Ultrastructural changes in kitten visual cortex after environmental modification

Brain Research, 66 (1974) 165-172 © Elsevier Scientific Publishing Company, Amsterdam - Printed in The Netherlands

165

Ultrastructural changes in kitten visual cortex after environmental modification

L. J. GAREY* AND J. D. PETTIGREW** Department of Physiology-Anatomy, University of California, Berkeley, Calif. 94720 (U.S.A.) (Accepted October 17th, 1973)

During the critical period of development o f the visual system of the cat, deprivation can profoundly affect the physiology of the cerebral cortex2, 9-11. Changes can be brought about by very brief exposure 3 and appear to be virtually permanentlL In a search for an underlying morphological change various workers have reported ultrastructural alterations in animals deprived o f visual cues to a greater or lesser degree4,7,1L The present experiments were performed on a preparation in which visual exposure could be given to one hemisphere while the other acted as an unstimulated control. The modifications were studied and their time course determined, the physiological results being presented in the previous paper 14, and the morphological findings here. Five kittens which had been reared in darkness for approximately 1 month received specific stimuli confined to one half of the visual field only. This was achieved by aligning the vertical meridian of the field with the edge of a pattern projected onto a screen in front of the anesthetized, paralyzed preparation and masking the rest o f the field. In this way one visual cortex received specific stimulation for up to 30 h while the other remained unstimulated. For 4 kittens the stimulus was a vertical grating moving over the right hemifield of one or both eyes while the fifth had a variety o f stimuli presented to the right hemifield of both eyes simultaneously. One animal (C11) was perfused through the heart with a buffered mixture of paraformaldehyde and glutaratdehyde 1 h after stimulation, and in the other 4 single unit recording was performed immediately or after a delay of some days using patterns projected onto the same screen used for stimulation ~4. These 4 animals were perfused at the termination of the physiological experiment. Blocks of area 17 of both sides were prepared for electron microscopy; the blocks were taken from the medial side of the lateral gyrus to avoid both the 17-18 border and the electrode tracks: micrographs were taken of" layers III and IV at a standard magnification o f × I4.000 and prints made at a further * Current address and address for reprint requests: lnstitut d'Anatomie, 9, Rue du Bugnon, CH 10tt Lausanne, Switzerland. ** Current address: Division of Biology, California Institute of Technology, Pasadena, Calif. 91109, U.S.A.

166 5c,r

SHORT COMMUNICATIONS 300

l

-

o

o o

%

2:20~

• unsl,mulaled

I

o

?OC

, S l ~ u l 0 t e~

• o

; g80o

oo

g

7,am~

o f

i ,

o

o

o

°

13

o

.

~o

o

": o~o~ . ~ i

oo:.o, •o,

ti i

,,

.. ,

o,-

,ao o%

: oo

•

o

,%,

•

I©0

"o

o o¢°

" ' ~ : . °" .'t "

•

o,

o•

•

, .°%" "

,o

0 ~>2

04

O6

rermln:l a,ea , ~

08

IC

i2

r 0

02

i -O4

Termmoi

i

q

06

O8

o,eo .Jm ~'

Fig. 1. Synaptic vesicle density (sv/sq. ffm) as a function of axon terminal area in the right (O) and left ( ) visual cortex of two kittens (C17 and FBP7) following a period of visual stimulation confined to the left hemisphere. In both experiments there is a population of small terminals with high vesicle density after stimulation which is not present on the unstimulated side.

enlargement of x 3. In order to achieve as constant a scale as possible a carbon grating was used for calibration and the groups of micrographs from the same brain were taken in one session whenever possible without changing the magnification. All axon terminals with clear post-synaptic membrane thickenings were selected and their area, synaptic vesicle content and post-synaptic membrane length measured. A marked difference between the two sides was apparent in all experiments: on the stimulated side there were more small, dense ternfinals than on the unstimulated side (Fig. 33 4), and a quantitative study of this appearance was made by comparing the density of synaptic vesicles in the axon terminals of each side. it was clear that the vesicle density in the stimulated hemisphere was consistently higher than in the other. In experiment C17 stimulation was by projection of a vertical grating in the right hemifield of the left eye for 6 h after dark-rearing for 3 weeks. At the end of the exposure the animal was revived and allowed to survive for 8 days bel'ore single unit recording for 12 h and perfusion. All terminals in lavers III and IV of the unstimulated side have a vesicle density of less than 140/sq. um and most of them less than 100/sq. ffm (Fig. 1), but in the stimulated hemisphere there is a population of terminals with densities significantly higher. Most of the dense terminals have an area of less than 0.4 sq. #m, and comparisons of vesicle densities in these small terminals s h o w that the mean density is lower on the unstimulated side. Layers ili and IV are essentially similar in this respect (Table I). The combined results from the small terminals of both layers (Fig. 2) indicate a significant shift in the distribution. The vesicle densitv of

SHORT COMMUNICATIONS

] 67

N= 48

i'it L| N = 43

0

!OZSZ 6 0

87'1~4 ~,~limuloted

~00 5vi,~m~

hernlst~re

2OO

'°l

3OO

0

IO0

2OO

3O0

Lv/um~

Fig. 2. Histograms of synaptic vesicle density in axon terminals. There is a progressive shift in the distribution of vesicle density from the lowest figures (dark reared animals C6, C7,CI0, and unstimulated hemisphere of CI7) to the highest, found following stimulation (FBP7 and the normal animal aged 2 months, K43).

large terminals in layer IV is similarly significantly higher in the stimulated cortex, but the large terminals in layer III do not show this change. In another experimem (FBP7) a different form of visual stimulus was used t3. After an initial period of I month in the dark room the kitten was placed for a total of 39 h in a dome-shaped enclosure, lighttight except for a series of holes in the roof so that the visual environment inside consisted of punctiform lights. The stimulus was reinforced 12 days later by further exposure to both lines and spots in the right hemifield only of both eyes during 3 days of single unit recording. In this way at the termination of the experiment the left visual cortex had had recent stimulation by patterns to which it had been specifically exposed at the height of the critical period, while the right cortex had had no recent stimulation at all. A comparison of the mean densities of synaptic vesicles in axon terminals o f the two sides (Table 1) shows a significantly greater density on the stimulated side in all terminals of layer lII of area 17 and in the small terminals of layer IV. When the results are expressed graphically in terms o f synaptic vesicle density as a function of size of terminal (Fig. 1) it is clear that most terminals in the unstimulated hemisphere have a density of less than 150 vesicles/sq, urn. but that in the stimulated hemisphere many of the endings have greatly higher densities, as in the previous experiment, and that most of these terminals again have an area of less than 0.4 sq. Urn. The relevant data are

168

SHORT COMMUNICATIONS

%

rC

8

~

I~-

:

E

q ~

'T

7

h 4' H E~ ii I~ -I ii ~ tJ ,.m. a',. ,q. ~

,q. ,'--! ~

oR. a'h. r-

-

i

,i ~.

E

!<

~,~,~,~,~o21

>

0

E

<

,4

>o

<

S r~ < Z

o J

Z D < Z )-

>

o ¢b

Fig. 3. Electron micrographs of area 17. - 28,000. 1 : dark reared kitten (C6) with a low density o~ synaptic vesicles (SV) in a typical terminal with an asymmetrical post-synaptic membrane onto t~ dendritic spine (SP). 2: normal kitten (K43) with a relatively high density of vesicles in terminal~ containing both spherical (S) and flattened (F) varieties. 3: unslimulated visual cortex in experiment FBP7. One terminal (A) ends at a symmetrical thickening onto a dendrite and contains a modera!e density of vesicles; another (B) has an asymmetrical thickening onto a dendritic spine and contains only 3 clear vesicles. 4: stimulated cortex of experiment FBP7 showing the uniformly high densit} of vesicles, both spherical (S) and flattened (F).

170

SHORT COMMUNICATIONS

T A B L E II S Y N A P T I C VESICLE DENSITY IN SMALL AXON TERMINALS

Summary of measurements of the actual number of synaptic vesicles (sv) per section of small axon terminal ( < 0.4 sq. /~ln) in layers Ill and 1V of area 17.

Layers 111 and I V

No. o f terminals

Mean sv/small terminal (, 0.4 sq. imp)

% Dtff~'rence

Signi/icam'e

C17

Unstim. Stim. Unstim. Stim. Unstim. Stim. Unstim. Stim. Unstim. Stim. Dark reared

41 48 43 85 45 60 90 77 20 22 76

14.3 23.4 18.5 31.7 14.0 18.1 18.7 20.8 12.2 17.3 13.7

t 63.6

*

t 71.3

*

29.3

***

Normal 2 months

97

21.1 1 1.3

FBP7 Cll C8 C23 C6 C7 C10 K43

~

* P < 0.001; ** P < 0.01; * * * P < 0.05; § P -

J 1.7 ± 2.0 1:1.4 ~ 3.0 ± 1.4 !: 1.4 4 1.3 t 1.6 I 1.7 ± 1.7 t I 1

!11.2 r41.8

*:::*

0.05.

presented in Table I and Fig. 2 and may be compared with those for a normal kitten of similar age (2 months) (K43, Table 1 and Fig. 32) in which the vesicle densities are broadly similar to those in the stimulated terminals of the experimental animal. Experiments are in progress to determine the time course of the changes, both the minimum amount of stinmlation required and the time which must elapse before physiological and anatomical changes are detectable. Experiment C ll illustrates that 1 h after a 30 h period of stimulation of the right hemifield of the left eye, although not significant in a t test, synaptic vesicle density appears to have increased in the small terminals of layer Ill. After longer post-stimulus times of 14 and 15 h (C8, C23, Table 1) the same trend towards higher density in stimulated terminals is more marked. An increased density of vesicles in an axon terminal could be due either to a real increase in vesicles or to a shrinkage of the terminal. In those experiments where a change in vesicle density was observed a similar increase in the number of vesicles per terminal is found (Table II) and, further, there is no consistent difference between the mean terminal areas on the stimulated and unstimulated sides. These results suggest that the phenomenon cannot simply be explained by a reduction in size of terminals but must represent a true proliferation of synaptic vesicles. In view of the fact that some visual experience is unavoidable in setting up the preparation, a litter of 2 kittens (C6 and C7) was raised in the dark for 1 month, then anesthetized in the dark and perfused. A further kitten (C10) was treated similarly,

SHORT COMMUNICATIONS

[7

except that perfusion was delayed for approximately 1 h during which time the normal surgical preparations were carried out, thus providing a control for the effects of these procedures. As the synaptic vesicle densities of all 3 animals were essentially similar they are presented together (Table I). M a n y of the terminals in these animals show extremely low concentrations of vesicles (Fig. 31). The mean synaptic vesicle density is consistently low in all terminals and closely resembles that on the naive side of experiment CI 7, that is cortex which has had virtually no visual experience. The results suggest that there is a population of terminals in the visual cortex of young, visually inexperienced kittens which develops a high concentration of synaptic vesicles after stimulation by patterned input. It is not yet clear whether the dense terminals retain their new characteristics permanently or whether they can reverl to their primitive form. So far this population has been shown to include relatively small terminals, less than 0.4 sq. # m in cross-sectional area, and it consists predominantly of axo-dendritic and axo-spinous endings with asymmetrical thickenings. The reaction o[' large endings is more variable and their appearance may depend on a time factor. More work is needed to further narrow the classification of the axons, and ultimately the cells, involved in the 'learning' process. It has recently been shown 6 that stimulation by sound can cause an increase in synaptic vesicle density in small terminals in the auditory cortex. Exposure of the cortex to specific pattern vision is associated with an increase in synaptic vesicle density in certain neurons after a delay similar to that found before physiological changes were detectable in related experiments l~. A gradation of responsiveness can be found from 'naive' cortex, through the experimentally stimulated cortex in the first few hours after stimulation to the final situation on the second day. In visually inexperienced kittens, firing patterns are vague and broadly tuned 1,14, and axon terminals have low vesicle densities. During the first 10--20 h after stimulation responses become gradually more sharply tuned and the vesicle density increases, and finally in the second day most units behave in a mature fashion and vesicle density reaches a high level. The findings suggest that the remarkable plasticity observed in the neonatal months may be related to changes in efficacy of individual synapses as well as to the rapid growth of neurons 8 and the increase i1~ numbers of synapses 5 which are occurring at this time. We should like to thank H. B. Barlow for help and encouragement during this work which was supported by Grant EY00276 from the U.S. Public Health Service and a Medical Research Council Travelling Fellowship to L. J.G. The collaboration of R. D. Freeman and H. V. B. Hirsch and the technical assistance of Barbara Sternitzke are gratefully acknowledged.

1 BARLOW, H. B., AND PETTIGREW, J. D., Lack of specificity of neurones in the visual cortex of

young kittens, J. Physiol. (Lond.), 218 (1971) 98-100P. 2 BLAKEMORE,C., AND COOPER, G. F., Development of the brain depends on visual environmenL

Nature (Lond.), 228 (1970) 477-478. 3 BLAKEMORE,C., AND MITCHELL, O. E., Environmental modification of the visual cortex and the

neural basis of learning and memory, Nature (Lond:); 241 (1973) 467-468.

172

SHORT COMMUNICATIONS

4 CRAGG, B. G., Changes in visual cortex on first experience of rats to light, Nature (LomL), 215 (1967) 251 253. 5 CRAGG, B. G., The development of synapses in cat visual cortex, Invest. Ophthal., 1 I 11972) 377 385. 6 FEHI~R,O., FERENC,J., AND HAL,~SZ,N., Effect of stimulation on the number ofsynaptic vesicles in nerve fibres and terminals of the cerebral cortex in the cat, Brain Research, 47 (1972) 37 48. 7 FIFKOVX,E., The effect of monocular deprivation on the synaptic contacts of the visual cortex, J. Neurobiol., 1 (1970)285-294. 8 GAREY, L. J., FISKEN, R. A., AND POWELL, T. P. S., Observations on the growth of cells in the lateral geniculate nucleus of the cat, Brain Research, 52 (1973) 359 362. 9 HIRSf'H, H. V. B., AND SPINELLI, D. N., Visual experience modifies distribution of horizontall? and vertically oriented receptive fields in cats, Science, 168 (1970) 869-871. 10 HIRSCH, H. V. B., AND SPINELLI, D. N., Modification of the distribution of receptive [icld orientation in cats by selective visual exposure during development, Exp. Brain Res., 12 (1971) 509 527. I 1 HUBEI_,D. H., AND WJESEL, T. N., The period of susceptibility to the physiological effects of unilateral eye closure in kittens, J. Physiol. (Lond.), 206 (1970) 419 436. 12 MOLLGAARD, K., DIAMOND, U. C., BENNETT, E. L., ROSENZWEIG, M. R., AN[) LINI)NIR, B., Quantitative synaptic changes with differential experience in rat brain, Int. J. Nearosci., 2 (1971) 113 128. 13 PETTIGREW, J. D., AND FREEMAN, R. D., Visual experience without lines: effect on developing cortical neurons, Science, 182 (1973) 599 601. 14 PET'II'dREW,J. D., AND GAREY, L. J., Selective modification of single neuron properties in the visual cortex of kittens, Brain Research, 66 (1974) 160-164. 15 PI~TTIGREW,J. D., OLSON, C., AND HIRSCH, H. V. B., Cortical cffect of selective visual experience: degeneration or reorganization? Brain Research, 51 (1973) 345-351.