glaucomatous right eye and is
characterized by smaller CO blobs without the interconnecting band of
extrablob COR. In the lower panel, the CO-rich ocular dominance columns
from layer 4C, having input from the normal eye (N), and the CO-poor
columns with afferent input from the glaucomatous right eye (G), are
seen juxtaposed to comparable rows of blobs in the upper panel.
Figure 3 shows the relative COR values from the six visual cortices of the
individual experimental animals for blobs associated with input from
the glaucomatous eye (G) with those blobs with input from the normal
eye (N). The average relative reduction in COR in blobs associated with
the glaucomatous right eye for all the monkeys was 9% (range, 2–14%.
The average COR tended to be reduced in the blob associated with the
glaucomatous eye, with four of the six comparisons being significant
reductions (paired
t-test;
P < 0.01).
Normally, the level of COR in the blobs is approximately the same in
both visual cortices, as shown in the normal eye data from OHT18 (R, L)
displayed in
Figure 3 . The overall percentage of reduction in COR was
insignificantly greater between blobs on the ipsilateral right cortex
(14%) compared with those blobs in the cortex contralateral to the
glaucomatous eye (12%), as might be expected owing to the
characteristic spread of the glaucomatous lesion.
1
Overall, the average COR data for the five experimental monkeys
(Fig. 3) shows that there was a significant reduction in the COR between the
blobs having input from the glaucomatous right eye (RE) compared with
those blobs having input from the normal left eye (LE). The
distribution of the loss in COR within the blob is illustrated in
Figure 4A , showing the COR profiles of 10 blob pairs. The shape of curve of the
means and SD for 10 glaucomatous blobs was comparable with that
of the companion 10 normal blobs, differing only in the lower
relative level of COR in the glaucomatous blobs throughout. This
suggests a uniform loss in COR throughout the blob, which is borne out
in
Figure 4B , showing a linear regression fit to the 24 mean values
from the center of the blob into the interblob space. The parallel COR
data and the regression curves attest to a uniform loss in COR
throughout the glaucomatous blob.
Experimental glaucoma dramatically reduced the relative size of the CO
blob, as is shown in
Figure 5 . In every case, the size of the CO blob receiving input from the
glaucomatous eye was significantly smaller than the normal companion CO
blob (six paired
t-tests:
t > 9.0,
df = 22,
P < 0.001) with an average
shrinkage of 53%. The smallest amount of shrinkage of the CO blob is
seen in the left visual cortex (L) of OHT18, where the average reduced
size was still a significant 23%. Therefore, experimental glaucoma
dramatically reduces the spatial domain of the affected eye in
superficial primary visual cortex, with a specific impact on the
metabolic activity in the CO blobs.
The reduction in blob size and COR distribution is apparent in the
rendering of the tangential surface plot of a field of CO blobs shown
in
Figure 6 .
Figure 6A shows the unfiltered image of a field of CO blobs from
layers 2 and 3, with the ODCs running almost vertically. The dashed
line contour connects four blood vessels that serve as fiducial marks
for spatial registration of the
Figures 6A 6B 6C 6D .
Figure 6B shows
the contrast-enhanced field of CO blobs for better viewing.
Figure 6C shows the density reversal of the image, with the holes of the blood
vessels appearing as black spots and the COR appearing white.
Figure 6D shows the image after the thresholding and application of a hole
zippering routine, which effectively removed the blood vessel and hole
artifacts from the image, leaving the COR displayed as a
light-gray–scale image. The rows of CO blobs associated with the
experimental glaucoma of the right eye (R) and those associated with
the normal eye (L) are marked, as is the subfield of CO blobs presented
in
Figure 6E .
Figure 6E is a surface plot of the subfield of CO blobs marked in
Figure 6D . The COR density is color-coded, and the field of blobs
appears as rows of parallel peaks. Alternate rows of blobs associated
with the normal eye appear as robust, broad peaks interconnected along
the row by a higher level of COR, whereas the adjacent rows of blobs
connected with the glaucomatous eye appear much smaller and do not have
the interconnecting bridge of COR.
In summary, as the afferent input from ganglion cells to the visual
brain is reduced by experimental glaucoma, the CO blobs of the
superficial layers of primary visual cortex become smaller in a manner
consistent with a uniform loss of metabolism within the blob.
These results show that experimental glaucoma dramatically altered
metabolic activity as well as the spatial domain of the affected eye in
superficial layers of primary visual cortex. As retinal ganglion cells
became nonfunctional (and eventually died), the entire downstream chain
of neurons became metabolically reduced, and the cortical spatial
domain formerly controlled by the affected eye became quiescent. That
is not to say that these neurons were inactive, just that they had lost
much of their primary afferent signal and were relatively reduced in
their neural activity, as the CO marker for neuronal metabolism
indicated.
The effects of glaucoma on the superficial CO-rich blobs of the V1
cortex are interesting in that the pattern of loss in COR suggests a
uniform loss throughout the blob. Trusk et al.
14 and
Edwards et al.
20 have characterized the density and
pattern of change in COR within the normal CO blob, showing that COR is
highest in the blob center, indicating that there is normally an
underlying anisotropy in metabolic activity within the blob. Trusk et
al.
14 also showed that the COR reduced uniformly
throughout the blob after deafferentation by enucleation, TTX blockade,
and monocular eyelid suture conditions similar to that associated with
deafferentation by glaucoma as we have shown here. Therefore, a uniform
reduction in COR throughout the blob accounts for both the smaller blob
size and the consistent COR profile in the blobs associated with the
glaucomatous eye reported here.
Because the CO blob gets multiple inputs, it is surprising that the
effect of glaucoma was uniform throughout the blob. Because we have
shown
1 that both P- and M-cell LGN layers supplying
signals to V1 are metabolically, but differentially, inactivated by
glaucomatous blockade, it is worthwhile to consider the contribution of
those other pathways having input to the CO blobs.
One obvious source of stimulation to the CO blobs of layers 2 and 3 is
the direct thalamic input from the koniocellular (K)-cells from the
extralaminar zones of the LGN.
21 22 23 As the P- and M-cell
input to the blobs weaken and the blob shrinks, the K-cell input could
sustain elevated metabolism within the shrinking blob. However, because
the K-cell itself gets a major afferent input from a third morphologic
and functional class of retinal ganglion cell, the small bistratified
ganglion cell
23 24 (thought to carry blue–yellow color
information; see recent review
25 ), it too would probably
be subject to the same deleterious effects of glaucoma. Although the
K-cell receives other input (first from the superficial layers of the
superior colliculus [SC], as well as from the parabigeminal nucleus[
PG]), it is hard to see how these indirect afferent routes into the
CO blobs could sustain metabolic activity, because both these sources
are themselves dependent on retinal ganglion cell input. For example,
the SC superficial layers get a large projection from V1, layer 5,
which would be downstream of the impaired P-, M-, and K-cell
geniculocortical projection just discussed. Moreover, it would be
expected that advanced glaucoma would weaken the direct ganglion cell
input to SC in a similar manner to the effects shown for the
geniculocortical projection.
Moreover, if one uses the cell size argument in the sequence of
impairment and death of ganglion cells in glaucoma
26 (i.e., that it is the ganglion cells with the largest soma size that
are first affected in glaucoma) the small bistratified ganglion cell
supplying input to the K-cell would have no particular survival
advantage, in that these ganglion cells have soma sizes intermediate to
midget and parasol ganglion cells having input to the P- and M-cell LGN
layers, respectively. Therefore, it is unlikely that K-cells would have
any survival advantage in glaucoma.
Because it has been suggested that the collective input to the blobs
from the three LGN sources (layers 4Cα, 4Cβ, and K-cell) represents
less than 20% of the synaptic connections within the CO
blob,
25 most of the synaptic contacts within the blob must
come from intracortical sources. For example, if the mosaic of blobs
has some binocular interconnections (either excitatory or inhibitory)
the active blobs connected with the normal eye could affect stimulation
to the CO blob connected with the glaucomatous eye. What evidence is
there for such a pathway?
Several studies have shown that there are extensive intracortical
lateral projections (spreading axons with periodic terminations) within
the superficial layers of V1 of primates. Orthograde tracers have been
injected within the layer 3 CO blob, and the characteristics of the
lateral axonal spread described.
27 Malach et
al.
28 showed that biocytin injections within one
CO blob marked axonal spread that tended to contact adjacent blobs
connected with the same eye, but not to the blobs connected with the
opposite eye. However, using similar methods, Yoshioka et
al.
29 have shown that these lateral projections spread
predominantly orthogonal to the course of the ODCs, where axonal
branches contact most of the adjacent CO blobs. These two studies
reinforce the original paper on the subject by Livingstone and
Hubel
30 who used horseradish peroxidase to show that blobs
are connected with blobs, and nonblob areas are interconnected with
other nonblob areas. These studies agree that there is a statistical
preference of these lateral projections to link CO blobs in ODCs
associated with the same eye, but that there are significant exceptions
observed in each study. Therefore, although most of the lateral
interconnections seem to be between like compartments (e.g., blobs to
blobs; interblobs to interblobs, and interblobs within ODCs of the same
eye) a significant percentage of the axonal terminals make contact with
the CO blobs associated with the opposite eye. Such reciprocal contacts
between pairs of CO blobs could support interblob neural coherence and
provide a synchrony between ODC domains analyzing the same part of
visual space. It could be speculated further that a neural coherence is
necessary to knit together a higher order binocular perceptual map,
with the CO blobs serving as neuronal fiducial points for keeping the
monocular maps synchronized and functionally aligned. From our results,
there seems to be no differential spatial input from these sources, in
that the loss in COR appears to uniform throughout the blob. As the
afferent inputs to the blob from layers 4Cα and 4Cβ and from the
K-cells of the LGN fail with the progressive death of the retinal
ganglion cells, the lateral input from the adjacent column driven by
the normal eye are apparently insufficient to sustains metabolic demand
within CO blobs of the glaucomatous eye. Of course, a strong counter
argument to this possibility is the absence of binocularity of neurons
within the blobs.
Experimental glaucoma produces a differential reduction in the layer 4C
input sublaminae, with a greater loss in 4Cβ than in
4Cα.
1 This result, which we reported earlier, was
consistent with what we had found in the LGN (that experimental
glaucoma had a greater effect on the metabolism of the P-cellular
pathway, than on the M-cell pathway), but incompatible with the earlier
suggestions of Glovinsky et al.,
26 who presented evidence
to the contrary—that the larger parasol ganglion cells, which make up
the primate M-cell retinogeniculocortical pathway are affected first
and more severely in glaucoma in humans and in experimental glaucoma in
monkeys. The results occurring in the CO blobs reported herein suggest
a uniform reduction in COR throughout the blob, consistent with a
uniform spatial projection to the blob from the 4Cα and the 4Cβ
sublaminae. By contrast, Edwards et al.
20 have reported
recording from a small number of neurons within the center of the blob
having a higher contrast sensitivity (a characteristic of the M-cell
pathway) than those neurons near the periphery of the blob, suggesting
an anisotropy in the inputs from the M-cell and P-cell pathways. The
uniform reduction of COR throughout the blob reported here provides no
evidence for a differential spatial projection of M- and P-cell input
within the blob and is consistent with the results of Trusk et
al.
14
These results further extend the description of the pathophysiology of
glaucoma along the afferent chain of anatomic sites from the eye to the
superficial layers of primary visual cortex.
The authors thank Shilpa Patel for technical assistance.