neurogenesis of the magnocellular basal …neurondevelopment.org/sites/default/files/bayer int j dev...

Inc. 1. Devl. Ncuroscimce. Vol. 3, No. 3. pp. 229-243. 19US. Printed in Great Britain.

0736s74tw-G $03.00+0.lx~ Pergamon Press Ltd.

@ 1985 ISDN

NEUROGENESIS OF THE MAGNOCELLULAR BASAL TELENCEPHALIC NUCLEI IN THE RAT

SHIRLEY A. BAYER

Department of Biology, Indiana-Purdue University, 1125 East 38th Street, Indianapolis, IN 46223, U.S.A.

(Accepted 13 September 1984)

AM&t-Neurogenesis in the magnocellular basal telencephalic nuclei of the rat was examined with [3H]thymidine autoradiography. The experimental animals were the offspring of pregnant females given two injections of [‘Hlthymidine on consecutive embryonic (E) days (E12-E13, E13-E14, E21-E22). On postnatal day (P) 60, the percentage of labeled cells and the proportion of cells originating during 24 h periods were quantified at several anatomical levels throughout the magnocellular basal telencephalic nuclei. The neurons of the horizontal limb of the diagonal band originate mainly between El3 and El6 in a combined rostra]-to-caudal and lateral-to-medial gradient. The neurogenetic gradients in the horizontal limb are continued by generation patterns of cells in the vertical limb of the diagonal band-media1 septal complex, the large cells in the polymorph layer of the olfactory tubercle, and the large cells of the anterior amygdaloid area. The substantia innominata originates between El3 and El7 in combined caudal-to-rostra1 and lateral-to-medial gradients. The globus pallidus originates between El3 and El7 in combined caudal-to-rostra& ventral-to-dorsal and medial-to-lateral gradients. The entopeduncular nucleus originates between El2 and El4 in a ‘sandwich’ gradient where neurons in the core of the nucleus are older than those in either the anterior or posterior ends. There is an overall superficial (ventral) to deep (dorsal) neurogenetic gradient between the magnocellular basal nuclei present at any given rostrocaudal level. An important finding is that neurogenetic gradients in the individual components of the magnocellular basal nuclei are alike (with. the possible exception of the entopeduncular nucleus) indicating they are part of a single system. Finally, evidence is presented that neurogenetic gradients in the magnocellular basal telencephalic neurons can be correlated with their anatomical projections to the cerebra1 cortex.

Key words: Globus pallidus, Substantia innominata, Horizontal limb of the diagonal band, Auto- radiography, Neurogenesis.

The diffuse magnocellular nuclei in the basal telencephalon contain large, isodendritic type neurons and get major input from either the neostriatum (caudate, putamen) or the ventral striatum (olfactory tubercle, nucleus accumbens).2G22,36 Sape? recently showed that these cells project in a topographic manner to the entire cerebral cortex. These large cells have only recently been considered as a single system. Some of the magnocellular groups have been included as part of the olfactory tubercle (polymorph cells in layer III), the preoptic area (magnocellular nucleus, nucleus of the horizontal limb of the diagonal band of Broca, basal nucleus of Meynert), the septal region (medial septal nucleus, nucleus of the vertical limb of the diagonal band), the globus pallidus and entopeduncular nucleus. Much interest has been generated by the finding that these nuclei are the source of cholinergic innervation to the cerebral cortex which may degenerate in patients with Alzheimer’s type senile dementia, providing a pathological substrate for the disease. 7.49

In previous studies, pulse labeling with single injections of [3H]thymidine has been used to establish approximate times for neuronal birthdates in some of the magnocellular basal nuclei in the mouses and hamster.48 The present st ud tions of [3H]thymidine.4

y utilizes comprehensive labeling with multiple injec- This method allows an accurate delineation of both the onset and cessa-

tion of neurogenesis as well as the determination of the proportion of neurons that originate during single days of embryonic life. Portions of the magnocellular basal nuclei have been examined with this technique in previous studies of the septal region,’ amygdala,* and olfactory tubercle.’ The aim of this study is to quantify neurogenetic patterns in the remaining magnocellular nuclei to be examined (nucleus of the horizontal limb of the diagonal band, substantia innominata, globus pallidus and entopeduncular nucleus), and to determine the pattern of neurogenesis throughout all of the magnocellular basal nuclei. Evidence will be provided that neurogenetic gradients in the various magnocellular basal nuclei are similar and can be correlated with their patterns of anatomical connections to the cerebral cortex.

229

230 Shirley A. Bayer

METHODS

The experimental animals were the offspring of Purdue-Wistar timed-pregnant rats given 7 s.c‘. injections of [3H]thymidine (Schwarz-Mann: sp. act. 6.0 Ci/mM; 5 FCiig body wt) to insure comprehensive cell labeling. The injections (given between 9 and 11 a.m.) to an individual animal were separated by 24 h. Two or more pregnant females made up each injection group. The onset of the [3H]thymidine injections was progressively delayed by 1 day between groups (El?-E13. E13-E14. . . . E21-E22). The day of sperm positivity was designated embryonic day 1 (El). Normally, births occur on E23, which is also designated as postnatal day 0 (PO). All animals were perfused through the heart with 10% neutral formalin on P60. The brains were kept for 2-l h in Bouin’s fixative, then were transferred to 10% neutral formalin until they were embedded in paraffin. The brains of at least six animals from each injection group were blocked coronally according to the stereotaxic angle of the Pellegrino et ~1.“’ atlas. Every 15th section (6 km) through the magnocellular basal telencephalic nuclei was saved. Slides were dipped in Kodak NTB-3 emulsion, exposed for 12 weeks, developed in Kodak D19. and post-stained with hematoxylin and eosin.

Anatomically matched sections were selected for quantitative analysis at 11 levels throughout the anteroposterior extent of the magnocellular basal telencephalic nuclei’” (drawings, Figs l-3). Large cells in the basal telencephalic nuclei were counted microscopically at 312.5X in either medial, lateral, or dorsal unit areas set off by an ocular grid (0.085 mm’). For quantification, all cells within a designated area were counted, and were further subdivided into two classes. labeled and nonlabeled. Cells with reduced silver grains overlying the nucleus in densities above background levels were considered labeled; obvious endothelial and glial cells were excluded. The proportion of labeled cells (% labeled cells/total cells) was then calculated from these data.

The determination of the proportion of cells arising (ceasing to divide) on a particular day utilized a modification of the progressively delayed comprehensive labeling procedure,J and is described in detail elsewhere.‘.’ Briefly, a progressive drop in the proportion of labeled neurons from a maximal level (>95%) in a specific population indicates that the precursor cells are pro- ducing nonmitotic neurons. By analyzing the rate \of decline in labeled neurons, one can determine the proportion of neurons originating over blocks of days (or single days) during development. Table 1 shows the data and calculations for the medial part of the horizontal limb of the diagonal band at level A7.4. To calculate the proportion of cells originating on day E14, for example, the proportion of cells labeled the morning of El5 (30.33% in the E15-El6 injection group) is substracted from the proportion of cells labeled the morning of El4 (66.2% in the E14- El5 injection group) to give 35.87% as the proportion of cells originating between the onset of injections on El4 and E15. Trends in cell labeling were analyzed in individual animals with the sign test;h the rationale for the use of this statistic is provided elsewhere.‘,’

Table 1. Neurogenesis in the nucleus of the horizontal limb of the diagonal band (A7.4)

Injection group N

% Labeled cells Day ot “‘% Cells (mean + S.D.) origin originating ’

El2+E13 7 A 100 El2 2.18 (A-B) El3+El4 11 B 97.82-t I.83 El3 31.62(B-C) El4+E15 7 c 66.2 i 10.3x El4 35.87 (C-D) E15+El6 6 D 30.33 t Il.41 El5 28.13 (D-E) El6+El7 12 E 1.9 i- 2.42 El6 1.9

* These data are graphed in Fig. 1C (second from top)

RESULTS

There were three prominent neurogenetic gradients shown by nearly all of the magnocellular basal telencephalic nuclei; the first two lie along the rostrocaudal and lateromedial planes, respectively; the third lies between structures in the superficial-deep plane. Since essentially all segments of the magnocellular basal nuclei quantified at the various anatomical levels were significantly different (sign test, P<O.O5), none of the data could be combined. The data illustrated

Ncurogcncsis of the magnocellular basal tclcnccphalic nuclei in the rat 231

“r

Fig. 1. (A), (B) Photomicrographs of the medial portion of the horizontal limb of the diagonal band (HLDB) at levels A7.2 (A) and AS.6 (B) in the brain of a rat exposed to [3H)thymidine on ElS-El6 and killed on P60. More neurons are labeled in (B) than in (A) illustrating the strong caudal-to-rostra1 neurogenetic gradient (6 pm paraffin sections, hematoxylin and eosin, bar = 50 pm). The bar graphs in (C) are the proportion of cells originating over single embryonic days in the medial HLDB. Cells were counted in the areas indicated in black in the drawings. Table 1 indicates how the data were calculated. The data at level A7.2 are counts of large cells in the anterior amygdaloid area, while those at level AlO. are counts of large cells in layer III of the olfactory tubercle. The prominent posterior-to-anterior neurogenetic gradient is indicated by the arrow next to the drawings and by a gradual shift in neuro-

genetic peaks from earlier to later times at progressively more anterior levels.


MY OF OMGIN-

Fig. 2. (A), (B) Photomicrographs of the medial part of the substantia innominata (SI) at levels A6.8 (A) and A8.6 (B) in the brain of a rat exposed to [%I]thymidine on EL%El6 and killed at P60. More neurons are labeled in (B) than in (A) illustrating the strong caudal-to-rostra1 neurogenetic gradient (6 km paraffin sections, hematoxylin and eosin, bar = SO +m). The bar graphs in (C) are the proportion of ceils originating during single embryonic days in the medial SI. Data are based on counts of celis in the black areas shown in the drawings. The prominent posterior-to-anterior neurogenetic gradient is indicated by the arrow and by a shift in peaks in neuro~nesis from earlier to later times at progressively

more anterior levels.

Neurogenesis of the magnocellular basal teiencephalic nuclei in the rat 233

Fig. 3. (A), (B) ~~omi~~r~hs of the ventral globus pallidus (GP) at levels AS.6 (A) and A6.H (8) in the brain of a rat exposed to @Jthymidine on ES-El6 and killed on Pf10. More neurons are labeled in (B) than in (A) illustrating the strong caudal-to-rostra1 neurogenetic gradient (6 Pm paraffin sections. hematoxylin and eosin. bar = St) km). The bar graphs in (C) are the proportion of cells originating during single embryonic days in the ventral GP. Data are based on counts of cells in the black areas shown in the drawings. The prominent posterior-to-anterior neurogenetic gradient is indicated by the arrow and by a shift in peaks in neurogenesis from earlier to later times at progressively more anterior levels.

Neurogenesis of the magnocellular basal telencephalic nuclei in the rat 235

below show representative examples of the neurogenetic gradients found within and between the various magnocellular pallidal nuclei examined.

Nucleus of the horizontal limb of the diagonal band

The nucleus of the horizontal limb of the diagonal band of Broca (HLDB) has had several names. Loo3’ called this the ‘magnocellular preoptic nucleus’, and Gurdjian”j”a named it the ‘interstitial nucleus of the septal portion of the medial forebrain bundle’. Finally, it is also homologous to the basal nucleus of Meynert in primates, and is often called the ‘basal nucleus’.44 Since it has clear structural continuity with neurons in the diagonal band of Broca,4’ this nucleus will be called its horizontal limb in the following discussion. At the caudal edge of the olfactory tubercle, the HLDB sweeps through the lateral preoptic area toward the medial edge of the piriform cortex. Rostrally, neurons in the HLDB blend with the polymorph cells in layer III of the olfactory tubercle; caudally, they blend with the anterior amygdaloid area.‘5~16~23-25~29~31~51

The large neurons in HLDB were quantified at lateral and medial levels between A7.2 and A8.6 (drawings, Fig. 1C). The photomicrographs in Fig. 1 show neurons in the medial part of the HLDB at levels A7.2 (A) and A8.6 (B) in the brain of a rat exposed to [“Hlthymidine on E15- El6 and killed on P60. The majority of the large neurons are unlabeled in (A), while at least half of the large neurons are labeled in (B), indicating a caudal-to-rostra1 gradient of neurogenesis. The graphs in Fig. 1C show the proportion of neurons originating during 24 h periods throughout the rostrocaudal extent of the medial HLDB. In all cases. caudal levels originate significantly before more rostra1 levels (P~O.0001, sign test). At A7.2, peak neurogenetic days are E13-E14, while at A8.6, the peak occurs on E15-E16. The large neurons at A7.2 were previously quantified in a study of the neurogenesis of the amygdala.* The data in the top graph of Fig. 1C are recounts of the large neurons only. The lateral part of HLDB also shows a caudal-to-rostra1 neurogenetic gradient (data not shown). Some scattered large neurons of this nucleus continue rostrally beyond A8.6 and extend into the olfactory tubercle. The time of origin of the large neurons in layer III in the medial olfactory tubercle have been previously reported in a neurogenetic study of the islands of Calleja and olfactory tubercle (see Fig. 3C of Bayer3). These data are repeated in the bottom graph of .Fig. 1C for comparison. Neurogenesis of the large cells in the olfactory tubercle occurs significantly later than the HLDB neurons at A8.6 (P<O.OOOl, sign test) and the rostrocaudal neurogenetic gradient seen throughout the HLDB continues into the olfactory tubercle.

There is also a prominent neurogenetic gradient between lateral and medial parts of the HLDB. Figure 6 shows data from level A7.4; data from other levels are not illustrated. Neurogenesis peaks laterally on E13, and medially between El3 and El5 (P<O.OOOl, sign test). Neurogenesis of the medial septal nucleus at level A7.4 (upper left graph, Fig. 6) continues the lateral-to-medial gradient throughout the HLDB by being generated significantly later (P<O.OOOl, sign test) than medial neurons in the HLDB. The data for medial septal neurogenesis was previously reported in a study of septal region development’ and is repeated here for comparison. The nucleus of the vertical limb of the diagonal band is also generated significantly later than medial neurons in the HLDB at more rostra1 levels (data not shown).

The substantia innominata

The substantia innominata (SI) contains medium-sized to large neurons lying beneath the striatum and globus pallidus. Classical anatomical descriptions of the globus pallidus mention large-to-medium sized spindle shaped cells extending rostra1 to the anterior border of the globus pallidus just beneath the anterior commissure in the ‘innominate region.“5V20,23,29V31*38*45 Loo31 and Humphreyz3 name this collection of cells the ‘interstitial preoptic nucleus’, while Yourigs cells it the ‘intercalated nucleus of the internal capsule’. Fox” refers to it as the ‘nucleus of the ansa lenticularis’. Today it is commonly called the ‘substantia innominata’.*’ The neurons in the SI were quantified at lateral and medial locations beneath the caudoputamen complex at levels AS.6 and A8.2, and medially back to level A6.2 (drawings, Fig. 2C); the lateral part is indistinct posterior to level A8.2. At medial levels, some large cells lie ventral to the limits of the SI, and extend to the medial HLDB (shown as dots in Fig. 2C); these scattered neurons were included in DI 3:,-B


the quantification of the medial SI. The photomicrographs in Fig. 2 show neurons in the medial part of the SI at levels A68 (A) and A8.6 (B) in the brain of a rat exposed to [“Hlthymidine on EKE16 and killed on P60. Nearly all of the large-to-medium sized neurons are unlabeled in (A) while most are labeled in (B), indicating a caudal-to-rostra1 gradient of neurogenesis. The graphs in Fig. 2C show the proportion of neurons originating during 24 h periods throughout the rostrocaudal extent of the SI. In all cases, caudal levels originate significantly before more rostra1 levels (P<0.005-0.0001, sign test). At level A6.2 the peak days of neurogenesis occur on E13-E14, while at A8.6, peak neurogenesis occurs on E16. El3 is the peak day for neurogenesis laterally at levels A8.2 and A8.6 (data not shown) which is 3-4 days earlier (PC 0.0001, sign test) than peak neurogenesis medially at these same levels.

The giobus pallidus

The globus pallidus (GP) is a mass of large cells lying ventrolateral to the internal capsule and ventromedial to the caudoputamen complex. ‘6.23.31.38s1 It becomes most prominent just posterior to the level where the anterior commissure crosses the midline (level A7.4) and continues as far caudally as level A4.4. The GP was divided into ventral and dorsal halves at six levels between levels A4.4 and A7.4 (drawings, Fig. 3C). The photomicrographs in Fig. 3A, B show neurons in the ventral part of the GP at levels A.5.6 (A) and A6.8 (B) in the brain of a rat exposed to [3H]thymidine on EWE16 and killed on P60. Nearly all large neurons are unlabeled in (A) while most are labeled in (B), indicating another caudal-to-rostra1 gradient of neurogenesis. The graphs in Fig. 3C show the proportion of neurons originating during 24 h periods throughout the rostrocaudal extent of the ventral GP. In all cases, caudal levels originate significantly before more rostra1 levels (P<0.013-0.0001, sign test). Caudally, El3 is the peak time of neurogenesis, while El5 and El6 are peak times rostrally. The dorsal GP shows a similar neurogenetic gradient along the rostrocaudal plane (some data are illustrated in Fig. 5B, C). There is also a significant neurogenetic gradient between ventral and dorsal parts of the GP (Fig. 5B, C). At all levels. neurogenesis in ventral parts occurs significantly before that in dorsal parts (PC 0.0001, sign test). Finally at level A7.4, the ventral GP shows a lateral-to-medial gradient (Fig. 6) with neurogenesis peaking on El5 laterally and on El6 medially (P<O.OOl, sign test). The data are combined for both medial and lateral cells in the ventral GP at level A7.4 in Figs 3 and 5.

The entopeduncular nucleus

The entopeduncular nucleus (EP) is a ventromedial continuation of the globus pallidus. 16*25,31*38*45 In the rat, the EP is small, containing sparsely scattered large neurons among the fibers in the cerebral peduncle. It is commonly considered to be homologous to the internal segment of the globus pallidus in primates. 37.45 EP neurons were counted at three levels between A4.4 and A5.6 (drawings, Fig. 4). The graphs in Fig. 4 show the time of neuron origin over 24 h periods during embryonic development. Neurons at level A5.0 begin to originate significantly earlier (peak on E12) than those at levels A4.4 (P<O.OOOl) and A5.6 (PcO.002) where neurogenetic peaks occur on E13. This is a ‘sandwich’ type of neurogenetic gradient along the rostrocaudal plane.

Neurogenetic gradients between pallidal structures

Most of the neurogenetic gradients between pallidal nuclei lie along a superficial-deep axis. Figure 5 illustrates data from several coronal levels cut perpendicular to the rostrocaudal plane. At level A7.4 (Fig. 5B) neurons in three different pallidal nuclei are stacked superficial to deep: HLDB neurons are oldest and most superficial; SI neurons are slightly younger and deeper; next, are ventral GP neurons still younger and deeper; finally, the youngest and deepest neurons lie in the dorsal GP (P<0.0001-0.0002, sign test). The HLDB and SI continue anterior to level A7.4. and there is a significant superficial-to-deep neurogenetic gradient between these structures at more anterior levels (P<O.OOOl, sign test; Fig. 5A). The SI, ventral and dorsal GP and the EP nuclei are posterior to level A7.4. In all cases, there is a ventral-to-dorsal (or superficial-to-deep) neurogenetic gradient between pallidal nuclei present at any one rostrocaudal level. Figure 5C shows data from the ventral and dorsal GP at levels A6.2 and A4.4; ventral neurons are significantly earlier in generation time (P<O.OOOl, sign test). In addition, the deeper SI neurons at


40

30

20

10

0 60

50

40

30

20

10

0 EEEEE 11 12 13 14 15

DAY OF ORIGIN

Fig. 4. Time of neuron origm in the entopeduncular nucleus. Cells were counted in the black areas shown in the drawings. The bar graphs are the proportion of neurons originating on single embryonic days. Level A5.0 (center graphs) has the earliest peak in neurogenesis, while levels A4.4 (top) and A5.6

(bottom) have later peaks indicating an old-center ‘sandwich’ neurogenetic gradient.

DAY OF ORIGIN

-iA+ OF ORIGIN DIW OF ORIGIN

Fig. 5. Quantification of the prominent superficial-to-deep gradient within and between pallidal nuclei. (A) Proportion of neurons originating during single embryonic days in the medial HLDB (black) and the medial SI (dark stipple) at levels AS.6 (on left) and A8.2 (on right). Large neurons just dorsal to the medial HLDB were included in counts of the SI (indicated as dots in the drawings). (B) Data as in (A) for neurons in the medial HLDB (black, bottom graph), medial SI (dark stipple, graph second from bottom). ventral GP (medium stipple. graph second from top), and dorsal GP (light stipple, top graph) at level A7.4. (C) Data as in (A) and (B) for neurons in the ventral GP (medium stipple) and dorsal GP (light stipple) at levels A6.2 (left) and A4.4 (right). In all three series of graphs, the peaks in neurogenesis shift from earlier to later times in progressively more dorsal areas (arrows in drawings).


% 40

30

20 %

10 SO

0 40

30 30

20 20

10 10

0 0

JO 50

40 40

30 30

20 20

10 10

0 0

12 13 14 1s 16 17 12 13 14 15 16 17 18

DAY OF ORIGIN

Fig. 6. Quantification of the lateral-to-medial neurogenetic gradient found in the magnocellular basal nuclei. Bar graphs are the time of neuron origin over single embryonic days in the medial septal- diagonal band complex (left column of graphs) and ventral globus pallidus (right column). In both columns of graphs, peaks in neurogenesis shift from earlier to later times in progressively more medial

locations (arrows in drawing).

level A6.2 are significantly earlier than the ventral GP neurons at this same level (PcO.043; top graph, Fig. 2C). The deeper EP neurons are significantly earlier than ventral GP neurons at level A4.4 (P<O.OOOl, top graph, Fig. 4). Note that it is the HLDB rather than the SI which extends rostrally into the olfactory tubercle. Cells in the anterior HLDB are generated significantly earlier than the polymorph cells of the olfactory tubercle which continues the rostrocaudal gradient seen throughout the most superficial plate of pallidal neurons (Figs 1C and 7). On the other hand, cells in the SI are located considerably deeper in the telencephalon and, due to the prominent superficial-to-deep neurogenetic gradient between pallidal nuclei, these cells are generated nearly simultaneously with the more rostra1 and superficial polymorph cells in the olfactory tubercle (Fig. 7).

DISCUSSION

The magnocellular basal nuclei as an integrated whole

Figure 7 diagrammatically illustrates the three major neurogenetic gradients seen throughout all of the magnocellular basal nuclei except the entopeduncular nucleus. First, there is a caudal- to-rostra1 gradient in the globus pallidus (top plate), substantia innominata (intermediate plate), and diagonal band-medial septal complex which extends into the olfactory tubercle (bottom plate). This is illustrated in Fig. 7 by an increase in shading in areas containing older neurons at the caudal limits of each of the three nuclear complexes, respectively. The visual effect of a rostral-to-caudal direction is created in Fig. 7 by a narrowing of the structures in the rostra1 end; in reality, many structures (e.g. globus pallidus) are wider at their anterior ends. Second, there is a lateral-to-medial gradient within the diagonal band-medial septal complex, confirming the findings of previous [3H]thymidine autoradiographic studies. ‘a48 This is illustrated by darker lateral shading in Fig. 7. Data for the vertical limb of the diagonal band and the medial septal nucleus were previously reported in a [“Hlthymidine autoradiographic study of the septal region. ’

Neurogenesis of the magnocellular basal telencephalic nuclei in the rat

A4.4

Al-O.0 I I

Fig. 7. A three-dimensional overview of the neurogenetic gradients seen within and between the ventral globus pallidus (top plate), substantia innominata (middle plate) and diagonal band-medial septal complex (bottom plate). Arrows on right indicate directions: R, rostral; C,.caudal; L, lateral; M, medial; D, dorsal; V, ventral. The numbers in each segment are the proportion of neurons which have already originated by the morning of El5 and can no longer be labeled. The level of shading is made darker or lighter to indicate more or ‘less accumulated cells, respectively. Darker to lighter shading goes from the caudal-to-rostra1 parts of each structure to represent a strong neurogenetic gradient along this same line. Darker to lighter shading goes from lateral-to-medial parts in the posterior segments of the bottom plate to indicate a similar neurogenetic gradient. The superficial-to-deep neurogenetic gradient is represented by darker shading along the more ventral plates lying in a single vertical dimension. The magnitude of this gradient is more easily seen in the inset where the three structures are superimposed directly over each other. For the sake of simplicity, the dorsal globus pallidus is not included, and the lateral-to- medial gradients within the rostra1 substantia innominata and rostra1 ventral globus pallidus are not

illustrated.

A lateral-to-medial neurogenetic gradient is also found rostrally in the substantia innominata and globus pallidus (not shown in Fig. 7). Third, there is a superficial-to-deep gradient between the various basal telencephalic nuclei so that at any one rostrocaudal level, more dorsal (deep) cells are generated later than ventral (superficial) cells. This is shown by an increase in shading in ventral areas along a given vertical dimension in Fig. 7 (note that the dorsal part of the globus pallidus is not illustrated). The prominent caudal-to-rostra1 gradient and the superficial-to-deep gradients between nuclei and within the globus pallidus have not been previously reported. When one views the magnocellular basal nuclei as a whole, it is striking that neurons scattered throughout such a large expanse of the telencephalon should show this uniformity of neurogenetic gradients.

The similarity in neurogenetic gradients between the various pallidal nuclei examined suggests they are components of a single system of magnocellular neurons spread throughout the telencephalon. Johnston’” observed several different mammalian brains and noted that the large neurons scattered beneath the anterior commissure and in the diagonal band of Broca could only be compared with those in the globus pallidus. 0benchain38 spoke of a ‘giant cell stream’ throughout the basal telencephalon of marsupials. Later descriptive anatomical studies traced these large cells from the globus pallidus all the way to the olfactory tubercle.‘5,23,24,2g,51 Heimef and coworkers2%2’ have put forth the hypothesis that an enlarged pallidal system is found in the


mammalian brain which has both dorsal (globus pallidus) and ventral (diagonal band, substantia innominata) components. One of the best examples of continuity between pallidal components is the presence of cholinergic neurons. In primates, the cholinergic cells of the basal nucleus of Meynert and rostra1 basal forebrain (as indicated by choline acetyltransferase content) do not continue into the globus pallidus, 33.35 although some globus pallidal neurons are acetylcholines- terase positive. 39 However in rats, the ventromediai giobus pallidus contains cholinergic ceils.26*27,32 Recent advances in histochemical techniques have shown a remarkable similarity between neurochemical characteristics of the globus pallidus and the more rostra1 basal telencephalic nuclei beneath the anterior commissure. In the rat, for example, both the dorsal and ventral pallidum contain prominent iron deposits and stain intensely for enkephahn.“.47 Thus, three separate lines of evidence (classical histology, neurochemical characteristics, and patterns in neurogenetic timetables), suggest that all neurons of the magnocellular basal telencephalic nuclei constitute a single system.

The interlocked synchrony between neurogenetic timetables in the globus pallidus, substantia innominata and diagonal band suggests that these magnocellular basal telencephalic neurons are generated in neuroepithelia which are closely related in time and space during morphogenesis, and possibly arise from a single neuroepithelial source. Richter,4” in a study of human embryonic brains, concluded that both the internal (entopeduncular nudeus) and external segments of the globus pallidus along with the subthalamic nucleus arise in a caudal-to-rostra1 sequence from an ‘early exhausted’ neuroepitheiium lining the third ventricle. The strong cauda~-to-rostra1 neurogenetic gradient in the globus pallidus reported here would predict such a morphogenetic gradient. However, the entopeduncular nucleus is the only group of pallidal neurons studied which has a ‘sandwich’ rather than a caudal-to-rostra1 gradient. A shift in neurogenetic gradients usually indicates multiple germinal sources. For example, the neurogenetic gradients observed in the corticomedial vs the basolateral amygdala are not closely interrelated and the various nuclear groups in the amygdala are probably derived from different germinal zones.’ On the basis of its neurogenetic gradients, the entopeduncular nucleus probably does not originate from the same germinal source as the globus pallidus. The gtobus pailidus and substantia innominata may be generated in a neuroepithelium lining the ventromedial floor of the lateral ventricle (Bayer, unpublished data) making it a telencephalic rather than a diencephalic structure. Loo31 mentioned that the globus pallidus may be derived from a transition zone between the diencephaIon and telencephalon. A study of morphogenesis of the rat basal te~encephalon (Bayer and AItman, in preparation) will be done once all the timetables of neurogenesis in the basal telencephalon have been completed to resolve some of these issues.

Neurogenetic gradients in the magnocellular basal nuclei correlated with their anatomical connections to the cerebral cortex

A host of experimental anatomical studies have demonstrated that the entire cerebral cortex receives input from the magnocellular basal nuclei. The experimental evidence is accumulating that this projection is primarily cholinergic. ‘9~*6~27,30.32.33,35 In the rat, ceils in the globus pallidus project to widespread neocortical areas. 14.2628.30.32*44 The substantia innominata projects mainly to the neOcOrtex ‘).IO.l4.2~2(3.311.42.44

, as well as to the piriform cortex. lx The horizontal limb of the diagonal band projects to the neocortex”‘.*X.“‘.42.44 as well as to the piriform cortex.# On the other hand, the vertical limb of the diagonal band-medial septal complex projects mainly to the entorhinal cortex and hippocampal region5,“-‘9.32.34.44.46 and to the cingulate cortex. **

Some of the more recent studies have reported that the magnocellular basal nuclear projection to the cortical mantle is topographic.2R.42.w Several of the connections mapped by Sapera can be correlated with neurogenetic patterns. Figure 8 shows the broad outlines of the correlation between neurogenetic gradients in the cortex, neurogenetic gradients in the magnocellular nuclei, and their anatomical connections. Throughout the entire length of the cortex, lateral parts near the rhinai sulcus originate before medial parts (Altman and Bayer, unpublished data).13.50 Within the magnocellular basal nuclei, the older neurons are located laterally and ventrally, while younger neurons are situated medially and dorsally. The large open arrows in Fig. 8 are pointing in the direction of the youngest neurons in the cortex and magnocellular basal nuclei, respectively. To give some specific examples of these correlations (based on Saper’j), the older neurons


CORRELATED WITH ANATOMICAL CONNECTIONS

CEREBRAL CORTEX

NEUROGENETIC GRAOIENTS

OLDER TO YOUNGER NEUROGENETIC GRADIENTS

ANATOMICAL CONNECTIONS Ok& to O&r Regions ---3 m-r w y@jm RgQ&s . . . . . . . . ;.+

Fig, 8. Neurogenetic gradients (large outlined arrows) in the cerebral cortex and magnocellular basal nuclei correlated with their anatomical conne~ions (arrows connecting dashed or dotted lines). The small arrow set on the left indicates directions as in Fig. 7. In the cerebral cortex, the medial parts develop later than more lateral parts. In the magnoceliular basal nuclei, medial and dorsal parts contain younger neurons than lateral and ventral parts. The anatomical connections are set up so that older basal nuclei neurons project to earlier developing cortical areas (dashed line), while younger basal nuclei

neurons project to later developing cortical areas (dotted line).

in the ventral parts of the globus pallidus project to the early developing granular insular cortex near the rhinal sulcus, while progressively younger neurons in the dorsal globus pallidus project to later developing dorsomedial cortical areas such as the retrosplenial and visual cortices. Within the diagonal band-medial septal complex, the older lateral cells in the horizontal limb project to the laterally placed and early developing piriform cortex while younger medial cells in the vertical limb and medial septal nucleus project to the late developing hippocampal region, along the medial cortical edge, again maintaining the older-to-older (dashed lines in Fig. 8) and younger-to- younger relationship (dotted lines in Fig. 8). Within the ventral part of the rostra1 globus pallidus, the older neurons in lateral parts project to earlier developing lateral frontal cortex while younger neurons in medial parts project to a later developing prelimbic cortex along the medial cortical wall. These many correlations suggest that the origin of neurons in both the cortex and magnocellular basal nuclei may be timed so that anatomical connections between them develop in a specific pattern.

Acknowkdgemenfs-1 wish to thank Joseph Altman for advice and encouragement. Technical assistance was provided by Peggy Cleary and Vicki Palmore. Kathy Shuster and Mark O’Neil drew the figures. Janis Malone did the photographic work. and Connie Flexman typed the manuscript. This research was supported by NIH Grant NS-19744.

REFERENCES

Bayer S. A. (1979) The development of the septal region in the rat. I. Neurogenesis examined with 3H-thymidine autoradiography. J. camp. Neural. 183, 89-106. Bayer S. A. (1980) Quantitative ‘H-thymidine radiographic analyses of neurogenesis in the rat amygdala. J. camp. Neural. 194, 845-875. Bayer S. A. (1984) Neurogenesis in the olfactory tubercle and islands of Calleja in the rat. Int. J. Deal Neurosci. 3, 13-s-147. Bayer S. A. and Altman J. (1974) Hippocampal development m the rat: cytogenesis and morphogenesis examined with autoradiography and low-level X-irradiation. J. camp. Neural. 158, 55-80. Beckstead R. M. (1978) Afferent connections of the entorhinal area in the rat as demonstrated by retrograde labeling with horseradish peroxidase. Brain Res. 152, 249-264. Conover W. J. (1971) Prarrical Nonparametric Staristics. John Wiley, New York. Coyle J. T.. Price D. l_. and Delong M. R. (1983) Alzheimer’s disease: a disorder of cortical chohnergic innervation. Science. t-l/ash. 219. I1&1-1190.


8.

9.

10.

11 12

13

14 15 16 17

18

19

20

21

22

23

24

25

26

27.

28.

29.

30.

31. 32.

33.

34

35.

36.

37.

38.

39.

JO.

41.

42.

43.

44.

Creps E. S. (1974) Time of neuron origin in preoptic and septal areas of the mouse: an autoradiographic study. J. camp. Neural. 157, 161-244. Divac I. (1975) Magnocellular nuclei of the basal forebrain proiect to neocortex. brain stem and olfactorv bulb. Review of some fu&tional correlates. Brain Res. 93, 385-398: _ Divac I., Kosmal A., Bjorklund A. and Lindvall 0. (1978) SubcortIcal projections to the prefrontal cortex in the rat as revealed by the horseradish peroxidase technique. Neuroscience 3, 785-796. Domesick V..B. (1976) Projectibns of the nucleus’of the diagonal band of Broca in the rat. Anar. Rec. l&&391-392. Fallon J. H. (1983) The islands of Calleja complex. II. Connections of medium and large sized cells. Bruin Res. Bull. 10, 775-793. Fernandez V. and Bravo H. (1974) Autoradiographic study of development of the cerebral cortex in the rat. &am Behav. Evol. 9, 317-332. Finch D. M., Derian E. L. and Babb T. L. (1984) Afferents to rat cingulate cortex. Esp. .Veurol. (in press). Fox C. A. (1940) Certain basal telencephalic centers in the cat. J. camp. Neurol. 72, 1-62. Gurdjian E. S. (1927) The diencephalon of the albino rat. J. camp. Neural. 43, l-114. Haber S. N. and Nauta W. J. H. (1983) Ramifications of the globus pallidus in the rat as indicated by patterns of immunohistochemistry. Neuroscience 9, 245-260. Haberly L. B. and Price J. L. (1978) Association and commissural fiber systems of the olfactory cortex of the rat. I. Systems originating in the piriform cortex and adjacent areas. J. camp. Neural. 178, 711-740. Hartgraves S. L.. Mensah P. L. and Kelly P. H. (1982) Regional decreases of cortical choline acetyltransferase after lesions of the septal area and in the area of nucleus basalis magnocellularis. Neuroscience 7, 2369-2376. Heimer L. (1978) The olfactory cortex and the ventral striatum. In Limbic Mechanisms: the Continuing Evolurion of the Limbic System Concept (eds Livingston K. E. and Hornykiewicz 0.). pp~ 95-187. Plenum Press, New York. Heimer L., Switzer R. D. and VanHoesen G. W. (1982) Ventral striatum and ventral pallidum: components of the motor system? Trends Neurosci. 2, 83-87. Heimer L. and Wilson R. D. (1975) The subcortical projections of the allocortex: similarities in the neural associa- tions of the hippocampus, the pirifotm cortex and the neocortex. In Golgi Centennial Symposium: Perspectives in Neurobiology (ed. Santini M.), pp. 177-193. Raven Press, New York. Humphrey T. (1936) The telencephalon of the bat. I. The noncortical nuclear masses and certain pertinent fiber connections. J. camp. Neurol. 65, 603-711. Johnson T. N. (1957) Studies on the brain of a guinea pig. I. The nuclear pattern of certain basal telencephalic centers. J. camp. Neurol. 107, 353-377. Johnston J. B. (1923) Further contributions to the study of the evolution of the forebrain. J. camp. Neural. 35, 337- 481. Johnston M. V., McKinney M. and Coyle J. T. (1981) Neocortical cholinergic innervation: a description of extrinsic and intrinsic components in the rat. Exp. Bruin Res. 43, 159-172. Kelly P. H. and Moore K. E. (1978) Decrease of neocortical choline acetyltransferase after lesion of th,e globus pallidus in the rat. Exp. Neurol. 61, 47w84. Lamour Y.. Dutar P. and Jobert A. (1982) Topographic organization of basal forebrain neurons projecting to the rat cerebral cortex. Neurosci. Lerr. 34, 117-122. Lauer E. W. (1945) The nuclear pattern and fiber connections of certain basal telencephalic centers in the macaque. 1. camp. Neurol. 82, 215-254. Lehmann J., Nagy J. I.. Admadja S. and Fibiger H. C. (1980) The nucleus basalis magnocellularis: the origin of a cholinergic projection to the neocortex of the rat. Neuroscience 5, 1161-1174. Loo Y. T. (1931) The forebrain of the opossum Didelphys virginia. Part II. Histology. J. camp. Neurol. 52, 1-148. McKinney M., Coyle J. T. and Hedreen J. C. (1983) Topographic analysis of the innervation of the rat neocortex and hippocampus by the basal forebrain cholinergic system. J. camp. Neurol. 217, 103-121. McKinney M., Struble R. G., Price D. L. and Coyle J. T. (1982) Monkey nucleus basalis is enriched with choline acetyltransferase. Neuroscience 7, 2362-2368. Meibach R. C. and Siegel A. (1977) Efferent connections of the septal area in the rat: an analysis using retrograde and anterograde transport methods. Brain Rex 119, I-20. Mesulam M. M.. Mufson E. J.. Levey A. I. and Wainer B. H. (1983) Cholinergic innervation of cortex by the basal forebrain: cytochemistry and cortical connections of the septal area, diagonal band nuclei. nucleus basalis (substantia innominata), and hypothalamus in rhesus monkey. J. camp. Neurol. 214, 170-197. Mogenson G. I., Swanson L. W. and Wu M. (1983) Neural projections from nucleus accumbens to globus pallidus. substantia innominata, and lateral preoptic-lateral hypothalamic area: an anatomical and electrophysiological investi- gation in the rat. J. Neurosci. 3, 189-202. Nauta W. J. H. and Mehler W. R. (1969) Fiber connections of the basal ganglia. In Psychofropic Drugs and Dysfunc- tions of‘the Basal Ganglia. A Multidisciplinary Workshop (eds Crane G. E. and Gardner R.. Jr.), pp. 67-74. Public Health Service Publication No. 1938. Obenchain J. B. (1925) The brains of the South American marsupials, Caenolestes and Orolestes. Field Mus. NUI. Hist. 224 (Zool. Ser.) 14, 175-232. Parent A., Gravel S. and Olivier A. (1979) The extrapyramidal and limbic systems relationship at the globus pallidus Icvel: a comparative histochemical study in the rat, cat, and monkey. In Advances in Neurology, Vol. 24 (eds Poirier L. J., Sourkes T. L. and Bedard P. J.), pp. l-l 1. Raven Press. New York. Pellegrino L. J.. Pellegrino A. S. and Cushman A. J. (1979) A Sfereotaxic Atlas of the Rat Brain. Second Edition. Plenum Press. New York. Price J. L. and Powell T. P. S. (1970) An experimental study of the origin and course of the centrifugal fibres to the olfactory bulb in the rat. J. Anar. 107, 215-237. Price J. L. and Stern R. (1983) Individual cells in the nucleus basalis diagonal band complex have restricted axonal projections to the cerebral cortex in the rat. Bruin Res. 269, 352-356. Richter E. (I96.S) Die entwicklung des globus pallidus und des corpus subthalamicum. Monogrn Gesamtgeb. Neural. Psychiut. 108, l-131. Sapcr C. B. (1984) Organization of cerebral cortical afferent systems in the rat. II. Magnocellular basal nucleus. J. wmp. Neural. 222, 313-342.


45. Smith 0. C. (1930) The corpus striatum, amygdala, and stria terminalis of Tamandua tetradacryla. J. camp. Neural. 51,65-127.

44. Swanson L. W. and Cowan W. M. (1976) Autoradiographic studies of the development and connections of the septal area in the rat. In The Septat Nuclei (ed. DeFrance J. W.), pp. 37-44. Plenum Press, New York.

47. Switzer R. C. III, Hill J. and Heimer L. (1982) The globus paihdus and its rostroventral extension into the olfactory tubercle of the rat: a cyto- and chemoarchitectural study. Neuroscience 7, 1891-1904.

48. tenDonkelaar H. J. and Dederen P. J. W. (1979) Neurogenesis in the basal forebrain of the Chinese hamster. 1. Time of neuron origin. Anat. Embryol. 156, 331-348.

49. Whitehouse P. J., Price D. L., Struble R. G., Clark A. W., Coyle J. T. and Delong M. R. (1982) Alzheimer’s disease and senile dementia: loss of neurons in the basal forebrain. Science, Wash. 215, 1237-1239.

50. Woodhams P. L., Basco E., Hajos F., Csillagy A. and Balazs R. (1981) Radial glia in the developing mouse cerebral cortex and hippocampus. Amt. Embryol. X63,331-343.

51. Young M. W. (1936) The nuclear pattern and fiber connections of the non-cortical centers of the telencephalon of the rabbit (Lepus cuniculus). J. camp. Neural. 65, 295-401.

neurogenesis of the magnocellular basal …neurondevelopment.org/sites/default/files/bayer int j dev...

Documents