Brain Research, 119 (1977) 73-86
73
© Elsevier/North-HollandBiomedicalPress, Amsterdam- Printed in The Netherlands
EFFECTS OF VISUAL DEPRIVATION ON POLYRIBOSOME AGGREGATION IN VISUAL CORTEX OF THE CAT
SIMON LeVAY Department o f Neurobiology, Harvard Medical School, Boston, Mass. 02115 (U.S.A.)
(Accepted May 3rd, 1976)
SUMMARY Neurons in the visual cortex of 48 normal and visually deprived kittens and cats were examined electron microscopically for the presence or absence of polyribosomes in their perikaryal cytoplasm. In normal animals at most ages the ribosomes of cortical neurons were aggregated into polysomes, but during the second and third months of life - - a period corresponding approximately to the physiologically defined critical period - - variable numbers of cells were found which contained ribosomes only in the monomeric form. The affected cells were spiny stellate neurons in the fourth layer of the cortex. Even within the critical period, however, cells with dispersed ribosomes were not found in every animal examined. In animals deprived of pattern vision in both eyes from birth the appearance of cells lacking polysomes at the beginning of the second month was more consistent. By 7 weeks half of the fourth-layer neurons were affected, and they remained in this condition for at least another 6 months. The dispersed ribosomes could be caused to reaggregate by prolonged pentobarbital anesthesia. In monocularly deprived animals the binocular segments of cortex showed little if any effects of deprivation, but in one kitten in which the two monocular segments were examined cells lacking polysomes were found only in the segment innervated from the closed eye. It was concluded that a short-lived polysomal disaggregation occurs during the critical period - - perhaps at the time of completion of synaptogenesis - - and that the subsequent reaggregation is dependent on a normal visual input. Comparison with the results of biochemical studies suggests that the polysomal disaggregation reflects a deficit in the synthesis of axonally transported proteins.
INTRODUCTION Visual deprivation during early life severely affects the metabolic, physiological
74 and structural development of the visual cortex, in animals with binocular vision, deprivation of vision in one eye leads to a physiological dominance of the experienced eye1,z0, which in the monkey, at least, has an anatomical basis in an abnormally extensive arborization of geniculocortical afferents serving that eye, and a loss ot cortical terrain by the deprived eye 15. An active inhibition of responses to the deprived eye may also be involved 7,1v. In binocular deprivation there is naturally no disturbance of the balance of the two eyes in the cortex, but there is a deterioration of responsiveness to visual stimulation which affects many, but not all cortical neurons31,zL The density of synaptic connections is abnormally low in the cortex of binocularly deprived cats 4 and in sightless or visually deprived rats major disturbances of RNA and protein metabolism have been demonstrated z,z6. Since the biochemical work has generally involved homogenization of the brain as an initial step, it has been hard to relate these findings to those of the anatomical and electrophysiological studies. During an examination of EM sections from the cortex of deprived kittens, however, large numbers of neurons were noticed which lacked polyribosomes, organelles which are normally prominent in the cytoplasm of nerve cells and indeed of nearly all cells actively engaged in protein synthesis. The occurrence of these cells seemed to offer an unusual opportunity to localize a metabolic effect of deprivation to a particular cell type whose connections and functional state may be studied. The identity of the affected cells, and the time course of their appearance, was therefore examined in greater detail. Particular attention was paid to their numbers in normaJ kittens of various ages, since Palay et al. 24 have described such cells in the visual cortex of normal monkeys. METHODS Observations were made on 48 kittens and cats, ranging in age from newborn to 5 years. Animals with Siamese coloring were excluded. Most of the animals were born and raised in our colony. Twenty-two of these animals were reared with normal vision, 17 with both eyes closed, and 9 with one eye closed. The eye closures were produced by suturing the eyelids together under halothane anesthesia during the second week of life. The animals were killed by perfusion with a mixture of glutaraldehyde (2 %), paraformaldehyde (2%) and sucrose (2%) in 0.1 M phosphate buffer at pH 7.3. All perfusions were performed in mid-morning. Except when the effects of prolonged anesthesia were being investigated, the perfusion was begun within 10-15 min of induction of anesthesia (sodium pentobarbital, 35 mg/kg i.p.). In order not to interfere with respiration during preparation, the perfusion was performed through the abdominal aorta. Blocks of cortex were taken from the suprasplenial sulcus in the intraaural plane, i.e. from the part of the primary visual cortex (area 17) which is binocularly innervated. In the monocularly sutured animals the cortex contralateral to the closed eye was examined. The tissue blocks were postosmicated, blocked-stained with uranyl acetate and phosphotungstic acid, dehydrated and embedded in Epon. Ultrathin sections were cut perpendicular to the pial surface, mounted on one-hole grids and
75 stained with lead citrate. The layers were identified by comparison with adjacent 1 #m thick sections stained with toluidine blue for light microscopy. In some normal and deprived animals blocks from area 18, somatic sensory cortex, auditory cortex, lateral geniculate nucleus and retina were also examined. Observations were also made on the primary visual cortex of some other species, including macaque monkeys, rats, mice and rabbits. RESULTS
Cytology The ultrastructural appearance of neurons with and without polyribosomes is illustrated in Fig. 1A and lB. These cells are both taken from the fourth layer of the primary visual cortex (area 17) of a 6-month-old kitten whose eyes had been closed by lid suture at 10 days of age. In the cells lacking polysomes there was no obvious change in the total numbers of ribosomes present, but they were dispersed as single particles. It was not possible to determine whether these particles represented monomeric ribosomes, ribosomal subunits, or a mixture of these. The change affected membrane-bound and cytoplasmic ribosomes equally, and there did not seem to be any great change in the proportion of ribosomes which were membrane-bound. On close inspection, one or two polyribosomes could usually be found in a single section. Occasionally, cells were encountered, most of whose ribosomes were disaggregated, but which also contained an appreciable number of polysomes. These cells, arbitrarily defined as containing at least 6 recognizable polysomes in a single section, were termed intermediate and were counted separately. The endoplasmic reticulum of the cells lacking polysomes was also unusual, and its appearance varied with the animal's age. In 1- to 2-month-old kittens it generally took the form of long rambling cisternae, usually running singly through the cytoplasm rather than in parallel stacks. Sometimes many of the cisternae were oriented radially in the cell instead of in parallel with the nuclear membrane, which is the more usual configuration. At later stages (Fig. lB) the endoplasmic reticulum was broken up into short, randomly-oriented cisternae. These changes in the endoplasmic reticulum were not completely consistent, however; occasional cells with a normal complement of polyribosomes showed endoplasmic reticulum in a fragmented state, and some cells lacking polysomes contained endoplasmic reticulum in a more or less normal configuration. The nucleolus of the cells lacking polysomes had a normal appearance, as did the Golgi apparatus, mitochondria, lysosomes and microtubules. The dictyosomes of the Golgi apparatus were in a normally active state, judging by the characteristic narrowing of the cisternae towards the mature face of the dictyosome. In older animals the elements of the Golgi apparatus and particularly of the vesicular and tubular elements of smooth endoplasmic reticulum, which did not form an integral part of Golgi fields, were somewhat reduced.
Fig. 1. Cytoplasm of two neurons showing (A) aggregation of ribosomes into polysomes and (B) ribosomes in dispersed form. Both micrographs were taken from the fourth layer of area 17 in a binocularly deprived, 6-month-old kitten. Note that the disaggregation in (B) affects both the free cytoplasmic ribosomes and those attached to the cndoplasmic retieulum. The arrangement of the membrane-bound ribosomes may be seen most clearly at points where the ER is sectioned tangentially (arrows). To the left side of Fig. IB many of the disaggregated ribosomes are associated with clumps of amorphous granular material. The complete lack of elements of the Golgi apparatus in B is accidental, n, nucleus. Bar I ,,m.
?7
o
c
o
o
o
o
o o o
,~0
o
o
o
o
o o
o
c,
•
4a+b
o
o
o
o
o
o
•
.....
o
o
~ ........................
4c •
0
o
o
~)
•
o
o o o o
c, o
o
5
o o
o o
°o
o
o
c,
o o
c, o o
Fig. 2. Reconstruction of a single EM section passing through layers 3-5 of area 17 in a 3-month-old binocularly deprived kitten, showing the position of every neuron as recorded from the EM stage micrometers. Cells with a normal complement of polyribosomes are indicated by open circles, cells lacking polysomes by filled circles, and cells of intermediate appearance by half-filled circles. The cells lacking polysomes are restricted to layer 4 and the lowest part of layer 3. They form a somewhat higher proportion of the total cells in 4c than in 4a ÷ b. Note the scarcity of intermediate cells.
Laminar distribution The great majority of cells lacking polysomes were found in the fourth layer of the cortex. Their laminar distribution was examined in greater detail by mapping the position of all cells with normal and disaggregated polysomes found in a section passing perpendicularly through the third, fourth and fifth layers. Such a reconstruction for a 3-month-old, binocularly deprived kitten is shown in Fig. 2. It was typical of such reconstructions that the cells lacking polysomes were slightly more common in the lower part of the fourth layer (layer IVc of Otsuka and Hassler 23) than in the upper part (called layer IVa + b because of a supposed equivalence to the two sublayers IVa and IVb found in primates). Overall, am•st half the cells in the fourth layer in this animal
78 lacked polysomes. A few such cells were found in the third layer, usually close to its border with layer IV, but only rarely were they found in layers 11, V or VI.
Cell type and synaptology The restriction of the affected cells to the fourth layer indicated that they were predominantly stellate rather than pyramidal neurons, for pyramids are scarce in this layer, particularly in its lower part, layer lVc 18. Stellate cells have been further subdivided into those whose dendrites carry numerous spines and those which have few or no dendritic spines ls,2°,21. These two cell types differ both in their laminar distribution - - spine-free stellates occur throughout the cortex, whereas spiny stellates are restricted to the fourth layer - - and in their synaptic arrangements. Particularly relevant to the problem of identifying these cell types in EM sections is the observation 18 that the axosomatic synapses on spiny stellate cells are exclusively of Gray's type 22,12, whereas spine-free stellate cells carry axosomatic synapses both of this type and of Gray's type 1. In the present study it was consistently found that the cells lacking polysomes carried axosomatic synapses of Gray's type 2 only. To illustrate this relationship, Table I shows the distribution of type 1 synapses of 57 cells examined in a single section from the fourth layer of a 6-month-old, binocularly deprived kitten. It was concluded that the affected cells were spiny stellate cells, and that the spine-free stellate cells were not susceptible to ribosomal disaggregation. Since the synaptic arrangements of pyramidal cells are similar to those of spiny stellate cells is, it is possible that pyramids, which occur in limited numbers in layer IVa and b, may also be affected. For the most part though cells with recognizable apical dendrites did not show disaggregation. The spiny stellate cells carry type 1 synapses on their dendritic spines is and this population of synapses includes many of those formed by geniculocortical afferentslL While it was not possible to trace individual dendrites from the somata of cells lacking polysomes to the loci of axospinous synapses, this type of synapse was present in large numbers in the neuropil of all animals examined, and showed no obvious abnormality in the deprived animals (but see Discussion).
Relationship of disaggregation to age and deprivation The time course of the appearance of ribosomal disaggregation, and its relationTABLE 1
Relationship between ribosomal aggregation and synapse distribution on perikaryon All 57 cells in a single section through the fourth layer in a 6-month-old, binocularly deprived kitten were examined for presence or absence of polysomes and for possession of Gray's type I axosomatic synapses. The lack of type 1 synapses on the perikarya of cells lacking polysomes indicates that these cells are spiny stellate cells (see text). This result is typical of all animals examined.
Type 1 synapses present Type 1 synapses absent
Polysomes present
Polysomes absent
Intermediate
12 16
0 27
0 2
79 SUTURE
BINOCULAR
III |
5b
,~o
I~o
225
2yr
I00. MONOCULAR
SUTURE
.90 .~ so-
@ @
5b
,6o
130
...~- . - , ~ .....~ 1 yr
3.Syr
5 yr
100-
NORMAL
50
g -
~
~
-
Bit
50
'
'~o
",5o
--: o1,i
-age (days)
Fig. 3. Time course of ribosomal disaggregation in binocularly deprived, monoculady deprived and normal cats. Each point represents for a single animal the percentage of neurons in layer 4 showing disaggregation (mean sample: 76 cells) against the animal's age at death. Eyelid sutures were performed in the second week of life, i.e. close to the time of normal eye opening. The exact ages of the adult normal cats was not known, but they were estimated to range from 1 to 5 years.
ship to deprivation, was examined in 21 normal kittens and cats ranging in age from newborn to adult, 13 binocularly deprived animals and 9 which were monocularly deprived. The deprivation was produced in all cases by suturing the eyelids together in the second week of life, i.e. close to the time of normal eye opening, and leaving the eye or eyes sutured until the animals were killed. In sections of cortex similar to that mapped in Fig. 2, every cell in the fourth layer was examined for the presence or absence of polysomes. The results are tabulated in Fig. 3, in which the proportion of cells lacking polysomes is shown in relation to age for each of the three groups of animals. The intermediate cells are not shown separately but are included with the normal cells; they did not exceed 1 ~ of the total cells except in one 3-month-old normal kitten in which they comprise 18 ~ . Because prolonged general anesthesia had
80 a dramatic effect on the number of cells showing disaggregation (see below), Fig. 3 only includes data for animals which were perfused within 10-15 min after induction of anesthesia. In the normal animals all neurons contained polysomes between birth and one month of age. At later times cells showing disaggregation were found in some animals but not in others. The great variability in the number of these cells is particularly obvious among the 5 normal kittens perfused between 53 and 59 days of age: one kitten had none, one had 50 ~ of the cells in the fourth layer affected, and the other three showed intermediate numbers. One apparently normal animal perfused at 10 weeks of age had an exceptionally large number of cells showing disaggregation, 78 ~. This figure was even higher than that found in any of the deprived animals. Older normal animals had much lower numbers of these cells - - of 6 adult cats examined, 5 had no cells lacking polysomes and one had 10 ~. In the binocularly deprived animals, as in the normals, no cells lacking polysomes were found before the end of the first month of life. In the following two weeks the proportion of affected cells rose rapidly and consistently to 50 ~ of the cell population in the fourth layer. This proportion then remained stable until at least 7 months of age, and only in the longest-surviving animal - - a 2-year-old adult - - was there a decline in their numbers. Taking the period between 40 and 100 days of age, within which the normal group of animals seemed most susceptible to ribosomal disaggregation, the mean number of cells lacking polysomes was twice as great in the binocularly deprived group (46 ~ against 22 ~ , P < 0.05, unpaired Wilcoxon test) and the variance was significantly less (P < 0.01, F-test). The significance of the difference between the two groups increases if older animals are included; thus for the age range 40-225 days P
73
© Elsevier/North-HollandBiomedicalPress, Amsterdam- Printed in The Netherlands
EFFECTS OF VISUAL DEPRIVATION ON POLYRIBOSOME AGGREGATION IN VISUAL CORTEX OF THE CAT
SIMON LeVAY Department o f Neurobiology, Harvard Medical School, Boston, Mass. 02115 (U.S.A.)
(Accepted May 3rd, 1976)
SUMMARY Neurons in the visual cortex of 48 normal and visually deprived kittens and cats were examined electron microscopically for the presence or absence of polyribosomes in their perikaryal cytoplasm. In normal animals at most ages the ribosomes of cortical neurons were aggregated into polysomes, but during the second and third months of life - - a period corresponding approximately to the physiologically defined critical period - - variable numbers of cells were found which contained ribosomes only in the monomeric form. The affected cells were spiny stellate neurons in the fourth layer of the cortex. Even within the critical period, however, cells with dispersed ribosomes were not found in every animal examined. In animals deprived of pattern vision in both eyes from birth the appearance of cells lacking polysomes at the beginning of the second month was more consistent. By 7 weeks half of the fourth-layer neurons were affected, and they remained in this condition for at least another 6 months. The dispersed ribosomes could be caused to reaggregate by prolonged pentobarbital anesthesia. In monocularly deprived animals the binocular segments of cortex showed little if any effects of deprivation, but in one kitten in which the two monocular segments were examined cells lacking polysomes were found only in the segment innervated from the closed eye. It was concluded that a short-lived polysomal disaggregation occurs during the critical period - - perhaps at the time of completion of synaptogenesis - - and that the subsequent reaggregation is dependent on a normal visual input. Comparison with the results of biochemical studies suggests that the polysomal disaggregation reflects a deficit in the synthesis of axonally transported proteins.
INTRODUCTION Visual deprivation during early life severely affects the metabolic, physiological
74 and structural development of the visual cortex, in animals with binocular vision, deprivation of vision in one eye leads to a physiological dominance of the experienced eye1,z0, which in the monkey, at least, has an anatomical basis in an abnormally extensive arborization of geniculocortical afferents serving that eye, and a loss ot cortical terrain by the deprived eye 15. An active inhibition of responses to the deprived eye may also be involved 7,1v. In binocular deprivation there is naturally no disturbance of the balance of the two eyes in the cortex, but there is a deterioration of responsiveness to visual stimulation which affects many, but not all cortical neurons31,zL The density of synaptic connections is abnormally low in the cortex of binocularly deprived cats 4 and in sightless or visually deprived rats major disturbances of RNA and protein metabolism have been demonstrated z,z6. Since the biochemical work has generally involved homogenization of the brain as an initial step, it has been hard to relate these findings to those of the anatomical and electrophysiological studies. During an examination of EM sections from the cortex of deprived kittens, however, large numbers of neurons were noticed which lacked polyribosomes, organelles which are normally prominent in the cytoplasm of nerve cells and indeed of nearly all cells actively engaged in protein synthesis. The occurrence of these cells seemed to offer an unusual opportunity to localize a metabolic effect of deprivation to a particular cell type whose connections and functional state may be studied. The identity of the affected cells, and the time course of their appearance, was therefore examined in greater detail. Particular attention was paid to their numbers in normaJ kittens of various ages, since Palay et al. 24 have described such cells in the visual cortex of normal monkeys. METHODS Observations were made on 48 kittens and cats, ranging in age from newborn to 5 years. Animals with Siamese coloring were excluded. Most of the animals were born and raised in our colony. Twenty-two of these animals were reared with normal vision, 17 with both eyes closed, and 9 with one eye closed. The eye closures were produced by suturing the eyelids together under halothane anesthesia during the second week of life. The animals were killed by perfusion with a mixture of glutaraldehyde (2 %), paraformaldehyde (2%) and sucrose (2%) in 0.1 M phosphate buffer at pH 7.3. All perfusions were performed in mid-morning. Except when the effects of prolonged anesthesia were being investigated, the perfusion was begun within 10-15 min of induction of anesthesia (sodium pentobarbital, 35 mg/kg i.p.). In order not to interfere with respiration during preparation, the perfusion was performed through the abdominal aorta. Blocks of cortex were taken from the suprasplenial sulcus in the intraaural plane, i.e. from the part of the primary visual cortex (area 17) which is binocularly innervated. In the monocularly sutured animals the cortex contralateral to the closed eye was examined. The tissue blocks were postosmicated, blocked-stained with uranyl acetate and phosphotungstic acid, dehydrated and embedded in Epon. Ultrathin sections were cut perpendicular to the pial surface, mounted on one-hole grids and
75 stained with lead citrate. The layers were identified by comparison with adjacent 1 #m thick sections stained with toluidine blue for light microscopy. In some normal and deprived animals blocks from area 18, somatic sensory cortex, auditory cortex, lateral geniculate nucleus and retina were also examined. Observations were also made on the primary visual cortex of some other species, including macaque monkeys, rats, mice and rabbits. RESULTS
Cytology The ultrastructural appearance of neurons with and without polyribosomes is illustrated in Fig. 1A and lB. These cells are both taken from the fourth layer of the primary visual cortex (area 17) of a 6-month-old kitten whose eyes had been closed by lid suture at 10 days of age. In the cells lacking polysomes there was no obvious change in the total numbers of ribosomes present, but they were dispersed as single particles. It was not possible to determine whether these particles represented monomeric ribosomes, ribosomal subunits, or a mixture of these. The change affected membrane-bound and cytoplasmic ribosomes equally, and there did not seem to be any great change in the proportion of ribosomes which were membrane-bound. On close inspection, one or two polyribosomes could usually be found in a single section. Occasionally, cells were encountered, most of whose ribosomes were disaggregated, but which also contained an appreciable number of polysomes. These cells, arbitrarily defined as containing at least 6 recognizable polysomes in a single section, were termed intermediate and were counted separately. The endoplasmic reticulum of the cells lacking polysomes was also unusual, and its appearance varied with the animal's age. In 1- to 2-month-old kittens it generally took the form of long rambling cisternae, usually running singly through the cytoplasm rather than in parallel stacks. Sometimes many of the cisternae were oriented radially in the cell instead of in parallel with the nuclear membrane, which is the more usual configuration. At later stages (Fig. lB) the endoplasmic reticulum was broken up into short, randomly-oriented cisternae. These changes in the endoplasmic reticulum were not completely consistent, however; occasional cells with a normal complement of polyribosomes showed endoplasmic reticulum in a fragmented state, and some cells lacking polysomes contained endoplasmic reticulum in a more or less normal configuration. The nucleolus of the cells lacking polysomes had a normal appearance, as did the Golgi apparatus, mitochondria, lysosomes and microtubules. The dictyosomes of the Golgi apparatus were in a normally active state, judging by the characteristic narrowing of the cisternae towards the mature face of the dictyosome. In older animals the elements of the Golgi apparatus and particularly of the vesicular and tubular elements of smooth endoplasmic reticulum, which did not form an integral part of Golgi fields, were somewhat reduced.
Fig. 1. Cytoplasm of two neurons showing (A) aggregation of ribosomes into polysomes and (B) ribosomes in dispersed form. Both micrographs were taken from the fourth layer of area 17 in a binocularly deprived, 6-month-old kitten. Note that the disaggregation in (B) affects both the free cytoplasmic ribosomes and those attached to the cndoplasmic retieulum. The arrangement of the membrane-bound ribosomes may be seen most clearly at points where the ER is sectioned tangentially (arrows). To the left side of Fig. IB many of the disaggregated ribosomes are associated with clumps of amorphous granular material. The complete lack of elements of the Golgi apparatus in B is accidental, n, nucleus. Bar I ,,m.
?7
o
c
o
o
o
o
o o o
,~0
o
o
o
o
o o
o
c,
•
4a+b
o
o
o
o
o
o
•
.....
o
o
~ ........................
4c •
0
o
o
~)
•
o
o o o o
c, o
o
5
o o
o o
°o
o
o
c,
o o
c, o o
Fig. 2. Reconstruction of a single EM section passing through layers 3-5 of area 17 in a 3-month-old binocularly deprived kitten, showing the position of every neuron as recorded from the EM stage micrometers. Cells with a normal complement of polyribosomes are indicated by open circles, cells lacking polysomes by filled circles, and cells of intermediate appearance by half-filled circles. The cells lacking polysomes are restricted to layer 4 and the lowest part of layer 3. They form a somewhat higher proportion of the total cells in 4c than in 4a ÷ b. Note the scarcity of intermediate cells.
Laminar distribution The great majority of cells lacking polysomes were found in the fourth layer of the cortex. Their laminar distribution was examined in greater detail by mapping the position of all cells with normal and disaggregated polysomes found in a section passing perpendicularly through the third, fourth and fifth layers. Such a reconstruction for a 3-month-old, binocularly deprived kitten is shown in Fig. 2. It was typical of such reconstructions that the cells lacking polysomes were slightly more common in the lower part of the fourth layer (layer IVc of Otsuka and Hassler 23) than in the upper part (called layer IVa + b because of a supposed equivalence to the two sublayers IVa and IVb found in primates). Overall, am•st half the cells in the fourth layer in this animal
78 lacked polysomes. A few such cells were found in the third layer, usually close to its border with layer IV, but only rarely were they found in layers 11, V or VI.
Cell type and synaptology The restriction of the affected cells to the fourth layer indicated that they were predominantly stellate rather than pyramidal neurons, for pyramids are scarce in this layer, particularly in its lower part, layer lVc 18. Stellate cells have been further subdivided into those whose dendrites carry numerous spines and those which have few or no dendritic spines ls,2°,21. These two cell types differ both in their laminar distribution - - spine-free stellates occur throughout the cortex, whereas spiny stellates are restricted to the fourth layer - - and in their synaptic arrangements. Particularly relevant to the problem of identifying these cell types in EM sections is the observation 18 that the axosomatic synapses on spiny stellate cells are exclusively of Gray's type 22,12, whereas spine-free stellate cells carry axosomatic synapses both of this type and of Gray's type 1. In the present study it was consistently found that the cells lacking polysomes carried axosomatic synapses of Gray's type 2 only. To illustrate this relationship, Table I shows the distribution of type 1 synapses of 57 cells examined in a single section from the fourth layer of a 6-month-old, binocularly deprived kitten. It was concluded that the affected cells were spiny stellate cells, and that the spine-free stellate cells were not susceptible to ribosomal disaggregation. Since the synaptic arrangements of pyramidal cells are similar to those of spiny stellate cells is, it is possible that pyramids, which occur in limited numbers in layer IVa and b, may also be affected. For the most part though cells with recognizable apical dendrites did not show disaggregation. The spiny stellate cells carry type 1 synapses on their dendritic spines is and this population of synapses includes many of those formed by geniculocortical afferentslL While it was not possible to trace individual dendrites from the somata of cells lacking polysomes to the loci of axospinous synapses, this type of synapse was present in large numbers in the neuropil of all animals examined, and showed no obvious abnormality in the deprived animals (but see Discussion).
Relationship of disaggregation to age and deprivation The time course of the appearance of ribosomal disaggregation, and its relationTABLE 1
Relationship between ribosomal aggregation and synapse distribution on perikaryon All 57 cells in a single section through the fourth layer in a 6-month-old, binocularly deprived kitten were examined for presence or absence of polysomes and for possession of Gray's type I axosomatic synapses. The lack of type 1 synapses on the perikarya of cells lacking polysomes indicates that these cells are spiny stellate cells (see text). This result is typical of all animals examined.
Type 1 synapses present Type 1 synapses absent
Polysomes present
Polysomes absent
Intermediate
12 16
0 27
0 2
79 SUTURE
BINOCULAR
III |
5b
,~o
I~o
225
2yr
I00. MONOCULAR
SUTURE
.90 .~ so-
@ @
5b
,6o
130
...~- . - , ~ .....~ 1 yr
3.Syr
5 yr
100-
NORMAL
50
g -
~
~
-
Bit
50
'
'~o
",5o
--: o1,i
-age (days)
Fig. 3. Time course of ribosomal disaggregation in binocularly deprived, monoculady deprived and normal cats. Each point represents for a single animal the percentage of neurons in layer 4 showing disaggregation (mean sample: 76 cells) against the animal's age at death. Eyelid sutures were performed in the second week of life, i.e. close to the time of normal eye opening. The exact ages of the adult normal cats was not known, but they were estimated to range from 1 to 5 years.
ship to deprivation, was examined in 21 normal kittens and cats ranging in age from newborn to adult, 13 binocularly deprived animals and 9 which were monocularly deprived. The deprivation was produced in all cases by suturing the eyelids together in the second week of life, i.e. close to the time of normal eye opening, and leaving the eye or eyes sutured until the animals were killed. In sections of cortex similar to that mapped in Fig. 2, every cell in the fourth layer was examined for the presence or absence of polysomes. The results are tabulated in Fig. 3, in which the proportion of cells lacking polysomes is shown in relation to age for each of the three groups of animals. The intermediate cells are not shown separately but are included with the normal cells; they did not exceed 1 ~ of the total cells except in one 3-month-old normal kitten in which they comprise 18 ~ . Because prolonged general anesthesia had
80 a dramatic effect on the number of cells showing disaggregation (see below), Fig. 3 only includes data for animals which were perfused within 10-15 min after induction of anesthesia. In the normal animals all neurons contained polysomes between birth and one month of age. At later times cells showing disaggregation were found in some animals but not in others. The great variability in the number of these cells is particularly obvious among the 5 normal kittens perfused between 53 and 59 days of age: one kitten had none, one had 50 ~ of the cells in the fourth layer affected, and the other three showed intermediate numbers. One apparently normal animal perfused at 10 weeks of age had an exceptionally large number of cells showing disaggregation, 78 ~. This figure was even higher than that found in any of the deprived animals. Older normal animals had much lower numbers of these cells - - of 6 adult cats examined, 5 had no cells lacking polysomes and one had 10 ~. In the binocularly deprived animals, as in the normals, no cells lacking polysomes were found before the end of the first month of life. In the following two weeks the proportion of affected cells rose rapidly and consistently to 50 ~ of the cell population in the fourth layer. This proportion then remained stable until at least 7 months of age, and only in the longest-surviving animal - - a 2-year-old adult - - was there a decline in their numbers. Taking the period between 40 and 100 days of age, within which the normal group of animals seemed most susceptible to ribosomal disaggregation, the mean number of cells lacking polysomes was twice as great in the binocularly deprived group (46 ~ against 22 ~ , P < 0.05, unpaired Wilcoxon test) and the variance was significantly less (P < 0.01, F-test). The significance of the difference between the two groups increases if older animals are included; thus for the age range 40-225 days P
