The effects of protein deprivation on the nucleus raphe dorsalis: A morphometric golgi study in rats of three age groups

Brain Research, 221 (1981) 243-255 Elsevier/North-Holland Biomedical Press

243

T H E E F F E C T S OF P R O T E I N D E P R I V A T I O N O N T H E N U C L E U S R A P H E D O R S A L I S : A M O R P H O M E T R I C G O L G I S T U D Y I N RATS OF T H R E E A G E GROUPS

SOFIA D[AZ-CINTRA*, LEON CINTRA*, THOMAS KEMPER, OSCAR RESNICK and PETER J. MORGANE** Worcester Foundation for Experimental Biology, Shrewsbury, and Neurological Unit, Boston City Hospital, Boston, Mass. (U.S.A.)

(Accepted February 12th, 1981) Key words: nucleus raphe dorsalis - - raphe dorsalis cells - - Golgi morphometric study - - undernutrition - - age-related changes and prenatal undernutrition - - dendritic spine density

SUMMARY In a previous study we identified 3 cell types in the nucleus raphe dorsalis ( N R D ) : fusiform, multipolar and ovoid. In the present study, we have investigated the effect of an 8 % casein diet on these 3 cell types using quantitative techniques on rapid Golgiimpregnated neurons from rats of 3 different ages: 30, 90 and 220 days. Major and minor axes of the cell body and dendritic diameter were unaffected and primary dendritic linear extent was only slightly affected by the diet. All 3 cell types in control rats showed an increase in synaptic spines on both primary and secondary dendrites between 30 and 90 days followed by a decrease for all 3 of the cell types at 220 days. Protein-deprived rats failed to show these age-related changes. Other parameters of comparison showed clear differences between the 3 cell types. These differences could be readily seen when total synaptic spine input to the primary and secondary dendrites was calculated from the data on dendritic number, linear extent and spine density. When viewed in this way the fusiform and ovoid cells show either little change or a decreased synaptic input at all ages, while the presumed serotonergic multipolar cells showed an increase. This is in agreement with neurochemical studies in these rats showing increased levels of this biogenic amine in protein malnourished rats.

*S. Diaz-Cintra and L. Cintra on leave from Depto. de Fisiologia, Instituto de Investigaciones Biom6dicas, UNAM, Universidad Nacional Aut6noma de M6xico, Ciudad Universitaria, M6xico, 20, D.F. **To whom reprint requests should be mailed.

244 INTRODUCTION Since the classic studies of Sugita 4° and Stewart 39 there have been numerous investigations of the effects of undernutrition on the developing rodent brain. The most marked effects noted have been when nutritional deprivation includes the period of lactation, the so-called critical period of rodent brain development 15. In these studies the cerebellum has shown the most striking effects. The cerebellar cortex has been found to be thin, particularly the molecular layer1,3,9,31, 3s and its transient germinal external granule cell layer delayed in its time of resolution 3,25,38. Purkinje cells, though not decreased in number 9, showed an increased cell packing densityS, as, decreased size of the cell body 9 and extent of their dendritic territories28,34, 3s, decreased total length of dendritic processes28, 34, alterations in the branching of the dendritic tree zl, decreased total number of dendritic spines 34, and a lag in the timetable of resolution of perisomatic processes 9. Granule cell neurons have been reported to show a deficit in number 4,9,14, either increased28, 31, or unchanged cell packing density 4, decreased neuronal cell size 9, and decreased linear extent, but not number, of dendrites 2s,3s. Basket cells have been reported to show a decrease in total number 9 and in length and spine density of their dendrites 38. The cerebral cortex has been reported in malnutrition to show a less striking effect with superficial cell layers more affected than deep cell layerse, 26,27. Cortical width is decreased2,~,7,8,14,37,4°, 41, neuronal cell packing density increased 7,1°,2°,27,39, and total neuronal number was probably unaffected z4,~s. Neuronal cell size has been reported to be either unaffected 20 or decreased z,7,1° and dendritic diameter and length decreased 2,36,4°. In Golgi studies synaptic spine density has been reported to be decreased26,~s, 41, and in EM studies decreased 16,2° or unaffected 11. Subcortical and brain stem formations have been reported to be much less affected by undernutrition~9,25, 35. In the lateral vestibular nucleus, neuronal cell packing density was noted to be increased but cell size unaffected 22. In the neostriatum and nucleus of the diagonal band of Broca dendritic length was unaffected and spine density only slightly decreased in the neostriatum 25. In the olfactory bulb, cell size and dendritic diameter were unaffected while dendritic number was decreased 35. In our studies on the effect of undernutrition on the developing rat brain we have used a model in which female rats are undernourished prior to pregnancy and undernutrition continued through gestation until the animals are sacrificed. One of the most striking findings in our model has been a marked increase of the biogenic amines (serotonin and norepinephrine) in several brain regions. This extended from our first sacrifice time at birth until adulthood (Morgane et al. 3° and Resnick et al.3~). No previous morphological studies of the effects of undernutrition have been done in the aminergic nuclei of the brain stem. Therefore, in the present study we have investigated the effects of protein undernutrition on one of the major serotonergic nuclei of the brain stem, the nucleus raphe dorsalis. We have used criteria for cell types and methodologies established in our prior study of the normal anatomy of this nucleus (Diaz-Cintra et al.13). In this study we recognized, based on cell shape, size, dendritic architecture, age-related changes, and site of origin of the axon, 3 cell types, fusiform,

245 multipolar and ovoid. Since effects of undernutrition on subcortical formations have been reported to be less marked than on cortical formations, and the effect of combined undernutrition during gestation and lactation found to be less marked than during lactation alone (Leuba and Rabinowicz z~,27), we applied a wide variety of parameters of morphometric analysis to the nucleus raphe dorsalis. MATERIALS AND METHODS Virgin female Charles River CD Sprague-Dawley descended rats were placed on either an 8 ~ or 25 ~ casein diet 5 weeks prior to conception and then maintained on these diets during gestation and lactation in accordance with the paradigm described by Morgane et al. a0. Litter size was kept constant from the day of birth at 8 male pups. The pups were left undisturbed until weaning at 21 days and then separated into 4 animals per cage and maintained ad libitum on their respective diets. At 30, 90 and 220 days 8 rats from each diet group were weighed and then anesthetized and perfused through the heart with neutral-buffered formalin. The brain was removed the following day and weighed. The modified rapid Golgi method and the criteria for the identification of cell types used were described in a previous publication (Diaz-Cintra et al.la). The entire nucleus raphe dorsalis was serially sectioned at a thickness of 120-160/~m in either the frontal or sagittal plane. From these, 5 well impregnated brain stems were selected at each age from each diet group. Cells were classified according to criteria previously established by Diaz-Cintra et al. lz. At each age and in both diet groups 45 fusiform, 30 multipolar, and 25 ovoid cells were selected as follows. For each rat 3 fusiform and two multipolar cells were quantitated from the rostral, middle and caudal third of the nucleus, and for the less abundant ovoid cells, one from the rostral third and two each from the middle and caudal third of the nucleus. Thus, for each rat 9 fusiform, 6 multipolar, and 5 ovioid cells were quantitated and examined. Cells selected for study were the first consecutivelyencountered at each sampling area that had two well impregnated or primary and respective secondary dendrites arising from opposite poles of the cell that were completely within the section. For each cell the following measurements were made with a calibrated ocular reticle: (1) major and minor axes of the cell body; (2) number of perisomatic spines; (3) number of primary and secondary dendrites; (4) linear extent of two primary and two secondary dendrites; (5) diameter of the primary and secondary dendrites at the midpoint of their extent; and (6) number of dendritic spines along a 50-~m segment at the midpoint of the dendritic extent on two primary and their secondary dendrites. The number of perisomatic spines represents all spines encountered at all focal planes of the cell body. The linear extent of dendrites in the plane of section were measured directly and those not in the plane of section by triangulation according to the method of Bok 6 and Kemper et al. z4. Correction factors for synaptic spine density were not used since the diet, with only minor exceptions noted in the text, had no effect on cell body size or dendritic diameter. Statistical significance was determined using the Student's t-test in the measurements on points 1, 4 and 5 above and the remaining 2, 3 and 6 were analyzed with the nonparametric Mann Whitney U-test.

* P < 0.05. ** P <~ 0.01. *** P < 0.001.

--8% --10% --0.9% --11%*

--9% --10% --8% --7%

- - 0 . 5 % +1%

Exp.

+8% +0.7% --7%

--4% --1% 4 1% --6%

Minor axis

--5% --0.1% --3%

--7% --6% --2% --7%

Major axis

Cont. Exp. Cont.

Cont. Exp. 90 vs 220 Cont. Exp.

30 vs 90

Ovoid

90 vs 220

30 vs 90

Multipolar

Cont. Exp. 90 vs 220 Cont. Exp.

30 vs 90

Fusiform

Perikarya

..10%*

--73%** --4% --17% --8% --40%** 4 13% --7% --3%

--53%*

Diameter

--8% ..12% ..1% --20%***

--9% --5% --7% --9%

--4%

--5% --6% 4-11% ,,15%

±0.9%

+0.9% --0.9% --6%

--5% --5% +2% --2%

Secondary Primary

..14% --3% +8% --13% --15%*** --2% --55%*** --3% --4%

0% ..1% --8% --6% - - 1 4 % --3% --48%** --10%*

Perisom- Number atic spines Primary

Dendrites

--2% --6% --1% --9%

..1%

..1% 0% --5°/

--6% --2% --3% --6%

--2% --5% --4% --7%

--8%

..0.1% --8%* --2%

--3% --0.4% --3% --6%

Secondary Primary

..16%** ...18%** --3% --4% --9%* --15%*** --9%* --4%

--2% +22%* --5% --10%

--13%*

..15%* --12% --9% ..19%**

--6%

--34%*** --11% --18%** --0.7%

--11%

--2% --24%*** --26%*** --25%** --1% --1% --10%* --19%***--15%**

--6% --23%*** --0.8% --9%

Secondary

Number o f spines/50 itm

Secondary Primary

Linear extent

Table to show age-related changes in the morphometric data in the 3 cell types in nucleus raphe dorsalis in the 25 % casein diet and 8 % casein diet rats

TABLE I

t,~ 4~

247 RESULTS The results are shown in Table I and summarized in Figs. 1-5. For all cell types in the control rats, with a single exception, there were no significant age-related changes between 30 and 90 days and 90 and 220 days in the major or minor axes of the cell body and number, diameter, and linear extent of the primary and secondary dendrites. The exception was a 10 % (P < 0.05) decrease in the linear extent of the secondary dendrites of the multipolar cell between 90 and 220 days. In the 8 % casein diet rats many of these measurements showed age-related changes. Between 30 and 90 days the multipolar cell showed a 15 % decrease in number (P < 0.001) and 8 % decrease in linear extent (P < 0.05) of their primary dendrites and a 25 % increase in linear extent of the secondary dendrite (P < 0.01). In register with the latter, the fusiform cell and ovoid cell also showed an increase in extent of their secondary dendrites, respectively 23 % (P < 0.001) and 22 % (P < 0.01). Between 90 and 220 days the ovoid cell showed an 11% increase in its maj or axis (P < 0.05), the fusiform cell showed a 10 % decrease in the number of primary dendrites (P < 0,05), and a 20% decrease in number of secondary dendrites (P < 0.001). The multipolar cells showed a 10 % increase in number of primary dendrites (P < 0.05) and a 13 % decrease in the linear extent of their secondary dendrites (P < 0.05) (Table I). When these measurements were compared at each age for the two diet groups there were no significant differences in the 8 % casein diet rats when compared to controls in the major and minor axes of the cell body (Fig. 1). The number of primary MAJOR AXIS

~'~

~

~

MINOR AXIS

....|

,

~ 20 I i

I

H/C"

I

I

I

..i/,I

I

3o

,i--, .... , f .......

a.i5

5~

I

~ ..........

~

I

z/c"

I

'

/f¢"

--',1~ ~

/

I

I

~fz

J

._I ::a, ~ ~.___ 30I 20 o

I0 J

30

90

I

7'20 30 AGE(DAYS}

90

.zzl "

i

220

Fig. 1. Graphs to show the effect of the 8 % and 25 % casein diet on measurements of major and minor axes of fusiform, multipolar and ovoid cells at 30, 90 and 220 days of age. Solid lines show 25 % casein diet and dashed lines the 8 % casein diet animals. R d- S.E.

248 PRIMARY DENDRITES NUMBER

DIAMETER {~)

5.5

84s

zo

,,o 3.5 -

• I

I~11

u.

7 ........

t ...... I 1 - - I

I

i

~I

"'"

2.5

4O0

~,--I[--,

i.5

I

250 I

I

zo

__.~ I - 3.5

~. . . . .

sso

~

.:,'

I

55

l ~ i

I

.~."

I

1 30

' 90

/.¢

I

~ ..... I 220

I

.;¢

I

L

400

i/ll/i .... II--!

I

250 I

25

~-

.;,"

300

l.O I

I

_i

T~----JJ~

1,5

2J

I

40O

i ....

2.5

.;¢

2.5

i ....

i ..... l .... I I-5

300

I

-3

0

s5o

IO .:¢

5.5

o~4.5

LINEAR EXTENT (p)

25

1.0

["-"-'1 I~[

2,50 I 30

' ~'~ 90 AGE (DAYS)

I 220

I 30

9I O

.;l'

I 220

Fig. 2. Graphs to show the effect of 8 % and 25 % casein diet on the primary dendrites of the fusiform, multipolar and ovoid ceils at 30, 90 and 220 days of age. Solid fines show 25 % casein diet and dashed lines the 8% casein diet animals. ~ ± S.E. *P < 0.05, **P << 0.01, * * * P <: 0.001.

dendrites was decreased at 30 days by 12% on the fusiform cells (P < 0.05) and increased by 12 % on the multipolar cells (P < 0.05). At 90 days there were no significant differences in the 3 cell types. At 220 days their number was decreased by 20 % on the fusiform cells (P < 0.001) and 13 % on the ovoid cells (P < 0.05) (Fig. 2). At 30 days of age the number of secondary dendrites was increased by 32 % on the multipolar cells (P < 0.001) and at 90 days by 33 % on the fusiform cells (P < 0.001) and 24 % on the multipolar cells (P < 0.05). At 220 days they were increased by 12 % on the fusiform cells (P < 0.05) and by 23 % on the multipolar cells (P < 0.01) (Fig. 3). The only significant difference in primary and secondary dendritic diameter was a 13 o/ increase of the primary dendrites of the multipolar cells at 220 days (P < 0.05). The linear extent of the primary dendrites increased by 10 % on the multipolar cells at 30 days (P < 0.05) and by 10 % on the fusiform cells at 220 days (P < 0.05) (Fig. 2). For all 3 cell types the linear extent of the secondary dendrites was decreased at 30 days, respectively 25 %, 20 % and 16 % for the fusiform cells (P < 0.001), multipolar cells (P < 0.001), and ovoid cells (P < 0.01). At older ages the only significant change was a 10 % decrease in the linear extent of the secondary dendrites for the fusiform cells at 220 days (P < 0.05) (Fig. 3). In contrast to the lack of age-related changes in measurements of cell body size

SECONDARY DENDRITES NUMBER

DIAMETER(p)

4.5

~'

~15

;i2

i .......... il" .

~ 2.5

I--

LINEAR EXTENT(p)

2..

2oo

[lO

175

,fill'l""l ].

--I

b15

I0 I

I

/.

I

12E I

I

.~l"

I

~ ~5 i / i ~ , ' ~ I zo V_Z5

1.5

1.5

}......... ~11-:'-=~

150

E

.;t

I 2.5

z.5

,.5

o

1.5

I

.;."

I

I 90

.'.,'

I 220

I"

I

I

/.,"

I

.51

I 220

200

t .....

,5o

~.... l,~l

I.O I 30

I

125 I

4.5

.5¢

I

,'"'" I~'ll"~"~" " I

,75

LO J

I

f

,'"

125 I 30

I 90

,¢ ""

I 220

I 30

~

AGE (DAYS)

Fig. 3. Graphs to show the effect of the 8 ~ and 25 ~ casein diet on the secondary dendrites of the fusiform, multipolar and ovoid cells at 30, 90 and 220 days of age. Solid lines show 25 ~ casein diet and dashed lines the 8 ~ casein diet animals. ~ 4- S.E. *P < 0.05, **P < 0.01, ***P < 0.001. 45

35 o25

__I.

I..... I',,

b-

"Ii. "'I I

2

I

,r

I

45

~

3.5

~2.5

"I',.'\i "II,

z

I

I

1

1

.;r

I

45

.n 55

o25 o 15

510

L ,, t I 90 " 220 AGE (DAYS)

Fig. 4. Graphs to show the effect of the 8 ~ and 25 ~ casein diet on the spine density on the somal surface of the fusiform, multipolar and ovoid cells at 30, 90 and 220 days of age. Solid lines show 25 casein diet and dashed lines the 8 ~ casein diet animals. 4- S.E. *P < 0.05, **P < 0.01, ***P < 0.001.

250 SECONDARY

PRIMARY DENDRITES

DENDRITES

8

9

I ~8 L~

~, i/\"\i

~7 Lt"E

5

6

o

I

~3DOO

I

,;¢

9

I I

I

-f I

8

.~.,"

I

[

J

z hJ

//\"\i F_ ?

I/ 5 I

I

,sz

I

I

I

,~(

I

I

.?~

30

90

i

9 t°,, 8 _1

i~

o

~rl-,~ I -

-I1"

l

.?F

I 220

5~ f 50

I 90

T P

220

AGE (DAYS)

Fig. 5. Graphs to show the effect of 8 % and 25 % casein diet on the spine density in a 50-#m segment on primary and secondary dendrites of the fusiform, multipolar and ovoid cells at 30, 90 and 220 days of age. Solid lines show the 25 % casein diet and dashed lines the 8 % casein diet animals. ,X ± S.E. *P < 0.05, **P < 0.01, ***P < 0.001. and number, diameter, and linear extent of dendrites, control rats showed several significant age-related changes in synaptic spines. Between 30 and 90 days control rats showed an age-related increase in perisomatic spines on ovoid cells of 73 % (P < 0.01). Between 90 and 220 days both the multipolar cells and ovoid cells in controls showed a significant decrease in perisomatic spines o f 5 5 % (P < 0.001) and 40% (P < 0.01), respectively. In the 8 % casein diet rats the only significant age-related changes were a decrease in perisomatic spines between 90 and 220 days of 48 % and 53 %, respectively, for the fusiform cells (P < 0.01) and multipolar cells (P ~ 0.01) (Table I). When the two diet groups were compared at each age the ovoid cells in the 8 % casein diet rats showed a 69 % increase in perisomatic spines at 30 days (P < 0.001), the multipolar cell a 32 % decrease at 90 days (P < 0.05), and the fusiform cells a 52 % decrease at 220 days (P < 0.001) (Fig. 4). In control rats the spine density on the primary dendrites increased between 30 and 90 days by 16 %, 24 % and 15 %, respectively, for the fusiform cells (P < 0.001), multipolar cells (P < 0.001), and ovoid cells (P ~ 0.05). Between 90 and 220 days control rats showed a decrease of 9 % in spine density on primary dendrites on the

251 fusiform cells (P < 0.05) and 19 % on the multipolar cells (P < 0.001). In the 8 % casein diet rats the only significant age-related changes were between 90 and 220 days with the fusiform cells showing a 9 ~oincrease (P < 0.05) and the ovoid cells a 19 % increase (P < 0.01) (Table I). When the two diet groups were compared at each age significant differences in the 8 % casein diet rats were found only at 90 days. All 3 cell types showed a decrease in spine density, respectively, of 12~, 17% and 19% for the fusiform cells (P < 0.01), multipolar cells (P <~ 0.001), and ovoid cells (P < 0.01) (Fig. 5). The spine density on the secondary dendrites in the control animals showed an increase between 30 and 90 days of 18 %, 26 %, and 34 %, respectively, for the fusiform cells (P < 0.01), multipolar cells (P < 0.001), and ovoid cells (P < 0.001). Between 90 and 220 days these cells showed a respective decrease of 15 % (P <: 0.001), 15 % (P < 0.01) and 18 ~o (P < 0.01). In the 8 % casein diet rats there were no significant agerelated changes in spine density between 30 and 90 days and 90 and 220 days (Table I). When compared to controls at each age the fusiform cells in the 8 % casein diet rat showed an 11% decrease in spine density (P < 0.05) at 30 days. At 90 days all cell types showed a significant deficit of 22 %, 19 % and 20 %, respectively, for the fusiform cells (P < 0.001), multipolar cells (P < 0.001), and ovoid cells (P < 0.001). At 220 days this deficit persisted for the fusiform cells and multipolar cells, respectively, 11% (P < 0.05) and 15 ~o (P < 0.05). DISCUSSION In control rats all 3 raphe cell types: fusiform, multipolar and ovoid, showed little change in the major and minor axes of the cell body and diameter, number, and linear extent of the primary and secondary dendrites between 30 and 220 days while synaptic spine density on the cell body and on primary and secondary dendrites showed marked changes during this time. The 8 ~ casein diet rats showed a similar almost complete lack of age-related changes for the major and minor axes of the cell body, dendritic diameter, and linear extent of the primary dendrite. When these measurements were compared to controls at each age only two of 45 comparisons showed a statistically significant change. This lack of effect of the diet on these parameters is in agreement with previous published studies on subcortical formations z2,25. The remaining parameters, which have received little attention in the literature, showed age-related changes that were different from controls. These include perisomatic spines, number of dendrites, linear extent of secondary dendrites, and number of dendritic spines. These changes fell into two categories, one in which these changes in the experimental rat were similar for all cell types and those in which they were different for each cell type. In the first category were age-related changes in secondary dendritic extent between 30 and 90 days and in synaptic spine density between 30 and 90 and 90 and 220 days on both dendritic processes. Between 30 and 90 days all cell types in the experimental rats showed an increase in dendritic extent while controls failed to show any significant change during this time. As a result a deficit in dendritic extent for all cell types at 30

252 days was eliminated at 90 days of age. Between 30 and 90 days both dendritic processes in control rats showed an increase in synaptic spine density followed by a decrease at 220 days. In the experimental rats these age-related changes failed to occur. As a result at 90 days these dendritic processes in the experimental rats all showed a deficit. Prior to that time, at 30 days, only the secondary dendrites on the fusiform cell had shown a deficit and at 220 days only this dendritic process and the secondary dendrites on the multipolar cell continued to show a persistent deficit. The only exceptions were agerelated changes in synaptic density on the primary dendrites of the ovoid cells. In control rats, although showing a decrease between 90 and 220 days, it failed to achieve statistical significance and during this time the experimental rats showed a significant increase in spine density. Thus a major effect of the 8 ~ casein diet on all cell types was a lag in the development in linear extent of the secondary dendrites and the failure of a transient increase in synaptic spines on primary and secondary dendrites. Although the significance of this effect is unknown at the present time, a possibility is that between 30 and 90 days control rats produce an excess of synaptic spines on their dendrites and that those not needed are eliminated by 220 days. This process of synaptic elimination has been well documented during development of innervation of skeletal muscle and autonomic ganglia and previously postulated for the CNS (Purves and Lichtmana3). A possibility is that in the undernourished rats this process is curtailed or eliminated. In the second category there was no consistent pattern of age-related change or in the changes when the two groups were compared at each age, indicating distinct differences between the 3 cell types. These include the changes in secondary dendritic extent between 30 and 90 days and between 30 and 90 and 90 and 220 days in agerelated changes in the number of secondary and primates dendrites and number of perisomatic spines. Of these, the most striking changes were noted in dendritic numbers. This effect was most readily seen when the two diet groups were compared at each age. The fusiform cell in the experimental rat showed a deficit in primary dendritic number at 30 and 220 days and the ovoid cell at 220 days. The multipolar cell showed an increase in primary dendrites at 30 days and in secondary dendrites at all ages. The fusiform cell showed an increase in secondary dendrites at 90 and 220 days. Taking these effects on dendritic number together with the effects on dendritic length and spine density, it is possible to estimate the synaptic spine input to the primary and secondary dendrites and then compare both diet groups at each age. When viewed in this way, the multipolar cell in the experimental rat was the only cell to show an increase in synaptic input on the primary dendrite, which occurred at 30 days of age, and the only cell to show increased synaptic input at all ages on the secondary dendrites. The only other cell to show an increased synaptic input on the secondary dendrites was the fusiform cell which showed a small transient increase at 90 days. All other comparisons showed either no effect of the 8 ~ casein diet or a decrease. Perisomatic spines also showed a different pattern of synaptic input for each cell type, presumedly reflecting a different afferent relationship than that shown by the dendritic tree. The fusiform cells in the experimental rat showed a marked deficit at 220 days, a time when only the secondary dendrite was showing a decrease in spine density. The multipolar cells showed a deficit at 90 days which was in register with a deficit in

253 primary and secondary dendritic spines at this age. The ovoid cells showed an increase in perisomatic spines at 30 days, a time when there were no significant differences in dendritic spines. The significance of these differences between the 3 cell types can only be speculated on since the distribution of afferents to the various parts of the dendritic apparatus of these cells and to their cell bodies is not known. However, some evidence is available regarding their function and axonal projections. According to Felten and Cummings 17 and Felten and Crutcher is all, or virtually all, neurons in the nucleus raphe dorsalis histofluoresce for serotonin. However, Descarries et al. 12, in a study in the adult rat, noted that only 32 ~o of raphe dorsalis cells contain 5-HT. In agreement with thisis the more recent study of Pfister and Danner3L They noted in Nissl-stained and rapid Golgi-impregnated material 3 cell types in the nucleus raphe dorsalis corresponding closely to those noted in our studies. Their type 1 neuron or polygonal neuron with somatic spines appears to correspond to our multipolar cell, their type 2 neuron was referred to as a fusiform neuron and corresponds to our cell of the same name, and their type 3 neuron or pyriform neuron corresponds to our ovoid cell. In histofluorescence studies they noted that only the type 1 neuron, the multipolar cell, showed fluorescence for serotonin. Further, they felt, based on the Golgi-stained material, that the type 2 neuron or fusiform cell was a projection neuron and the pyriform cell (corresponding to our ovoid cell) was a local circuit cell. This correlation of their type 1 neuron with serotonin histofluorescence localization is of interest since the multipolar cell in our 8 ~ casein diet rats is the only neuron that showed increased dendritic synaptic spine input on a primary dendrite at any age and was the only cell to show an increased secondary dendritic spine input at all ages studied. In this regard, our experimental model (8 ~o casein rats) showed serotonin levels that are significantly elevated throughout the brain, particularly in the brain stem30, a5. ACKNOWLEDGEMENTS Supported by Grants BNS 79-22507 (NSF), HD-06364 ( N I C H H D ) , Public Health Service International Research Fellowship NO. 5F05 T W O 2693-02 and C O N A C Y T , Consejo Nacional de Ciencia y Tecnologia, M6xico - - Research Fellowship No. 27234.

REFERENCES 1 Altman, J. and McCrady, B., The influence of nutrition on neural and behavioral development. IV. Effects of infantile undernutrition on the growth of the cerebellum, Develop. Psychobiol., 5 (1972) 111-122. 2 Angulo-Colmenares, A. G., Vaughan, D. W. and Hinds, J. W., Rehabilitation following early malnutrition in the rat: body weight, brain size, and cerebral cortex development, Brain Research, 169 (1979) 121-138. 3 Barnes, D. and Altman, J., Effects of different schedules of early undernutrition on the preweaning growth of the rat cerebellum, Exp. NeuroL, 38 (1973) 406-419. 4 Barnes, D. and Altman, J., Effects of two levels of gestational-lactational undernutrition on the post-weaning growth of the rat cerebellum, Exp. Neurol., 38 (1973) 420-428.

254 5 Bass, N. H., Netsky, M. G. and Young, E., Effect of neonatal malnutrition on developing cerebrum. I. Microchemical and histologic study of cellular differentiation in the rat, Arch. Neurol., 23 (1970) 289-302. 6 Bok, S. T., Histonorny of the Cerebral Cortex, Elsevier, Amsterdam, 1956, pp. 60-98. 7 Callison, D. A. and Spencer, J. W., Effect of chronic undernutrition and/or visual deprivation upon the visual evoked potential from the developing rat brain, Develop. Psychobiol., 1 (1968) 196-204. 8 Clark, G. M., Zamenhof, S., Marthens, E., Grauel, L. and Kruger, L., The effect of prenatal malnutrition on dimensions of cerebral cortex, Brain Research, 54 (1973) 397-402. 9 Clos, J., Favre, C., Selme-Matrat, M. and Legrand, J., Effects of undernutrition on cell formation in the rat brain and specially on cellular composition of the cerebellum, Brain Research, 123 (1977) 13-26. 10 Cordero, M. E., Diaz, G. and Araya, J., Neocortex development during severe malnutrition in the rat, Amer. J. Clin. Nutro, 29 (1976) 358-365. 11 Cragg, B. G., The development of cortical synapses during starvation in the rat, Brain, 95 (1972) 143-150. 12 Descarries, L., Beaudet, A., Watkins, K. C. and Garcia, S., The serotonin neurons in nucleus raphe dorsalis of adult rat, Anat. Rec., 193 (1979) 520. 13 Diaz-Cintra, S., Cintra, L., Kemper, T., Resnick, O. and Morgane, P. J., Nucleus raphe dorsalis: a morphometric Golgi study in rats of three age groups, Brain Research, 207 (1981) 1-16. 14 Dobbing, J., Hopewell, J. W. and Lynch, A., Vulnerability of developing brain: VII. Permanent deficit of neurons in cerebral and cerebellar cortex following early mild undernutrition, Exp. Neurol., 32 (1971) 439-447. 15 Dobbing, J. and Sands, J., Vulnerability of developing brain: IX. The effect of nutritional growth retardation on the timing of the brain growth-spurt, BioL Neonate, 19 (1971) 363-378. 16 Dyson, S. E. and Jones, D. G., Some effects of undernutrition on synaptic development - - a quantitative ultrastructural study, Brain Research, 114 (1976) 365-378. 17 Felten, D. L. and Cummings, J. P., The raphe nuclei of the rabbit brain stem, J. comp. Neurol., 187 (1979) 199-244. 18 Felten, D. L. and Crutcher, K. A., Neuronal-vascular relationships in the raphe nuclei, locus coeruleus and substantia nigra in primates, Amer. J. Anat., 155 (1979) 467-482. 19 Fish, I. and Winick, M., Effect of malnutrition on regional growth of the developing rat brain, Exp. Neurol., 25 (1969) 534-540. 20 Gambetti, P., Autilio-Gambetti, L., Rizzuto, N., Shafer, B. and Pfaff, L., Synapses and malnutrition: quantitative ultrastructural study of rat cerebral cortex, Exp. Neurol., 43 (1974) 464-473. 21 Griffin, W. S. T., Woodward, D. J. and Chanda, R., Malnutrition-induced alterations of developing Purkinje cells, Exp. Neurol., 56 (1977) 298-311. 22 Johnson, J. E. and Yoesle, R. A., The effects of malnutrition on the developing brain stern of the rat: a preliminary experiment using the lateral vestibular nucleus, Brain Research, 89 (1975) 170-174.

23 Lewis, P. D., Bal~zs, R., Patel, A. J. and Johnson, A. L., The effect of undernutrition in early life on cell generation in the rat brain, Brain Research, 83 (1975) 235-247. 24 Kemper, T. L., Caviness, W. F. and Yakovlev, P. I., The neuronographic and metric study of the dendritic arbours of neurons in the motor cortex of Macaca mulatta at birth and at 24 months of age, Brain, 96 (1973) 765-782. 25 Kemper, T. L., Pasquier, D. A. and Drazen, S., Effect of a low protein diet on the anatomical development of subcortical formations, Brain Res. Bull., 3 (1978) 443-450. 26 Leuba, G. and Rabinowicz, Th., Long-term effects of postnatal undernutrition and maternal malnutrition on mouse cerebral cortex. I. Cellular densities, cortical volume and total numbers of cells, Exp. Brain Res., 37 (1979) 283-298. 27 Leuba, G. and Rabinowicz, Th., Long-term effects of postnatal undernutrition and maternal malnutrition on mouse cerebral cortex. 1I. Evolution of dendritic branchings and spines in the visual region, Exp. Brain Res., 37 (1979) 299-308. 28 McConnell, P. and Berry, M., The effects of undernutrition on Purkinje cells dendritic growth in the rat, J. comp. Neurol., 177 (1978) 159-172. 29 McConnell, P. M. and Berry, M., Effects of refeeding after neonatal undernutrition on Purkinje cell dendritic growth in the rat, J. Anat. (Lond.), 126 (1978) 423-425. 30 Morgane, P. J., Miller, M., Kemper, T., Stern, W., Forbes, W., Hall, R., Bronzino, J., Kissane, J.,

255

31 32 33 34 35

36 37

38 39 40

41

Hawrylewicz, E. and Resnick, O., The effects of protein malnutrition on the developing central nervous system in the rat, Neurosci. Biobehav. Rev., 2 (1978) 137-230. Neville, H. E., and Chase, H. P., Undernutrition and cerebellar development, Exp. NeuroL, 33 (1971) 485497. Pfister, C. and Danner, H., Fluorescenzhistochemische und neurohistologische untersuchungen am nucleus raphe dorsalis der ratte, Acta histochem., 66 (1980) 253-261. Purves, D. and Lichtman, J. W., Elimination of synapses in the developing nervous system, Science, 210 (1980) 153-157. Pysh, J. J., Perkings, R. E. and Singer, B., The effect of postnatal undernutrition on the development of the mouse Purkinje cell dendritic tree, Brain Research, 163 (1979) 165-170. Resnick, O., Miller, M., Forbes, W., Hall, R., Kemper, T., Bronzino, J. and Morgane, P. J., Developmental protein malnutrition: influences on the central nervous system of the rat, Neuroci. Biobehav. Rev., 3 (1979) 233-246. Salas, M., Diaz, S. and Nieto, A., Effects of neonatal food deprivation on cortical and dendritic development of the rat, Brain Research, 73 (1974) 139-144. Siassi, F. and Siassi, B., Differential effects of protein-calorie restriction and subsequent repletion on neuronal and nonneuronal components of cerebral cortex in newborn rats, J. Nutr., 103 (1973) 1625-1633. Sima, A. and Persson, L., The effect of pre- and postnatal undernutrition on the development of the rat cerebellar cortex. I. Morphological observations, Neurobiology, 5 (1975) 23-34. Stewart, C. A., Weights of various parts of the brain in normal and underfed albino rats at different ages, J. comp. neuroL, 29 (1918) 511-528. Sugita, N. Comparative studies on the growth of the cerebral cortex. VII. On the influence of starvation at an early age upon the development of the cerebral cortex, albino rat, J. comp. Neurol., 29 (1918) 177-240. West, C. D. and Kemper, T. L., The effect of a low protein diet on the anatomical development of the rat brain, Brain Research, 107 (1976) 221-237.