Postnatal development of the locus coeruleus in Trisomy 19 mice: Morphological and morphometric study

Brain Reseawh Builctrn. Vol. 28, pp. 923-930, Printed in the USA. All rights reserved.

1992 Copyright

0361-9230/92 $5.00 + .OO 0 1992 Pergamon Press Ltd.

Postnatal Development of the Locus Coeruleus in Trisomy 19 Mice: Morphological and Morphometric Study DIETRICH

E. LORKE’

AND

LA.RS P. KLIMASCHEWSKI

Anatomisches Institut, Ahteilung,Ji’ir Neuroanatvmi, Universitiitskrunkenhaus Eppendorf; D 2000 Humhur~ 20, Germun) Received

8 October

1991

LORKE, D. E. AND L. P. KLIMASCHEWSKI. Postnurui devrl~pmt~n! cfthr /octa mmria~.r in Triswny 19 mice: Morphological and ~~#rp~z~~~f~i~ srrc&.BRAIN RES BULL 28(6) 923-930, 199X---To study the effect of trisomy upon a brain region that is generated very early during development. the locus coeruleus (LC) has been examined mo~hologi~ally and mo~homet~cally in 23 Trisomy 19 (Ts19) mice and their chromosomally balanced control littermates aged 2-18 days postpartum. Gross morphological alterations of the Ts 19 LC could neither be observed by light nor by electron microscopy. The LC was properly located. Ultrastructural features indicating increasing protein synthesis such as nucleolus-like bodies and a rise in the amount of granular endoplasmic reticulum and in the size of the nucleoli have been observed both in Ts19 and control mice. Maturation of the LC was delayed in Ts19. Morphometric studies on the volume, cell number, and cell density revealed that, apart from a 2-day delay in development, the Ts19 LC was of normal size. The present study supports the observation that the no~drenergi~ svstem is not affected in the Tsl9 CNS. Taking into account that the cerebellum of Tsl9 mice is markedly hypoplastic. the resultsjndicate a differential pathogenic effect of trisomy upon different neural systems. Aneuploidy

Brain stem

Development

Mouse

Locus coeruleus

TRISOMY of the human chromosome 2 1, the Down syndrome, is the most common genetic cause for mental retardation (25). The precise mechanisms by which the additional chromosome leads to pathogenetic changes of the CNS are poorly understood. An experimental system allowing the systematic investigation of autosomal trisomies is provided by a mouse model described by Gropp (IO). Decreased viability is a genera1 characteristic associated with murine trisomies: Only Trisomy 19 (Ts 19) mice survive beyond birth. Murine Ts I9 therefore permits detailed studies on the effect of aneuploidy upon postnatal development. The majority of Ts19 mice die during the first postnatal week, and survival beyond the second postnatal week is extremely rare ( 14). Body weight is greatly reduced. The expression of developmental abnormalities of the cardiovascular system, palate, testes, and CNS can vary with the genetic background of the parental strains; in the progeny of NMRI-outbred mice, gross malfo~ations of Ts19 mice have not been observed (24,14,15,27,41). In the CNS, development of neurons, glia, myelin sheaths, and blood vessels in the retina, optic nerve, and cerebellar cortex is delayed by 2 days, but the orderly sequence of development is not disturbed ( 14,16,17). In addition, a significant hypoplasia of the cerebellum has been observed ( 16). It has been suggested that a premature arrest in neurogenesis, most notice-

Morphometry

Trisomy 19

able in brain regions maturing late, is responsible for impaired CNS function in trisomic organisms (24). This would imply that neurons arising very early during development are less affected by trisomy than brain regions developing mainly postnatally, such as the cerebellum. Among the first definite neuroblasts to be generated are the cells of the locus coeruleus (LC); the neurons of this pontine nucleus, which has been implicated in a variety of functions such as respiration, motivation, learning, and sleep, are already born between embryonic days 9 and 11 (36,38). Being the largest norepinephrine-containing nucleus in the mammalian brain and forming the major source of nom~ene~c inne~ation for the entire brain [see (23) for review], the LC is of particular interest because a variety of neurochemical alterations have been described in murine Ts16 and Ts19 (11,26,30,35). Postnatal development of the LC has therefore been studied in Ts19 mice by morphological and morphometric methods. METHOD

Animals The breeding design for the generation of trisomic mice is based on the observation that the probability of meiotic nondisjunction is increased in parental stocks heterozygous for two metacentric Robertsonian (Rb) chromo~mes ( i 0). Ts 19 mice

’ Requests for reprints should be addressed to D. E. Lorke, Anatomisches Injtitut, Abteilung fiir Neuroanatomie. U. K. E., Martin&r. 52, D 2000 Hamburg 20, Germany.

923

924

were obtained by mating laboratory-strain NMRI (Han.) females with males carrying the Rb(9.19)163H and Rb(8.19)lCt metacentrics, (Mu Liibeck), each of which contains one chromosome 19. The birth date, denoted as postnatal day 0 (PO), was determined by two inspections of pregnant females per day. To provide sufficient nutrition and maternal care for the Ts19 pups, litters were reduced to three to four animals at birth. For electron microscopy, one Ts19 mouse and its chromosomally balanced littermate were taken for analysis on P2, P4, P5, P6, P8, and P9; for morphometric analysis, one Ts 19 mouse was compared to its euploid littermate at each developmental stage (P2, P4, P6, P12, Pl4, P16, and P18), and on P8 and PlO five Ts19 pups and their euploid littermates were studied.

Histology Animals were weighed, deeply anaesthetized with an intraperitoneal injection of pentobarbital (100 mg/kg body weight), the abdomen was opened, and the spleen and the femur were removed for chromosome analysis. After incision of the jugular veins, a fixative consisting of 6% glutaraldehyde in 0.05 M phosphate buffer (pH 7.3; 760 mosM) was perfused through the left ventricle of the heart for 15 min under a pressure of 100 mm Hg. After overnight immersion of the whole head in the fixative, the skull was opened, the brain was carefully removed, and the brainstem with adjacent cerebellum was dissected from the rest of the brain rostra1 to the colliculi superiores. For electron microscopy, specimens were halved in the midline, the lateral, and the ventral parts of the brainstem and the cerebellum were cut away under a dissecting microscope, and the remaining tissue was immersed in fresh fixative for a further 24 h. Then, the tissue was rinsed in a mixture of 0.1 M phosphate buffer and 0.1 M saccharose, postfixed with buffered osmium tetroxide (1% for 2 h), dehydrated in an ascending series of ethanol, cleared in propylene oxide, and embedded in Epon 8 I2 ( 18). One-pm transverse semithin sections of the brainstem were cut with a Reichert OMU 3 (Wien) ultramicrotome and stained with toluidine blue and pyronine red. To obtain a defined plane of section, the brainstem was cut at closely spaced intervals

LORKE AND KLIMASCHEWSKI

starting from the colliculi inferiores so a section of the compact caudal part of the LC was obtained. Immediately adjacent ultrathin sections (70-90 nm) were prepared for electron microscopy and stained with saturated alcoholic uranyl acetate and lead citrate before examination with a Zeiss 09 (Oberkochen) transmission electron microscope. For morphometric analysis, entire brainstems were embedded in epoxy resin without postfixation in osmium, serially cut on a Reichert Supercut at 10 pm, and stained with toluidine blue/ pyronine red. Care was taken to section the brainstem perpendicular to its long axis. To determine the volume of the LC, the thick sections were projected by a Zeiss light microscope and a Panasonic (Osaka) camera into a MOP-Videoplan (Kontron, Munchen) semiautomatic image analyser and the outline of the LC was traced manually with a cursor on each section (magnification, X350). Only those cells were included in the measurement that were part of the densely packed group of neurons exhibiting the typical cytological criteria of LC neurons: a large, lightly stained nucleus with a prominent nucleolus and abundant cytoplasm with Nissl bodies occupying mostly the periphery. The total volume of the LC was determined by multiplying the sum of the cross-sectional areas by the thickness of the sections (IO pm). For estimation of the cell number. the nucleoli of all LC neurons were counted in every second section using an eyepiece reticule (magnification, X625) and the sum was multiplied by two. The total cell number was determined by correcting the crude nucleolar number according to Abercrombie’s (1) formula for double counting taking into account fragments too small to be resolved ( 12). The mean diameter of the nucleoli and the size of the smallest fragment visible were obtained for each animal by tracing the nucleoli of 100 LC neurons with the Kontron Videoplan (magnification, X2,250). In addition, the number of nucleoli per nucleus was counted in 100 LC neurons of three Ts19 mice and their control littermates aged 2, 10. and I8 days. Ninety-eight percent of the nuclei contained one nucleolus, whereas only 2% contained two: there was no variation with age. Mean section thickness was determined microscopically by focussing the top and bottom surfaces of sections and measuring the difference on the fine-focus micrometer. Cell density was

FIG. I. Semithin section of the caudal part of the LC on P2. Both in the (a) Ts.19 mouse and (b) control littermate the LC appears as a compact group of neurons with a light nucleus, a prominent nucleolus, and abundant cytoplasm with Nissl bodies. In the Ts19 mouse (a), most of the small dark nuclei of glioblasts (arrowheads) are visible at the medial margin of the LC, and only few have invaded the LC. In the control littermate (b), the majority of the glioblasts are already found within the LC. Neurons of the Ts19 LC are more densely packed than those of the control littermate, and the LC occupies a smaller area. X 170.

LOCUS COERULEUS

OF TRISOMY

925

19 MICE

FIG. 2. Ten-pm section of the caudal part of the LC on PIO. The LC is correctly located at a right angle to the floor of the IV ventricle (asterisks) and is of normal morphology. At its lateral margin, it is bounded by the darkly stained neurons of the mesencephalic nucleus of the trigeminal nerve (arrows); the medial border is made up by the central grew - . matter. Both in (a) Tsl9 mice and (b) control littermates LC neurons are more widely dispersed than at younger ages. X 120.

by dividing the total number of LC neurons by the volume of the LC. For statistical analysis, the Student’s t test (two tailed) was used on P8 and PlO.

determined

Chromosome Analysis All animals examined were karyotyped from preparations of minced spleen (P2-P8) or femoral bone marrow (P 1O-P 18). After incubation in tissue culture medium in the presence of Colcemid (37°C 90 min), hypotonic treatment with 0.075 M KCl, and methanol:acetic acid (3: 1) fixation, air-dried slides stained with Giemsa were evaluated at a magnification of about 1,000X [for details, see (14)]. Ts19 mice are characterized by the presence

of 2 metacentric chromosomes and a total of 4 1 arms, whereas control littermates only show 1 metacentric chromosome and 40 arms. RESULTS

Shortly after birth (PZ), the LC already appeared as a clearly distinguishable group of densely packed cells both in Ts19 mice and control littermates (Figs. la and b). Situated at the lateral margin of the fourth ventricle, it had the form of a crescent in coronal sections. Laterally, it was bounded by the darkly stained neurons of the mesencephalic nucleus of the trigeminal nerve (V); the medial border was made up by the central grey matter.

FIG. 3. Electron micrograph of the dorsal part of the LC of a 2-day-old (a) Ts19 mouse and (b) its control littermate. Both neurons show the typical indentations of the nuclear membrane (arrows) and a nucleolus-like body (asterisks). In LC neurons of the control mouse, more granular endoplasmic reticulum (arrowheads) has already been formed. X4,000.

926

LORKE AND IUIMASCHEWSIU

FIG. 4. Electron micrograph of the dorsal part of the LC on P8. In the LC of both the (a) Ts19 mouse and (b) its control littermate cell size, amount of granular endopl~mic reticulum {arrowheads),and number of Golgi complexes (arrows) have increased. As a result of a dday in development, this increase is less marked in Ts19 neurons. X4,000.

Whereas the LC was compact caudally and could easily be delimited from surrounding structures (Figs. la and b), it consisted of few scattered celis at its rostrai pole. Already at this stage of development, the perikarya exhibited the typical cytological characteristics of LC neurons: a large, pale nucleus with a prominent nucleolus and a marked cytoplasmic basophilia that was accentuated in the periphery of the perikarya by the presence of Nissl bodies. In contrast to the human LC, neuromelanin pigment could not be observed by light microscopy in the mouse LC. At the medial margin of the Ts 19 LC, numerous small dark nuclei of glioblasts were visible at this early stage of development; only few glioblasts had invaded the LC (Fig. la). In control mice, the majority of glioblasts were already found within the LC (Fig. 1b); otherwise, no morphological differences were observed between Ts19 mice and control littermates. However, the neurons of the Ts19 LC were more densely packed, and the LC was therefore occupying a smaller area as compared to control littermates. In addition, the LC of the Ts19 mouse extended over fewer sections than that of the euploid control. During the postnatal period (P2-P12), the LC perikarya increased considerably in size with both genotypes and became separated by the developing neuropil; as a result, cells were more widely dispersed and the LC did not appear as compact as before (Figs. 2a and b). From P4 on, glioblasts were also observed within the LC of Ts19 mice.

Both in Ts 19 and control mice, two main types of neurons could be distinguished: fusiform and multipolar. The predominant cell type in the ventral part of the LC was the multipolar neuron, whereas the dorsal part consisted mainly of fusiform neurons. These fusiform cells were studied by electron microscopy. Already on P2, they had the typical ultrastructural appearance of LC neurons both in Ts19 and control mice (Figs. 3a and b): Their large nuclei had prominent nucleoli, relatively light ka~oplasma, and deep indentations of the nuclear membrane. Especially in the Ts 19 LC, the cytoplasm still occupied a relatively small area and contained relatively sparse granular endoplasmic reticulum and only few Golgi complexes. In addition, nucleolus-like bodies (NLBs) were observed in both genotypes, round to oval organelles lacking a delimiting membrane. They consisted of a dark amo~hous condensation of granules surrounding an electron-lucent core. Occasionally, osmiophilic lysosomal dense bodies were visible. In the course of development, the size of the perikarya increased in both genotypes (Figs. 4a and b); a massive increase in the amount of granular endoplasmic reticulum was observed that condensed into Nissl bodies preponde~ntly located in the periphery of the cells. Golgi complexes became more numerous and the number of lysosomal dense bodies increased. At the ultrastructural level, LC neurons of Ts19 mice did not exhibit any cytological abnormalities either. However, the perikarya of LC neurons were generally smaller in Tsl9 mice, and the granular

LOCUS COERULEUS

OF TRISOMY

TOTAL

i:::::: ::::::: i::::::: :::::.:: ii:::::: ::::::: i::::::: ::::::: .::::::: ,::::::: I MEAN DIAMETER OF NUCLEOLI

VOLUME

volume

T

x106(pm3)

l2

921

19 MICE

2,8-

j--J control

lo q

/

2,6-

Ts,Q

8

&4-

::::::: ii::::: ::::::: ::::::: ii::::: ::::::: ::::::: ::::::: ::::::: :i::::: ::::::: i:::::: i:::::: ::::::: ::i:::: ::::::: ::::::: i:::::: ~ 4

Y,’

22

I

14

16 18 age(days)

FIG. 5. Development of the total volume of the LC for Ts19 mice and control littermates determined by tracing the contours of the LC on IOpm sections. Values represent single measurements; on days 8 and IO, the means f SD from five animals are depicted.

endoplasmic reticulum was less abundant as a consequence of a delay in development. The volume of the LC increased until PlO in control animals and P12 in Tsl9 mice (Fig. 5). Initially, the volume of the LC was smaller in Ts 19 mice (p 5 0.00 1); at later stages, a difference was not obvious. The uncorrected number of nucleoli for Ts 19 mice and control littermates is shown in Table 1. It increased until P14 in control animals and P16 in Tsl9 mice. However, these uncorrected figures take into account neither the thickness of the sections nor the size of the nucleoli. Whereas section thickness and the size of the profiles too small to be identified do not vary with age, the diameter of the nucleoli increased during development, as depicted in Fig. 6. Taking these morphometric data into consideration, the number of nucleoli was more accurately estimated by applying Konigsmark’s ( 12) correction formula (Fig. 7). Determination of the number of nucleoli per LC neuron revealed that both in Ts19 mice and in control littermates aged 2, 10, and 18 days the majority of the neurons contained a single nucleolus; the number of cells containing two nucleoli was about 290, a percentage within the range reported for adult rats (32,37). Therefore, the corrected number of nucleoli can be regarded as an accurate estimate for the cell number of the LC. Both in Ts19 and control mice, the number of LC neurons increased slightly during the first postnatal week (Fig. 7). Before reaching its maximum on P 16 in Ts 19 mice, the cell number was reduced in the

/’

.---.

x

I

I

I

I

I

I

I

2

4

6

8

10

12

14

To19

1 I 16 18 age (days)

6. Mean diameter of the nucleoli of the LC in Ts19 and control mice. One hundred nucleoli per animal have been traced on lo-Mm sections. Values represent single measurements; on days 8 and 10, the means f SD from five animals are depicted. FIG.

trisomics (p 5 0.001); thereafter, it was essentially the same in both genotypes. Cell density of the LC decreased until PlO in control mice and P 12 in Ts 19; afterward, it remained relatively constant (Fig. 8). Whereas LC neurons were more densely packed in Ts19 mice than in control animals during the first 10 days (p 5 O.OOl), no difference was observed at later stages of development. DISCUSSION

The LC is a useful model for assessing the cell deficit in the CNS of trisomic mice, whose viability is greatly reduced. This anatomically discrete, relatively well-defined tegmental pontine nucleus is homogeneous in the sense that the great majority of its neurons contain norepinephrine. Its neurons develop rather early during murine development, that is, between embryonic days 9 and 11 (36,38), and are therefore generated long before the death of the Ts19 mice. In addition, it is one of the few regions in the CNS where no significant neuronal degeneration has been observed during development (6); therefore, a cell count is not compromised by neuronal cell death. At the ultrastructural level, postnatal development of the LC was characterized by a rapid enlargement of neuronal perikarya and a rise in both the number of polyribosomes and amount of granular endoplasmic reticulum. An increase in cell size accompanied by a multiplication of ribosomes and their consecutive structural incorporation into granular endoplasmic reticulum is a general characteristic of neuronal maturation (13,39); these

TABLE 1 NUMBER

OF NUCLEOLI

COUNTED

(CRUDE

NUMBER)

IN LC NEURONS

OF Tsl9

MICE AND CONTROL

LITTERMATES

Age (days)

Ts19 Control

2

4

6

8

10

12

14

16

18

676 848

712 968

804 I.004

887 f 19.54 1.052 + 19.72

917 f 14.21 1,058 f 16.95

984 1,016

1,004 1,124

1,128 1,092

1,052 1,092

Values represent single measurements. On days 8 and 10, values are the means k SEM from five animals each.

LORKE

928

NUMBER OF NEURONS number

lOOO800600400-

-

control

.---.

Ts19

2001

I

I

I

1

I

I

2

4

6

8

10

12

14

1

I

16 18 age(days)

FIG. 7. Number of neurons of the LC of Ts19 and control littermates determined by counts of the nucleoli (12). Values represent single measurements; on days 8 and 10, the means f SD from five animals are depicted.

changes are related to an increased production of proteins necessary for dendritic and axonal outgrowth and subsequent synaptogenesis (2 I ,34). This increase in cell size and in the amount of granular endoplasmic reticulum was observed during development of both

control and Ts19 LC. As compared to littermate controls, the size of the perikarya and amount of granular endoplasmic reticulum were reduced in Ts 19, an observation reflecting the delay in neurogenesis already observed in other brain regions, for example, cerebellum ( 16) and the visual system ( 14,17). Otherwise, no morphological abnormalities were detectable. Nuclear contour irregularities or increased celhtlar membrane fragmentation as described in electron microscopic studies of Ts16 neurons (28) were not observed in the present study. The LC was properly located and exhibited the typical ultrastructural characteristics, such as prominent nucleolus, indentations of the nuclear membrane, and NLBs. NLBs, first described by Rohde (29) in the cytoplasm ofganglion cells, have been observed in various types of neurons including rat and mouse LC ($3 I). Because developmental studies on the LC indicate the largest number of NLBs is observed during maximal metabolic activity (34) and because an increase in the number of these organelles is observed in young LC neurons engaged in intense regenerative efforts (33), they are thought to be protein stores during phases of high metabolic activity (34). In the present study, the cell number of the LC has been determined by counting the nucleoli. This procedure seems appropriate because 98-99% of LC neurons contain a single nucleolus [present study; (32,37)]. However, although the section thickness (10 pm) was several times larger than the diameter of the nucleoli (about 2.5 pm) it cannot be excluded that fragments of nucleoli are counted twice (1). Therefore, the number of nucleoli counted overestimates the exact cell number. The crude nucleolar count has therefore been corrected according to a formula described by Konigsmark (12) that takes into account settion thickness, size of the smallest resolvable profile, and mean nucleolar diameter. Measurements of nucleolar diameter revealed that the size of the nucleoli increased by 25% between P2 and P 14. This rise in nucleolar size during development has been observed in several brain regions, for example, the rat LC (32) and the oculomotor nucleus of Tupaia belangeri (42). It can be explained by the function of the nucleolus as a production site for ribosomal nucleoproteins: The enlargement of the nucleoli reflects the enormous increase in granular endoplasmic

AND

KLIMASCHEWSKI

reticulum (13,19). The present results show the importance of taking into account the size of the nucleoli when determining the cell number by nucleolar count in developmental studies because the diameter of the nucleoli changes according to the functional activity of the cell. The cell number of the LC increased until P8 in control mice and PI6 in Ts19 mice. This was an unexpected observation because in the mouse, LC neurons are already generated very early during development, that is. between embryonic days 9 and 11 (36,38). It does not seem likely that the rise in cell number observed was due to an increase in glia cells, which immigrate into the LC until P14 (34) because the morphology ofglioblasts is clearly distinguished from that of LC neurons and care was taken only to include LC neurons into the cell count. An increase in the number of nucleoli per cell during development has not been observed either; on the contrary, it decreases with age (13,42).The most plausible explanation for the rise in the number of LC neurons seen in control mice during the early postnatal days, in agreement with the results described by Touret et al. (40) seems that maturation of LC neurons still proceeds during the early postnatal period so that not all LC neurons can be identified morphologically before the second postnatal week. Evidence for this hypothesis is supplied by a developmental study

of the rat LC (34) showing that cytoplasmic differentiation of LC neurons mainly takes place during the first two postnatal weeks. With Ts19 mice, an increase in cell number until P16 has been observed. This may be explained by a considerable delay in the differentiation of Ts19 LC neurons. However, it seems more likely that this observation was due to the sampling of the specimens examined. More than 90% of Ts19 mice die within the first 10 postnatal days (14). It is known that an additional chromosome impairs development to a variable degree. The oldest surviving animals are a self-selected population and their very survival marks them as different from the rest of the Tsl9 mice studied. Therefore, it seems likely that Tsl9 mice surviving until P16 and Pl8 were less severely affected by trisomy; consequently, LC number was higher in these mice. Even taking these considerations into account, the morphometric analysis shows the number of LC neurons is only reduced by less than 10% in Ts19 mice. Otherwise, morphometric data only revealed a 2-day lag in maturation of the Ts 19 LC. CELL DENSITY neurons per~06 ,,m3

1401

-

60

l

control ---•

Ts 19

40

I

I

I

I

1

1

2

4

6

8

10

I

12

1

14

1

1

16 18 agetdays)

FIG. 8.Changes in cell density (number of neurons per lo6 pm’) of the LC for Ts 19 mice and control littermates calculated from the data depicted in Figs. 5 and 7. Values represent single measurements; on days 8 and 10, the means f SD from five animals are depicted.

LOCUS

COERULEUS

OF TRISOMY

19 MICE

929

The present results indicate that the noradrenergic LC of Ts19 mice, although delayed in development and slightly reduced in size, is not pathologically altered. These findings are in good agreement with biochemical studies of norepinephrine levels in the brainstem and telencephalon of embryonic Ts19 mice (30) and support the observation that the noradrenergic system is not affected in the Ts19 CNS. Two mechanisms underlying the phenotypic effects of trisomy have been proposed: specific gene dosage effects and a generalized disruption of the genetic balance (8,9,25,30). Whereas dosage effects of specific genes result in distinctive phenotypic features characteristic of the particular trisomy, gene imbalance results in more generalized effects associated with a variety of trisomies. One possibility in distinguishing between these two mechanisms is the comparison between Ts 16 and Ts 19 mice. Because of a partial genetic homology between human chromosome 21 and mouse chromosome 16, Ts16 mice can be considered a model for specific gene dosage effects associated with the Down syndrome, whereas Ts19 mice help to identify the effects of gene imbalance, as abnormal features seen both in Ts.16 and Ts19 mice are probably due to general effects of trisomy. Such general features associated with a variety of trisomies, including growth retardation and hypoplasia, have also been observed in the LC of Ts19 mice. Because in the human Ts21, the Down syndrome, hypoplasia is most marked in those brain regions generated late during development, such as the cerebellum and hippocampus, it has been suggested that late-maturing neuronal components are preferentially affected by trisomy due to a premature arrest in neurogenesis (24). The present results support this hypothesis: Whereas a considerable hypoplasia of the late-maturing cerebellum has been found in Ts19 mice ( 16), the LC whose neurons are generated very early during development is hardly reduced in cell number. A reduction in the cell number of the LC by over 70% has been described in older patients with Down syndrome (20).

However, conflicting results have been reported concerning the number of LC neurons in younger Down patients. A significant loss of LC neurons has been described in younger Down patients prior to acquisition of Alzheimer’s-type changes (20). In contrast, cell counts of the LC of a 5.5month-old infant revealed a neuron number within the normal range; tyrosine hydroxylase (TH) activity did not indicate any profound defect in the catecholaminergic system (22). In embryonic Ts 16 mice, reduced levels of TH and dopamine-P-hydroxylase (DBH), the two rate-limiting enzymes of catecholaminergic neurons, have been observed in the whole brain (26) TH being decreased both in the cerebral hemispheres and diencephalon/brainstem (35). In addition, the number of TH immunoreactive cells in the LC was reduced by 30% (11). However, these results are difficult to interpret as only data of embryonic days 17 and I8 are available. Because there is a six- to tenfold rise in both TH and DBH levels between embryonic day 18 and postnatal day 14 (7) the reduced values in Ts16 can reflect either a delay in development or a selective decrease in the noradrenergic neurons of the LC, as suggested by Kiss et al. (11). Since Ts16 does not allow survival beyond term, postnatal studies on the LC cell number are not feasible in this trisomy. The present results on the postnatal development of Ts19 mice suggest that an early loss of LC neurons is not an invariant consequence of trisomy. Taking into account that the cerebellum is severely hypoplastic in Ts 19 mice, they also indicate a differential pathogenetic effect of trisomy upon different neuronal systems. ACKNOWLEDGEMENTS

We thank E. BGhm,I. Schade,and K. Siebert for their skillful technical assistance, and are grateful to PD Dr. Heinz Winking, Medizinische Universitat zu Liibeck, for the generous gift of mice carrying Robertsonian translocations.

REFERENCES 1. Abercrombie, M. Estimation of nuclear population from microtome sections. Anat. Rec. 94:239-246; 1946. 2. Bacchus, C.; Sterz, H.; Buselmaier, W.; Sahai, S.; Winking, H. Genesis and systematization of cardiovascular anomalies and analysis of skeletal malformations in murine trisomy 16 and 19. Two animal models for human trisomies. Hum. Genet. 77:12-22; 1987. 3. Baranov, V. S.; Vaisman, B. L.; Udalova, L. D. Some morphological peculiarities and state of rRNA synthesis, in the mouse embryos with trisomy for chromosome 19. Ontogenez 13:46-54; 1982. 4. Bersu, E. T. Morphologic development of the fetal trisomy 19 mouse. Teratology 29: 1I7- 129; 1984. 5. Coutsen, J. L. The locus coeruleus: A combined morphological study in the mouse. Ph.D. thesis, University of Arizona; 1978. 6. Cowan, W. M.; Fawcett, J. W.; O’Leary, D. D. M.; Stanfield, B. B. Regressive events in neurogenesis. Science 225:1258-1265; 1984. 7. Dreyfus, C. F.; Markey, K. A.; Goldstein, M.; Black, I. B. Development of catecholaminergic phenotypic characters in the mouse locus coeruleus in vivo and in culture. Dev. Biol. 97:48-58; 1983. 8. Epstein, C. J. Trisomy 21 and the nervous system: From cause to cure. In: Epstein, C. J., ed. The neurobiology of Down syndrome. New York: Raven Press; 1986:1-15. 9. Gearhart, J. D.; Oster-Granite, M. L.; Reeves, R. H.; Coyle, J. T. Developmental consequences of autosomal aneuploidy in mammals. Dev. Genet. 8:249-265; 1987. 10. Gropp, A. Animal model: Autosomal trisomies in fetal mice, exencephaly in mice with trisomy 12. Am. J. Pathol. 77:539-542; 1974. 11. Kiss, J.; Schlumpf, M.; Balazs, R. Selective retardation of the development of the basal forebrain choline@ and pontine catechol-

12.

13.

14.

15.

16. 17. 18. 19. 20.

21.

aminergic nuclei in the brain of trisomy I6 mouse, an animal model of Down’s syndrome. Dev. Brain Res. 50:25 l-264; 1989. Konigsmark, B. W. Methods for the counting of neurons. In: Nauta, W. J. H.; Ebbesson, S. 0. E., eds. Contemporary research methods in neuroanatomy. Berlin: Springer Verlag; 1970:3 15-340. LaVelle, A.; LaVelle, F. W. Cytodifferentiation in the neuron. In: Himwich, W. A., ed. Developmental neurobiology. Springfield, IL: Charles C. Thomas; 1970:117-164. Lorke, D. E.; Latter, M. Gliogenesis and myelination in the optic nerve of trisomy 19 mice. A quantitative electron-microscopic study. Acta Anat. (Base]) 137:222-233; 1990. Lorke, D. E.; Schulze, W.; Winking, H. Early postnatal development of the testes in mice with trisomy 19. Prog. Clin. Biol. Res. 296:7176; 1989. Lorke, D. E.; Stan, A.; Lierse, W. Morphogenesis of the cerebellum of trisomy 19 mice. Biomed. Res. 10:385-396; 1989. Lorke, D. E.; Winking, H. Histogenesis of the retina in murine trisomv 19. Brain Res. Bull. 16:845-851: 1986. Luf< J. H. Improvements in epoxy resin embedding methods. J. Biophys. Biochem. Cytol. 9:409-414; 196 1. Mann, D. M. A. Nerve cell protein metabolism and degenerative disease. Neuropathol. Appl. Neurobiol. 8:161-176; 1982. Mann, D. M. A.; Yates, P. 0.; Marcyniuk, B.; Ravindra, C. R. Loss of neurones from cortical and subcortical areas in Down’s syndrome patients at middle age. Quantitative comparisons with younger Down’s patients and patients with Alzheimer’s disease. J. Neural. Sci. 80:79-89: 1987. Matthews, M. A.; Narayanan, C. H.; Narayanan, Y.; Onge, M. F. St. Neuronal maturation and synaptogenesis in the rat ventrobasal

LORKE

930

22.

23.

24.

25.

26.

21.

28.

29.

30.

complex: Alignment with developmental changes in rate and severity of axon reaction, J. Comp. Neurol. 173:745-772; 1977. McGeer, E. G.; Norman, M.; Boyes, B.; O’Kusky, J.; Suzuki, J.; McGeer, P. L. Acetylcholine and aromatic amine systems in postmortem brain of an infant with Down’s syndrome. Exp. Neurol. 87: 557-570; 1985. Moore, R. Y.; Bloom, F. E. Central catecholamine neuron systems: Anatomy and physiology of the norepinephrine and epinephrine systems. Annu. Rev. Neurosci. 2: 1I3- 168; 1979. Nadel, L. Down syndrome in neurobiological perspective. In: Epstein, C. J., ed. The neurobiology of Down syndrome. New York: Raven Press: l986:239-25 1. Oster-Granite, M. L.; Gearhart, J. D.; Reeves, R. H.; Singer, H. S.; Moran, T. H.; Coyle, J. T. Down syndrome and trisomy 16 mouse. In: Coyle, J. T., ed. Animal models of dementia. New York: Alan R. Liss; 1987:279-307. Ozand, P. T.; Hawkins, R. L.; Collins, R. M.; Reed, W. D.; Baab, P. J.; Oster-Granite, M. L. Neurochemical changes in murine trisomy 16: Delay in cholinergic and catecholaminergic systems. J. Neurothem. 43:401-408; 1984. Pexieder, T.; Miyabara, S.; Gropp, A. Congenital heart disease in experimental (fetal) mouse trisomies: Incidence. In: Pexieder, T., ed. Perspectives in cardiovascular research, vol. 5. New York: Raven Press; 198 I :389-399. Plioplys, A. V.; Bedford, H. M. Murine trisomy 16 model ofDown’s syndrome: Central nervous system electron microscopic observations. Brain Res. Bull. 22:233-243; 1989. Rohde, E. Ganglienzellkern und neuroglia. Ein Kapitel Uber Vermehrung Und Wachsthum Von Ganglienzellen. Arch. Mikr. Anat. 47:121-135; 1896. Saltareili, M. D.; Forloni, G. L.; Oster-Granite, M. L.; Gearhart, J. D.; Coyle, J. T. Neurochemical characterization of embryonic brain development in trisomy 19 (Ts19) mice: Implications of selective deficits observed for abnormal neural development in aneuploidy. Dev. Genet. 8~267-279; 1987.

AND

~I~ASCHEWS~

3 1. Shimizu, N.; Imamoto, K. Fine structure of the locus coeruleus in the rat. Arch. Histol. Jpn. 3 1:229-246; 1970. 32. Sievers, J. Experimentelle Studien zur Frame der intrazerebralen Induktion: De; locus coeruleus und sein t&minales Zielgebiet, das Kleinhirn. Habil. Schr., University of Hamburg; 1980. 33. Sievers, J.; Lolova, 1. Effects of pre- and postnatal treatment with 6-hydroxydopamine on the development of the rat locus coeruleus. Bibl. Anat. 19:258-263; 1981. 34. Sievers, J.; Lolova, I.; Jenner, S.; Klemm, H. P.; Sievers, II. Morphological and biochemical studies on the ontogenesis of the nucleus locus coeruleus. Bibl. Anat. 19:52- 130; 198 1. 35. Singer, H. S.; Tiemeyer, M.; Hedreen, J. C.; Gearhart, J.; Coyle, J. T. Mo~holo~c and neur~hemi~l studies of embryonic brain development in murine trisomy 16. Dev. Brain Res. 15155-166; 1984. 36. Steindler, D. A.; Trosko, B. K. Two types of locus coeruleus neurons born on different embryonic days in the mouse. Anal. Embryol. (Berl.) 179:423-434; 1989. 37. Swanson, L. W. The locus coeruleus: A cytoarchitectonic, Golgi and immunohistochemical study in the albino rat. Brain Res. 1l&3956; 1976. 38. Taber Pierce, E. Time of origin of neurons in the brain stem of the mouse. Prog. Brain Res. 40:53-65; 1973. 39. Tennyson, V. M. The fine structure of the develonina nervous system. In: Himwich, W. A., ed. Developmental neurobioio8y. Springfield, IL: Charles C. Thomas: 1970:47- 116. 40. Touret, M.; Valatx, J.-L.; Jouvet, M. The locus coeruleus: A quantitative and genetic study in mice. Brain Res. 250:353-357; 1982. 41. White. B. J.; Tiio. J.-H.: Van de Water. L. C.: Crandall. C. Trisomy 19 in the labomtory mouse. I. Frequency in different c&es at specific developmental stages and relationship of trisomy to cleft palate. Cytogenet. Cell Genet. 13:217-231; 1974. 42. Zilles, K.; Schleicher, A.; Kretschmann, H.-J.; Wingert, F. Semiautomatic morphometric analysis of the nucleolar development in the nucl. n. oculomotorii of Tupaia belangeri during ontogenesis. Anal. Embryol. (Berl.) 149: 15-28; 1976.