| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Graduate Program in Neurobiology and Behavior, Department of Biology, Georgia State University, Atlanta, Georgia 30303
Submitted 13 November 2002; accepted in final form 2 January 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
,
1965) originally proposed that
some response properties in visual cortex, such as receptive field (RF) size,
stimulus orientation tuning, and the simple or complex categories of receptive
field substructure, depend critically on summation of the proper number,
source, and type of afferent inputs to each target cell. This view has been
supported by numerous subsequent studies
(Alonso et al. 2001;
Chapman et al. 1991;
Ferster and Miller 2000;
Martinez and Alonso 2001;
Roerig and Chen 2002;
Usrey et al. 2000). Others
have argued, however, that stimulus tuning is shaped substantially by lateral
inhibition within cortex (Allison et al.
1996; Crook et al.
1998; Eysel et al.
1998; Shevelev et al.
1998; Sillito
1975). That there is still substantial disagreement about the
relative roles of excitatory convergence and lateral inhibition in sculpting
visual response properties suggests that insight may be gained by examination
of a different model system. In this study, we have employed an experimental
procedure in hamsters that increases the degree of convergence to examine
whether RF properties in the superior colliculus (SC) depend on the
convergence of afferents from retinal ganglion cells.
Another question that can be addressed by our model system is how
construction and maintenance of RF properties might be affected by brain
damage. The extent to which plasticity after perinatal brain damage is
behaviorally adaptive depends on how functional properties of the altered
brain region are affected, either by the damage itself or by the compensation
process underlying the plasticity (Payne
and Lomber 2001). Yet there have been few studies addressing this
issue. Because our experimentally induced increase in retinocollicular
convergence involves neonatal brain lesions, we can study the effect of
increased afferent convergence on construction of RF properties after brain
damage.
One major advantage of using hamsters as a model is that the
retinocollicular projections form largely after birth, and thus the
convergence and map refinement processes can be challenged prior to retinal
activation of the SC by manipulations done on the day of birth. Responses to
visual stimulation are not present in SC until approximately postnatal day 12,
1 day before natural eye opening (Binns and
Salt 1997; Clancy et al.
2001; Huang and Pallas
2001). Experimental manipulation of the degree of convergence in
our model system is made possible by an interesting form of plasticity
exhibited by the hamster retinocollicular pathway. Partial ablation of the SC
at birth results in a compressed yet complete map of visual space within the
remaining SC (Finlay et al.
1979). Despite reorganization of retinocollicular connections
after the lesion, afferent/target convergence ratios are conserved at the
single-cell level. The compensatory nature of the reorganization is reflected
in the conservation of normal response properties including RF size, stimulus
velocity tuning, and stimulus size tuning
(Pallas and Finlay 1989). This
remarkable conservation of visual response properties was proposed to result
from activity-dependent mechanisms that preserve single neuron convergence
ratios (Pallas and Finlay
1991; Xiong et al.
1994).
N-Methyl-D-aspartate (NMDA) receptors have been
proposed to play a role in maintaining afferent/target convergence ratios
during normal development in the cerebellum (e.g.,
Rabacchi et al. 1992) and the
thalamocortical pathway (Fox et al.
1996). It has been demonstrated that NMDA receptors are necessary
not only for normal map refinement (Huang
and Pallas 2001; Simon et al.
1992) but also for the plasticity that conserves RF size in SC
neurons within compressed retinocollicular maps
(Huang and Pallas 2001).
Chronic blockade of NMDA receptors in both intact and partially ablated SC
resulted in significantly larger RF sizes compared with normal SC without
affecting visual responsiveness. This effect appears to occur through a
combined blockade of the normal map refinement and through preventing the
compensation for map compression in the partially lesioned SC. These
observations suggested that NMDA receptors play a key role in maintaining
afferent/target convergence ratios in the SC during normal development and map
compression, likely via spike-timing dependent plasticity
(Zhang et al. 1998) and
competition between afferents for target space
(Cline 1998;
Debski and Cline 2002).
Furthermore, they suggested the possibility that afferent/target convergence
ratios could be experimentally manipulated via pharmacological blockade of
NMDA receptors, allowing an examination of the extent to which construction of
response properties depends on summation of afferent RF properties.
Neurons in the SC are selective for stimulus velocity and size
(Chalupa and Rhoades 1977;
Finlay and Pallas 1989;
Huerta and Harting 1984;
Stein and Dixon 1979;
Tiao and Blakemore 1976).
These properties are thought to arise within the SC and are important for SC
mediated behaviors such as visuomotor integration and orientation
(Chalupa and Rhoades 1977;
Lomber et al. 2001;
Sparks et al. 2001;
Tiao and Blakemore 1976;
Wallace et al. 1998). If
spatiotemporal interactions between converging inhibitory and excitatory
retinal inputs are responsible for sculpting neural selectivity for stimulus
velocity and size in SC as has been proposed
(Chalupa and Rhoades 1977;
Goodwin and Henry 1978;
Stein and Dixon 1979;
Tiao and Blakemore 1976;
Waleszczyk et al. 1999), then
these response properties are likely to be affected by natural or
experimentally induced alterations in afferent/target convergence ratios. The
present study was initiated to study the dependence of complex RF properties
on afferent/target convergence and to determine the role of NMDA receptors in
the construction of stimulus velocity and size tuning circuits in the SC of
the hamster during both normal development and reorganization after early
injury. To test the hypothesis that NMDA receptor-controlled afferent
convergence plays a role in creating these response properties in intact and
partially lesioned SC, NMDA receptors were pharmacologically blocked from the
day of birth, and at adulthood, stimulus velocity and stimulus size tuning of
single SC neurons in these experimental groups were compared with those in
normal SC. If the construction of complex response properties is dependent on
convergence of afferent input and if the convergence ratio depends on NMDA
receptor function, then it can be predicted that NMDA receptor blockade would
alter those response properties. On the other hand, if response properties do
not result from summation of afferent input, increasing the afferent/target
convergence by NMDA receptor blockade should not affect stimulus tuning.
Moreover, if NMDA receptors are required for the compensatory plasticity that
conserves velocity and size tuning in partially lesioned SC, their chronic
blockade should prevent not only the normal development of response properties
but should also prevent their conservation in compressed maps.
We found that despite significant increases in RF sizes, NMDA receptor blockade in either intact or partially lesioned SC did not alter stimulus velocity or size tuning of neurons. These results suggest that velocity and size tuning of SC neurons are not dependent on convergence of retinal inputs onto single SC neurons. Although NMDA receptors play a role in refinement of RF size in the intact SC and in refinement of RF size and compensation for mismatched afferent/target ratios in the partially lesioned SC, these data argue that they are not involved in the construction or conservation of velocity and size tuning circuitry in either developmental scenario. This independence from NMDA receptor activity may facilitate preservation of function after SC damage, while allowing topographic maps to conform to changing convergence ratios during either development or evolution.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Forty-two Syrian hamsters (Mesocricetus auratus) were used in this study. Experimental animals were bred in the Georgia State University animal facility from breeding stock purchased from Charles River Laboratories (Wilmington, MA). Normal animals were either purchased as adults or bred in the colony. All animal procedures were approved by the Institutional Animal Care and Use Committee, and met standards of accepted care developed by the National Institutes of Health, the American Physiological Society, and the Society for Neuroscience.
Experimental design
Three groups of animals were used. The normal (N) group (n = 18
animals) received neither surgical nor drug treatment. The
D-2-amino-5-phosphonovaleric acid (D-APV) group
(n = 14 animals) had the biologically active form of the NMDA
receptor antagonist D-APV (Tocris Neuramin, Langford, UK) in Elvax
polymer (DuPont, UK) implanted over the SC at birth and throughout postnatal
development. This group was designed to determine if NMDA receptors play a
role in sculpting stimulus velocity and stimulus size tuning of SC cells
during normal development. In the PT/D-APV group (n = 10
animals), D-APV treatment was combined with heat cauterization of
the caudal part of SC at birth. This group was included to determine if NMDA
receptors play a role in the maintenance of stimulus velocity and stimulus
size tuning under increased afferent availability after partial SC ablation
(Pallas and Finlay 1989).
After rearing the hamsters to adulthood under pharmacological blockade of NMDA
receptors, response properties were assessed using in vivo extracellular
single-unit recording.
Elvax preparation
The D-APV impregnated Elvax polymer was generously donated by
Dr. Adam Smith (University of Oxford, UK) and was prepared according to
published methods (Schnupp et al.
1995; Silberstein and Daniel
1982; Smith et al.
1995). The polymer was prepared to contain a final concentration
of 10 mM D-APV, and a small amount (1: 100,000) of tritiated APV to
provide a measure of drug release rate. The initial procedures prior to
implantation and the drug release characteristics of the polymer have been
reported previously (Huang and Pallas
2001). Briefly, 100- to 200-µm-thick Elvax sheets were
preincubated for 48 h in phosphate-buffered saline (PBS; pH 7.4; 0.5 ml) to
prevent exposing the SC to an initial burst of drug release. The Elvax was
then inserted under the skull and over the SC in the experimental animals. On
implantation on the surface of the SC, the polymer continues to release the
drug at a gradually declining rate for ≤12 mo
(Huang and Pallas 2001;
Smith et al. 1995). We have
demonstrated that this Elvax preparation successfully blocks a substantial
proportion of the NMDA receptor-dependent glutamate component of
retinocollicular transmission, without reducing the AMPA receptor-dependent
component (Huang and Pallas
2001). The same Elvax preparation was used in both studies.
Surgical procedures
Neonatal surgery was performed within 12 h of birth. Hamster pups were initially anesthetized with 4% isoflurane in 0.5 l/min oxygen and then maintained in a deep surgical plane of anesthesia with 1–2% isoflurane. For the D-APV group, an incision was made through the skull at the boundary between the SC and the inferior colliculus, and a sheet of Elvax was cut to fit and slipped under the skull and over the right SC. For the PT/D-APV group, the superficial layers of the caudal portion of the right SC were ablated after insertion of the drug-impregnated Elvax. The pups were returned to maternal care after closure of the wound and recovery from anesthesia. All subsequent procedures were done at adulthood (≥3 mo of age).
Adult hamsters were prepared for physiological recordings by
anesthetization with urethan (0.7 g/ml ip; 0.03 ml/kg in 3–4 aliquots
spaced at 20-min intervals). The pupils were then dilated with a 10%
ophthalmic atropine solution. Respiration rates and withdrawal reflexes were
monitored to ensure a deep level of anesthesia appropriate for surgery, with
supplemental doses of urethan given as needed. After performing a craniotomy
over the SC, the sagittal sinus was ligated and cut. The visual cortex was
bilaterally aspirated to eliminate influences of corticocollicular projections
(Rhoades and Chalupa 1976,
1978) and to facilitate
viewing the surface of the SC for electrode placement. For the experimental
groups, the position of the Elvax strip was noted prior to removing it for
scintillation counting of residual drug content. The brain was kept covered
with sterile saline. To stabilize the eye during the recording session, the
conjunctiva at the nasal corner of the left eye was anchored by a suture to a
stereotaxic frame in which the hamster's head was held. The eye was covered
with a fitted plano contact lens for protection during the recording
session.
Electrophysiological recording
After removal of the Elvax polymer, recording sessions commenced with the
determination of the position of the optic disk with a reversing
ophthalmoscope. For extracellular single-unit recording, tungsten
microelectrodes (FHC, Bowdoinham, ME) with a tip diameter of 1–2 µm
and impedance of 1–3 M
were used. Using a penlight as a search
stimulus, electrode penetrations were made perpendicular to the surface of the
SC to locate visually responsive cells in the retino-recipient superficial
gray layer (SGS) (Pallas and Finlay
1989). A rapid multi-unit mapping of the rostrocaudal extent of
the SC was performed in both D-APV and PT/D-APV cases to
ensure that the entire visual field was represented in the SC. Only those
animals with a full representation of the visual field within the right SC
(the side of experimental manipulation) were studied further. Once the mapping
was complete, electrode penetrations were targeted to isolate single units
residing within 100 µm of the SC surface and with RFs centered within
15° of the optic disk. This region of the SC exhibits regular compression
of the retinal representation after partial SC ablations
(Finlay et al. 1979).
Moreover, restricting the recording sites to this region reduced the
likelihood that differences in response properties at different retinal
eccentricities affected comparisons across the three groups of animals
(Fortin et al. 1999;
Tiao and Blakemore 1976).
Visual stimulation and response selectivity
The location of the excitatory RF (eRF) of each neuron was first determined
using a penlight. A 14-in computer display monitor was then placed 40 cm in
front of the hamster's eye such that the neuron's eRF was in the center of the
monitor. A Sergeant Pepper graphics board (Number Nine, Cambridge, MA) was
used in conjunction with "STIM" software (developed by K.
Christian and Rockefeller University) to generate visual stimuli consisting of
single, smoothly moving light spots that could be varied in diameter,
direction, and velocity. A minimum intertrial interval of 5 s was used to
prevent adaptation. Data were acquired by CED 1401 hardware and processed by
Spike2 software (Cambridge Electronic Design, Cambridge, UK). The nasotemporal
diameter of the eRF was determined by sweeping 1° spots of light from the
top to the bottom of the monitor screen at different nasotemporal locations
with an interstimulus distance of 2° of visual field. The choice of
stimulus velocities and sizes used in this study was guided by previous
results showing that the majority of hamster SC neurons are selective for
small (<7.5°) slowly moving (<10°/s) stimuli
(Stein and Dixon 1979;
Tiao and Blakemore 1976).
Neural selectivity for stimulus velocity was determined by sweeping a 2.5°
diam spot of light in a temporal to nasal direction through the center of the
eRF, using velocities from 5 to 45°/s increasing at 5°/s intervals.
Selectivity for stimulus size was determined by sweeping a spot of light
through the center of the eRF with a velocity of 5°/s and ranging in size
from 2.5 to 15° in diameter at 2.5° size intervals. Each stimulus set
was repeated at least five times, and the responses were expressed as the mean
(±SE) number of spikes for each stimulus velocity or size. The location
of the eRF was re-plotted after the determination of size and velocity tuning
to ensure that the entire stimulus set remained centered on the eRF. The
response selectivity curve for each neuron was normalized to the response
elicited by the best stimulus to facilitate comparison across the three groups
of animals, independent of variations in response magnitude between neurons.
The best stimulus velocity or size was defined as one that not only elicited
the maximum response from the neuron, but also elicited a response that was at
least two times the response to the least preferred stimulus. If a continuous
range of stimulus velocities or sizes satisfied these criteria, the best
stimulus was noted as the middle value of the range. Neurons were also
classified according to the profile of velocity or size selectivity curves as
low-pass, band-pass, high-pass, and nonselective. A neuron was categorized as
low-pass in its velocity tuning if its response to a stimulus moving <15
°/s was at least twice that of the least preferred stimulus. High-pass
neurons were those that responded strongest to stimuli moving >25 °/s,
and had >50% of their maximum response at the highest velocity tested. The
response of neurons classified as band-pass peaked at an intermediate
velocity, with the response falling <50% of the maximum at the slowest and
the fastest velocities tested. Nonselective neurons responded to each stimulus
velocity tested with a response >50% of the maximum. A similar
classification scheme was used for stimulus size tuning.
Histology
At the termination of each recording session, under deep urethan anesthesia, hamsters were perfused via the left ventricle with PBS (0.1 M, pH 7.4) followed by 10% neutral buffered formalin. Brains were removed, postfixed in the same fixative for ≥24 h, and stored in PBS containing 30% sucrose for cryopreservation. They were then sectioned coronally at 50 µm and mounted for Nissl staining with cresylecht violet (Chroma Gesellschaft, Münster, Germany).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
). We
estimated D-APV release from the Elvax strip based on scintillation
counts. The mean D-APV release was 534 pmol/mm2 x
48 h prior to implantation and 98 pmol/mm2 x48 h on the day
of recording response properties. These values of D-APV were
comparable to those seen previously (Huang
and Pallas 2001).
To ensure that the population of neurons characterized in this study was
comparable to those from our previous study
(Huang and Pallas 2001), we
determined whether the effects of NMDA receptor blockade on RF size were
similar across the two studies. We found that the increase in RF size of
single SC neurons in the D-APV and PT/D-APV groups
compared with the normal group in this study was comparable to that reported
previously (Huang and Pallas
2001). The mean (± SE) rostrocaudal RF diameter of single
SC neurons in Normal, D-APV and PT/D-APV animals was
9.32 ± 0.34, 13.5 ± 0.44, and 19.5 ± 1.3°,
respectively (Fig.
1A). The mean RF size of neurons in each group was
significantly different from the other two groups (1-way ANOVA, P
< 0.001). As seen in the previous study, APV treatment shifts the entire
population to larger RF diameters (Fig.
1B). These results confirm that velocity and size tuning
of SC neurons in the experimental animals were created under conditions of
increased afferent/target convergence ratios and therefore allowed us to
determine the effects of increased convergence on these response
properties.
|
SC neurons fall into three categories of velocity-selectivity profiles regardless of rearing conditions
Retinorecipient SC neurons in the superficial gray layer generally exhibit
velocity tuning (Chalupa and Rhoades
1977; Tiao and Blakemore
1976). In single-unit recordings from the SC of normal animals,
consistent with previous studies, we found that most neurons preferred
slow-moving stimuli (low-pass velocity tuning), but a small proportion of
neurons exhibited high-pass or band-pass velocity tuning. We hypothesized that
if velocity tuning results from spatiotemporal convergence of retinal
afferents, then the increased convergence of afferents in the two experimental
groups should affect the velocity tuning of retinorecipient SC neurons. To
test this hypothesis, we recorded the responses from single SC neurons while
presenting stimuli moving across the RF at varying velocities from 5 to
45°/s. We found that chronic NMDA receptor blockade during both normal and
postlesion development did not alter the categorical distribution into three
velocity tuning categories. Figure
2 shows how these categories were defined by presenting examples
of single units from the three categories of velocity tuning profiles under
the different rearing conditions. Low-pass neurons such as those shown in
Fig. 2A responded best
to stimuli moving <15°/s, and their response level decreased gradually
with increasing velocities. Of 45 neurons recorded in normal SC, the vast
majority (42 cells or 93.3%) exhibited a similar low-pass velocity-selectivity
profile (Fig. 2D).
High-pass velocity-selective cells, such as the examples in
Fig. 2B, were
found rarely (1 cell or 2.2%) and responded best to faster stimuli
(>25°/s), with responses falling <50% of maximum at lower velocities
but >50% of maximum at the highest velocity tested. Neurons such as those
shown in Fig. 2C (2
cells or 4.4%) were broadly defined as band-pass because they responded better
to a stimulus moving at an intermediate velocity with responses falling
<50% of maximum at the highest and lowest velocities tested. Bandpass and
high-pass neurons are rare, and together comprised only 6.7% of the population
in the SC of normal hamsters (Fig.
2D). None of the SC cells recorded in this study could be
classified as nonselective for stimulus velocity; they were all velocity
selective based on our 50% criterion.
|
A wide range of best velocities was found in all three experimental groups,
but due to the rarity of high- and band-pass neurons, sufficient data were not
available to perform a population comparison of average best velocities.
However, a comparison across the three groups of animals showed that the
categories of velocity-selectivity profiles in the D-APV and
PT/D-APV groups were comparable to those seen in normal animals
(Fig. 2D;
2 test, 2 = 2.34 df. = 4, P
> 0.6). These results indicate that NMDA receptor blockade and the
resulting increase in RF sizes did not alter the distribution of neurons into
the three velocity tuning categories in the SC. This suggests that NMDA
receptors are not specifically involved in establishing categories of velocity
tuning during normal development. Moreover, activity mediated by the NMDA
receptors does not appear necessary for postlesion maintenance of the
categorical distribution of velocity tuning.
Chronic NMDA receptor blockade does not affect velocity tuning in either intact or partially lesioned SC
We also examined two other aspects of velocity tuning: best velocity of individual SC neurons and velocity tuning of the entire population of SC neurons that were recorded. Consistent with the results on velocity tuning categories, single-unit recordings in adult hamsters showed that mean best velocities in the D-APV (6.37 ± 0.44°/s) and PT/D-PV (6.58 ± 0.53°/s) groups were not significantly different from each other or from the mean best velocity in the normal (5.83 ± 0.32°/s) group (t-test, normal vs. D-APV, P > 0.3; normal vs. PT/D-APV, P > 0.2; D-APV vs. PT/D-APV, P > 0.7; Fig. 3A). Thus chronic NMDA receptor blockade during postnatal development did not alter the distribution of best velocity across the population of SC neurons.
|
To test whether NMDA receptor blockade had any effect on velocity tuning profiles across the population of SC neurons, the responses of the low-pass neurons were normalized and compared across the three groups of animals. The low-pass neurons alone were analyzed in this way because they formed the largest class of tuning profiles and because it was of interest to use a uniform population to compare responses to moving stimuli. No effect of NMDA receptor blockade was observed in low-pass neurons from either intact or lesioned SC (ANOVA, Tukey test for all pairwise comparisons, P > 0.05 in all cases; Fig. 3B). These results indicated that NMDA receptors are unlikely to be involved in the development of velocity selectivity in normal SC. Moreover, blockade of NMDA receptor activity did not appear to have a detrimental effect on the compensatory maintenance of velocity tuning after neonatal partial lesion of the SC. Thus increasing the convergence of afferent inputs on SC cells, and thereby significantly increasing RF size in both experimental groups, had no effect on stimulus velocity tuning in the SC.
SC neurons fall into four categories of stimulus size tuning selectivity profiles regardless of rearing conditions
SC neurons, unlike neurons in the retinogeniculostriate pathway, exhibit
characteristic tuning to stimuli much smaller than the classical RF. We
hypothesized that if this size tuning results from spatiotemporal summation
across multiple retinal afferents, then the experimentally induced increase in
afferent convergence should alter the size tuning profiles. To test this, we
recorded the responses of single SC cells in response to stimuli of varying
sizes. We found that NMDA receptor blockade during either normal or postlesion
development does not alter the three-category distribution of stimulus size
tuning. Figure 4 shows
representative examples of stimulus size tuning profiles from single units
recorded in the SC of normal, D-APV, and PT/D-APV
hamsters. Neurons such as those shown in
Fig. 4A were
classified as low-pass because they responded best to a light spot with a
diameter of 2.5°, and their responses decreased gradually with increasing
stimulus diameter. Neurons like those shown in
Fig. 4B responded best
to the largest stimulus sizes tested in this study and were classified as
high-pass. Neurons like the examples shown in
Fig. 4C were
classified as band-pass because of their strong response to intermediate
stimulus sizes and weak response to smaller and larger stimuli. As with
velocity tuning, the band-pass category for stimulus size was broadly defined,
with selectivity for any intermediate stimulus size considered as an example
of bandpass tuning. Neurons such as those shown in
Fig. 4D were
classified as nonselective because their responses did not fall <50% of
maximum for any stimulus tested. We did not observe any trend for the
experimental treatments to affect the distribution of band- or high-pass
neurons between groups. Of 39 neurons recorded in normal SCs, 24 (61.5%) were
low-pass, 4 (10.3%) were band-pass, 7 (17.9%) were high-pass, and 4 (10.3%)
were nonselective. A statistical comparison of the categories of stimulus size
selectivity profiles across the three groups of animals showed that, as with
velocity tuning, SC neurons were proportioned into the same classes of
stimulus size selectivity regardless of whether they were reared with NMDA
receptor blockade, in the context of either an intact or a partially lesioned
SC (Fig. 4E;
2 test, 2 =4.87, df =6, P > 0.5).
Because the distribution of neurons into categories of stimulus size tuning
remained unaltered, these results indicate that NMDA receptors are unlikely to
be involved in the generation of any particular size tuning category, either
during normal development or following partial SC lesion.
|
Chronic NMDA receptor blockade does not affect stimulus size tuning in either intact or partially lesioned SC
The distributions of best stimulus size and the SC neuronal population size tuning profiles were examined as outlined in the preceding text for velocity tuning. On quantitative analysis of the population of single units from each group, it was found that chronic NMDA receptor blockade had no effect on the distribution of best stimulus size (Fig. 5A) in either intact or compressed maps. The best stimulus size distributions in the D-APV (6.28 ± 0.67°) and PT/D-APV (5.01 ± 0.72°) groups were not significantly different from each other or from the best stimulus size distribution in normal SC neurons (5.21 ± 0.74°; t-test, normal vs. D-APV, P > 0.2; normal vs. PT/D-APV, P > 0.8; D-APV vs. PT/D-APV, P > 0.2).
|
As with population velocity tuning, only units classified as low-pass were compared for the analysis of the effect of NMDA receptor blockade on stimulus size tuning profiles across the population of SC neurons. A population-wide comparison of stimulus tuning was not performed on the band- or high-pass neurons due to their paucity in the samples. No difference was observed in the shape of tuning curves to stimulus size (Fig. 5B), showing that the decrease in response level with increasing stimulus size was similar in all three groups (ANOVA, Tukey test for pairwise comparisons, P > 0.05 for all comparisons). Thus as for stimulus velocity selectivity, chronic NMDA receptor blockade in intact or partially lesioned SC had no effect on stimulus size selectivity. These results indicate that NMDA receptors are unlikely to be involved in generating stimulus size selectivity in normal SC or in the compensatory maintenance of stimulus size selectivity following partial SC lesion. These results also show that altering the afferent/target convergence ratio has no effect on stimulus size tuning, suggesting that tuning has an origin independent of spatiotemporal summation of information from retinal inputs.
The results of this study show that NMDA receptor blockade during postnatal development has no effect on selectivity of neurons in superficial SC for stimulus velocity and size in normal hamsters, or on the conservation of these properties after neonatal partial SC lesions. These results taken together show that tuning of single SC neurons to the size and velocity of movement of a visual stimulus occurs independently of NMDA receptors in SC, map compression, afferent/target convergence, or RF size.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
) and that this is based on an NMDA
receptor-activity-dependent process (Huang
and Pallas 2001). This study evaluated the hypothesis that NMDA
receptor-dependent activity is also required for construction of velocity and
size tuning circuitry during normal development and for postinjury
compensation through its ability to regulate afferent/target convergence. Our
main finding is that stimulus velocity and size tuning of SC neurons are not
affected by blockade of NMDA receptor activity during postnatal development,
despite the resulting mismatch of afferent/target convergence ratios in intact
or partially lesioned SC. This suggests that NMDA receptor activity within the
SC is not necessary for either construction or refinement of stimulus velocity
and size tuning of SC neurons during normal development or after neonatal
partial SC ablation. Thus across large variations in map compression and
afferent convergence, and despite the possible decrease in acuity resulting
from enlarged RFs after early brain damage, the ability to code multiple
aspects of visual stimuli is maintained in SC by NMDA-receptor independent
mechanisms. Implications for neural mechanisms underlying stimulus velocity and size tuning in normal SC
Complex response properties such as stimulus velocity and size tuning could
arise from integration of activity converging from afferent structures to the
target structure (hierarchical processing), from circuitry intrinsic to the
target, from selective projections from afferent structures that are already
tuned (parallel processing) or from a combination thereof. These possibilities
have given rise to multiple hypotheses for how visual response properties are
constructed. For example, one proposed model for velocity selectivity depends
on spatiotemporal interactions between excitatory and inhibitory areas of a
visual neuron's RF (Barlow and Levick
1965; Goodwin and Henry
1978). This idea takes advantage of the fact that stimulus
velocity determines the time interval between the entrance of a moving
stimulus into the inhibitory and excitatory subfields of a neuron's RF. For
slowly moving stimuli, the transition from inhibitory to excitatory subfields
and vice versa takes longer than for fast-moving stimuli. This results in
differences in temporal overlap of inhibitory and excitatory synaptic inputs
and can result in velocity tuning. The model predicts that increasing the RF
diameter of visual neurons would change velocity tuning. Neurons with
different excitatory RF sizes are likely to have different dynamics of
temporal overlap between excitatory and inhibitory inputs for different
velocities, resulting in different velocity tuning. Consistent with this
argument, a correlation has been reported between RF size and velocity tuning
in visual cortex (Orban et al.
1981), lateral geniculate nucleus
(Hess 1979), and SC
(Waleszczyk et al. 1999).
Typically, neurons with larger RFs respond better to faster stimulus
velocities than neurons with smaller RFs. However, we did not find that
experimentally induced increases in RF size altered velocity tuning but
instead found that velocity tuning was unaffected in all respects in normal
and partially lesioned cases. This was true even though in our paradigm the
effect of RF size on velocity tuning could be assessed at similar map
locations. This shows that RF size of SC neurons is not a good predictor of
velocity tuning. The lack of effect of increased RF size on velocity tuning
suggests that spatiotemporal integration of afferent excitatory inputs to
individual SC neurons does not contribute significantly to velocity tuning of
SC neurons.
The results of this study point toward the likelihood that velocity tuning
in SC is not an emergent property but reflects tuning already present in the
retina imposed on the SC by selective parallel projections. The depth of
recording (<100 µm) in this study corresponds to the superficial part of
the stratum griseum superficiale (SGS) in the SC
(Mooney et al. 1985). Although
all types of retinal ganglion cells (RGCs) project to the SC in hamsters,
neurons innervated by rapidly conducting or Y-type RGCs are usually located
within the deeper layers of the SGS and within the stratum opticum
(Fukuda et al. 1978;
Mooney et al. 1985). Neurons
in the superficial SGS receive input only from the slowly conducting, W-type
RGCs (Johnson and King 1982).
The W-type RGCs are typically tuned to slow-moving stimuli, and their
selective projections to superficial SC may be responsible for the selectivity
for slow-moving stimuli that we observed. The lack of effect of altered
afferent/target convergence ratios due to NMDA receptor blockade in the intact
and partially lesioned SC could be explained if the experimental manipulations
resulted in increased input only or mainly from similarly tuned W-type RGCs.
However, the possibility that the NMDA receptor blockade alters inhibitory
circuitry within the SC to preserve velocity tuning cannot be excluded and is
currently under study.
In the hamster SC, close to 60% of neurons prefer stimuli that occupy less
than half of the RF (Pallas and Finlay
1989; Stein and Dixon
1979). The fact that the preferred stimulus size and distribution
of size tuning profiles did not change after the experimentally induced
increases in RF size suggests that the best stimulus size is not a fixed,
optimal percentage of the RF size but rather a reflection of interactions
between spatial summation and inhibition within the eRF
(Stein and Dixon 1979). In
cells that were classified as low- or band-pass, neural responses were maximal
when an appropriate level of spatial summation was reached. Increasing
stimulus size further apparently recruited inhibition and the response began
to decrease. The fact that stimulus size tuning did not change with
experimentally induced increases in RF size suggests that the thresholds for
spatial summation and inhibition are set independently of RF size and that
NMDA receptors are not involved in this process.
Conclusions
While NMDA receptors play a role in refinement of RF size in the intact SC and in refinement of RF size and compensation for mismatched afferent/target ratios in the partially lesioned SC, they are apparently not involved in the creation of velocity and size tuning in either developmental scenario. Thus it appears that NMDA receptor-independent mechanisms are responsible for the creation and conservation of response selectivity for attributes of stimulus motion and size in hamster SC neurons. This independence may facilitate preservation of visuomotor function that depends on stimulus attributes such as size and velocity of behaviorally relevant stimuli, while allowing topographic maps to conform to changing convergence ratios during developmental cell death and collateral elimination. The same mechanisms can act as a substrate to facilitate recovery of function after an injury. Such a lack of conflict between NMDA receptor-activity-dependent and -independent processes in creating multiple RF properties may be of broad applicability in other brain regions and during evolutionary changes in afferent/target population matching.
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
This work was supported by grants to S. L. Pallas from the National Institutes of Health, the National Science Foundation, and the Georgia State University Research Foundation.
Present address of L. Huang: Neurochip Program, Capital Biochip Corp., Beijing 100084, P.R. China.
|
FOOTNOTES |
|---|
Address for reprint requests: S. L. Pallas, Dept. of Biology, Georgia State University, 24 Peachtree Center Ave., Atlanta, GA 30303 (E-mail: bioslp{at}panther.gsu.edu).
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONACKNOWLEDGMENTSREFERENCES |
|---|
Alonso JM,
Usrey WM, and Reid RC. Rules of connectivity between geniculate cells and
simple cells in cat primary visual cortex. J Neurosci
21: 4002–4015,
2001.
Barlow HB and Levick WR. The mechanism of directionally selective units in rabbit's retina. J Physiol 78: 477–450, 1965.
Binns KE and Salt TE. Post eye-opening maturation of visual receptive field diameters in the superior colliculus of normal- and dark-reared rats. Dev Brain Res 99: 263–266, 1997.[Medline]
Chalupa LM and
Rhoades RW. Responses of visual, somatosensory, and auditory neurons in
the golden hamster's superior colliculus. J Physiol
270: 595–626,
1977.
Chalupa LM and
Rhoades RW. Directional selectivity in hamster superior colliculus is
modified ny strobe-rearing but not by dark rearing.
Science 199:
998–1001, 1978.
Chapman B, Zahs KR, and Stryker MP. Relation of cortical cell orientation selectivity to alignment of receptive fields of the geniculocortical afferents that arborize within a single orientation column in ferret visual cortex. J Neurosci 11: 1347–1358, 1991.[Abstract]
Clancy B, Darlington RB, and Finlay BL. Translating developmental time across mammalian species. Neuroscience 105: 7–17, 2001.[ISI][Medline]
Cline HT. Topographic maps: developing roles of synaptic plasticity. Curr Biol 8: R836–R839, 1998.[ISI][Medline]
Crook JM, Kisvarday ZF, and Eysel UT. Evidence for a contribution of lateral inhibition to orientation tuning and direction selectivity in cat visual cortex: reversible inactivation of functionally characterized sites combined with neuroanatomical tracing techniques. Eur J Neurosci 10: 2056–2075, 1998.[ISI][Medline]
Debski EA and Cline HT. Activity-dependent mapping in the retinotectal projection. Curr Opin Neurobiol 12: 93–99, 2002.[ISI][Medline]
Eysel UT, Shevelev IA, and Sharaev GA. Orientation tuning and receptive field structure in cat striate neurons during local blockade of intracortical inhibition. Neuroscience 84: 25, 1998.[ISI][Medline]
Ferster D and Miller KD. Neural mechanisms of orientation selectivity in the visual cortex. Annu Rev Neurosci 23: 441–471, 2000.[ISI][Medline]
Finlay BL, Schneps SE, and Schneider GE. Orderly compression of the retinotectal projection following partial tectal ablations in the newborn hamster. Nature 280: 153–154, 1979.[Medline]
Finlay BL and Pallas SL. Control of cell number in the developing mammalian visual system. Prog Neurobiol 32: 207–234, 1989.[ISI][Medline]
Fortin S, Chabli A, Molotchnikoff S, Penschuck S, Giorgetta O, and Fritschy JM. Maturation of visual receptive field properties in the rat superior colliculus. Dev Brain Res 112: 55–64, 1999.[Medline]
Fox K,
Schlaggar BL, Glazewski S, and O'Leary DDM. Glutamate receptor blockade at
cortical synapses disrupts development of thalamocortical and columnar
organization in somatosensory cortex. Proc Natl Acad Sci
USA 93:
5584–5589, 1996.
Fukuda Y, Suzuki DA, and Iwama K. A four group classification of the rat superior collicular cells responding to optic nerve stimulation. Jpn J Physiol 28: 367–384, 1978.[ISI][Medline]
Goodwin A and
Henry GH. The influence of stimulus velocity on the responses of single
neurons in the striate cortex. J Physiol
277: 467–482,
1978.
Hess RW. Responses of single cells in the cat's lateral geniculate nucleus and area 17 to the velocity of moving visual stimuli. Exp Brain Res 34: 273–286, 1979.[ISI][Medline]
Huang L and
Pallas SL. NMDA receptor blockade in the superior colliculus prevents
developmental plasticity without blocking visual transmission or map
compression. J Neurophysiol 86:
1179–1194, 2001.
Hubel DH and
Wiesel TN. Receptive fields, binocular interaction and functional
architecture in the cat's visual cortex. J Physiol
160: 106–154,
1962.
Hubel DH and
Wiesel TN. Receptive fields and functional architecture in two nonstriate
visual areas (18 and 19) of the cat. J Neurophysiol
28: 229–289,
1965.
Huerta MF and Harting JK. The mammalian superior colliculus: studies of its morphology and connections. In: Comparative Neurology of the Optic Tectum, edited by Vanegas H. New York: Plenum, 1984, p. 687–773.
Johnson MS and King AJ. The synaptic organization of the optic input to the guinea-pig superior colliculus. J Physiol 334: 20–21, 1982.
Lomber GL, Payne BR, and Cornwell P. Role of superior colliculus in analyses of space: superficial and intermediate layer contributions to visual orienting, auditory orienting, and visuospatial discriminations during unilateral and bilateral deactivations. J Comp Neurol 441: 44–57, 2001.[ISI][Medline]
Martinez LM and Alonso JM. Construction of complex receptive fields in cat primary visual cortex. Neuron 32: 515–525, 2001.[ISI][Medline]
McIlwain JT and
Fields HL. Interactions of cortical and retinal projections on single
neurons of the cat's superior colliculus. J
Neurophysiol 34:
763–772, 1971.
Mooney RD, Klein BG, and Rhoades RW. Correlations between the structural and functional characteristics of neurons in the superficial laminae and the hamster's superior colliculus. J Neurosci 5: 2989–3009, 1985.[Abstract]
Orban GA,
Kennedy H, and Maes H. Response to movements of neurons in areas 17 and 18
of the cat: velocity selectivity. J Neurophysiol
45: 1043–1058,
1981.
Pallas SL and Finlay BL. Conservation of receptive field properties of superior colliculus cells after developmental rearrangements of retinal input. Vis Neurosci 2: 121–135, 1989.[ISI][Medline]
Pallas SL and Finlay BL. Compensation for population size mismatches in the hamster retinotectal system: alterations in the organization of retinal projections. Vis Neurosci 6: 271–281, 1991.[ISI][Medline]
Payne BR and Lomber SG. Reconstructing functional systems after lesions of cerebral cortex. Nat Rev Neurosci 2: 911–919, 2001.[ISI][Medline]
Rabacchi S,
Bailly Y, Delhaye-Bouchaud N, and Mariani J. Involvement of the
N-methyl-D-aspartate receptor in synapse elimination
during cerebellar development. Science
256: 1823–1825,
1992.
Rhoades RW and Chalupa LM. Directional selectivity in the superior colliculus of the golden hamster. Brain Res 118: 334–338, 1976.[ISI][Medline]
Rhoades RW and
Chalupa LM. Functional and anatomical consequences of neonatal visual
cortical damage in superior colliculus of the golden hamster. J
Neurophysiol 41:
1466–1494, 1978.
Roerig B and
Chen B. Relationships of local inhibitory and excitatory circuits to
orientation preference maps in ferret visual cortex. Cereb
Cortex 12:
187–198, 2002.
Schnupp JWH, King AJ, Smith AL, and Thompson ID. NMDA-receptor antagonists disrupt the formation of the auditory space map in the mammalian superior colliculus. J Neurosci 15: 1516–1531, 1995.[Abstract]
Shevelev IA, Eysel UT, Lazareva NA, and Sharaev GA. The contribution of intracortical inhibition to dynamics of orientation tuning in cat striate cortex neurons. Neuroscience 84: 11–23, 1998.[ISI][Medline]
Silberstein GB and Daniel CW. Elvax 40P implants: sustained, local release of bioactive molecules influencing mammary ductal development. Dev Biol 93: 272–278, 1982.[ISI][Medline]
Sillito AM.
The contribution of inhibitory mechanisms to the receptive field properties of
neurones in the striate cortex of the cat. J Physiol
250: 305–329,
1975.
Simon DK and O'Leary DDM. Development of topographic order in the mammalian retinocollicular projection. J Neurosci 12: 1212–1232, 1992.[Abstract]
Simon DK,
Prusky GT, O'Leary DDM, and Constantine-Paton M.
N-Methyl-D-aspartate receptor antagonists disrupt the
formation of a mammalian neural map. Proc Natl Acad Sci
USA 89:
10593–10597, 1992.
Smith AL, Cordery PM, and Thompson ID. Manufacture and release characteristics of Elvax polymers containing glutamate receptor antagonists. J Neurosci Methods 60: 211–217, 1995.[ISI][Medline]
Sparks DL, Freedman EG, Chen LL, and Gandhi NJ. Cortical and sub-cortical contributions to coordinated eye and head movements. Vision Res 41: 3295–3305, 2001.[ISI][Medline]
Stein BE and Dixon JP. Properties of superior colliculus neurons in the golden hamster. J Comp Neurol 183: 269–284, 1979.[ISI][Medline]
Tiao Y-C and Blakemore C. Functional organization in the superior colliculus of the golden hamster. J Comp Neurol 168: 483–504, 1976.[ISI][Medline]
Usrey WM,
Alonso J-M, and Reid RC. Synaptic interactions between thalamic inputs to
simple cells in cat visual cortex. J Neurosci
20: 5461–5467,
2000.
Waleszczyk WJ, Wang C, and Dreher B. Velocity response profiles of collicular neurons: parallel and convergent visual information channels. Neuroscience 93: 1063–1076, 1999.[ISI][Medline]
Wallace MT,
Meredith MA, and Stein BE. Multisensory integration in the superior
colliculus of the alert cat. J Neurophysiol
80: 1006–1010,
1998.
Xiong M, Pallas SL, Lim S, and Finlay BL. Regulation of retinal ganglion cell axon arbor size by target availability: Mechanisms of compression and expansion of the retinotectal projection. J Comp Neurol 344: 581–597, 1994.[ISI][Medline]
Zhang LI, Tao HW, Holt CE, Harris WA, and Poo M-m. A critical window for cooperation and competition among developing retinotectal synapses. Nature 395: 37–44, 1998.[Medline]
This article has been cited by other articles:
|
|
 |
|
  K. A. Razak and S. L. Pallas Inhibitory Plasticity Facilitates Recovery of Stimulus Velocity Tuning in the Superior Colliculus after Chronic NMDA Receptor Blockade J. Neurosci., July 4, 2007; 27(27): 7275 - 7283. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. L. Pallas, P. Wenner, C. Gonzalez-Islas, M. Fagiolini, K. A. Razak, G. Kim, D. Sanes, and B. Roerig Developmental Plasticity of Inhibitory Circuitry J. Neurosci., October 11, 2006; 26(41): 10358 - 10361. [Full Text] [PDF]
|
|
|
|
 |
|
 K. A. Razak and S. L. Pallas Neural Mechanisms of Stimulus Velocity Tuning in the Superior Colliculus J Neurophysiol, November 1, 2005; 94(5): 3573 - 3589. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. M. Carrasco, K. A. Razak, and S. L. Pallas Visual Experience Is Necessary for Maintenance But Not Development of Receptive Fields in Superior Colliculus J Neurophysiol, September 1, 2005; 94(3): 1962 - 1970. [Abstract] [Full Text] |
