Anat Embryol (1996) 193:475-480
Seiji M a t s u d a 9P e t e r B a l u k
Takashi Fujiwara
9Springer-Verlag 1996
9D a i s a b u r o S h i m i z u
Dorsal root ganglion neuron development in chick and rat
Accepted: 27 September 1995
The aim of this study was to compare the
morphological development of dorsal root ganglion neurons in embryonic and early postnatal chicks and rats.
The three-dimensional architecture of neurons was observed in ganglia in situ and in dissociated neurons by
scanning electron microscopy after removal of the capsule and connective tissue. The percentages of neurons at
different developmental stages were determined. The
general morphological changes in the chick resembled
those in the rat but the timing was different. In both
chick and rat, the majority of neurons were bipolar at
early stages of development (embryonic day 6 in chick
and day 14 in rats) and later underwent pseudo-unipolarization to become mostly unipolar neurons at hatching or
birth. This maturation event started at an earlier stage in
chick embryos than in rats, with 57% unipolar neurons in
chick and only 7% in rat on embryonic day 14. However,
just after hatching or birth, at day 22 of development, a
larger proportion of immature unipolar neurons remained
in chicks (13%) than in rats (3%). We conclude that
these differences should be taken into consideration in
designing experiments on dorsal root ganglion neurons
grown in tissue culture.
Abstract
S. Matsuda (~)
Department of Anatomy,
Ehime UniversitySchool of Medicine, Shigenobu,
Ehime 791-02, Japan
Tel.: 81-899-64-5111 ext. 2504; Fax: 81-899-64-7483
E-mail:matsuda@m.ehime-u,ac.jp
E Baluk
CardiovascularResearch Institute, School of Medicine,
San Francisco, CA 94143-0131, USA
D. Shimizu
Central ResearchLaboratory,
Ehime UniversitySchool of Medicine, Shigenobu,Ehime,
Japan
T. Fujiwara
LaboratoryAnimal Center, Ehime UniversitySchool of Medicine,
Shigenobu, Ehime, Japan
Key words Spinal ganglia. Sensory ganglia.
Pseudo-unipolarization - Scanning electron microscopy 9
Differentiation
Introduction
Mature dorsal root ganglion (DRG) neurons of higher
vertebrates are unique in that they are devoid of normal
dendrites, but instead have a single trunk that divides into central and peripheral processes at some distance from
the cell body. During their early embryonic development
the neurons are bipolar, and they undergo a remarkable
morphological change, termed pseudo-unipolarization, to
assume their mature form (Pannese 1974; Lieberman
1976).
Pseudo-unipolarization has been examined in histological sections using classical silver impregnation methods
(His 1886; Cajal 1890; Lenhossek 1892), by transmission
electron microscopy (Tennyson 1965; Pannese 1968), and
more recently using immunohistochemistry for cytoskeletal markers (Riederer and Barakat-Walter 1992; De Koninck et al. 1993). However, with the above methods, the
precise time course of pseudo-unipolarization cannot be
determined accurately, because the neurons are visualized
in only two dimensions and only a few can be seen in
their entirety. Furthermore, there is little detailed information on the development of dorsal root ganglion neurons in birds, although these have been widely used as the
source of material for tissue culture studies (Mudge 1984;
Riederer and Barakat-Walter 1992).
Scanning electron microscopy (SEM) has overcome
some of these limitations of observing neurons, and using this technique, we have described the three-dimensional appearance of the neuronal cell bodies and processes in adult and in embryonic rat dorsal root ganglia
(Matsuda and Uehara 1981; Matsuda and Uehara 1984).
The aims of the present study were therefore to determine whether the general pattern of development of dorsal root ganglion neurons in the chick resembles that of
the rat, and to determine the percentages of immature
476
Fig. 1 Scanning electron micrographs of dorsal root ganglion
neurons types at successive stages of development in chick (a-d)
and rat (e). a Eccentrically bulging bipolar neuron on day E8. b
Bell-shaped bipolar neuron on day El3. c Short-stem unipolar
neuron on day El3. d Long-stem nnipolar neuron and bell-shaped
bipolar neuron on postnatal day 1. e Long-stem unipolar neurons
on postnatal day 1. Bars 5 gm
and mature neurons at different development stages by
using SEM. The hope is that these data will be useful in
designing experiments in tissue culture. In the present
study, we determine the percentages of different neuron
types in randomly selected, dissociated neurons from enzymatically digested chick and rat dorsal root ganglia,
because it has been reported that neurons in the superficial portion of the ganglion cells appear to develop earlier than those in the deeper portion (Cajal 1904).
477
Fig. 2a-e Scanning electron micrographs of dissociated chick
dorsal root ganglion neurons at successive stages of development.
a Eccentrically bulging bipolar neuron on day E8. b Bell-shaped
bipolar neuron on day El6. e Long-stem unipolar neuron on postnatal day 2. Bars 5 gm
Materials and methods
Scanning electron microscopy of intact developing dorsal
root ganglia
Chicks from embryonic day 6 (E6) to post-hatching day 2 (P2)
and rats from El4 to P1 were used. The ages of chick embryos
were confirmed using criteria described previously (Hamburger
and Hamilton 1951), and rat embryos were taken from timed-mated females. Rats were killed by an overdose of pentobarbital anesthetic, chicks by cervical dislocation. All animal experiments were
approved by the Ethical Committee of the University of Ehime
Medical School. Seven to 10 rats and 5 to 10 chicks were used at
individual stages and at least 10 thoracic DRGs were dissected out
from each animal. After removal of the DRG connective-tissue
capsules with fine forceps, the ganglia were treated with phosphate-buffered chick Ringer solution containing 0.25% trypsin for
20-30 rain at 37~ and fixed with 3% glutaraldehyde in 0.1 M
phosphate buffer (pH 7.4) for 60 min. They were dehydrated in a
graded series of ethanols, immersed in iso-amyl acetate, critical
point-dried, sputter-coated with platinum and examined with a
Hitachi S-500A scanning electron microscope as described previously (Matsuda and Uehara 1981, 1984).
Scanning electron microscopy of dissociated dorsal
root ganglion neurons
Ganglia were immersed in phosphate-buffered saline (PBS) at
4~ the connective-tissue capsules were removed, and the ganglia
were divided into small pieces with fine forceps. They were infiltrated with PBS containing 2.5 mg/ml trypsin for 60 min at 4~
and then incubated in the same solution for 20 min at 37~ The
specimens were rinsed in ice-cold PBS and gently triturated with a
siliconized Pasteur pipette so that ganglion neurons were dissociated and suspended. Glutaraldehyde was added to the cell suspension to a final concentration of 2.5% and the neurons were fixed
for 60 rain. The cell suspension was collected in a disposable syringe, filtered with a Millipore filter (0.22 gm) and postfixed for
15 rain by injecting 2% OsO 4 into the Millipore filter. Then PBS
and a graded series of ethanol were injected into the Millipore filter, and finally the neurons on the filter membrane were air-dried,
sputter-coated with platinum and examined as described above. A
total of 500 neurons were observed randomly for each stage of
chick and rat respectively and neurons were classified according to
the morphological criteria described previously (Matsuda and
Uehara 1984). During observation in the SEM, the specimens
were moved in a zigzag manner to avoid double counting of neurons.
Results
S c a n n i n g e l e c t r o n m i c r o s c o p y o f intact d e v e l o p i n g
c h i c k dorsal r o o t g a n g l i a
Until d a y E8, m o s t c h i c k D R G neurons w e r e spindleshaped; a few h a d an e c c e n t r i c a l l y b u l g i n g c y t o p l a s m .
T h e angle f o r m e d b y the extensions o f the central and
p e r i p h e r a l p r o c e s s e s o f these n e u r o n s was greater than
90 deg and the central p r o c e s s was thinner than the per i p h e r a l one (Fig. l a ) . B y d a y E l 3 , the two p r o c e s s e s app r o a c h e d each other as the cell b o d y b u l g e d m o r e in a direction o p p o s i t e to the initial s e g m e n t s o f the p r o c e s s e s .
T h e extensions o f these p r o c e s s e s f o r m e d an a n g l e less
than 90 deg, and thus the p r o c e s s e s and cell b o d y e x h i b ited a b e l l - s h a p e d c o n f o r m a t i o n (Fig. l b ) . A t the s a m e
d e v e l o p m e n t a l stages, there w e r e s o m e m o r e a d v a n c e d
g a n g l i o n n e u r o n s w h o s e cell b o d i e s e l o n g a t e d b e t w e e n
478
the initial segments of two processes to form a primitive
stem process (Fig. lc). At first, the stem processes were
larger in diameter than the sum of the central and peripheral processes, shorter in length than the diameter of the
cytoplasm and gradually became thinner as the cytoplasm elongated (Fig. ld). Two days after hatching, the
stem process was so long that the bifurcation of the central and peripheral processes was hardly visible in the
scanning electron microscope (Fig. ld). Besides the mature pseudo-unipolar neurons, immature ganglion neurons of various shapes as described above were also noted in the chick DRG at this stage (Fig. ld), in contrast to
the rat DRG, which contains only mature pseudo-unipolar neurons at birth (Fig. le and Matsuda and Uehara
1984).
Scanning electron microscopy of dissociated
chick ganglion neurons
The dissociation of chick ganglion neurons made it easy
to observe the morphological characteristics of large
numbers of neurons with the scanning electron microscope; they were classified into spindle-shaped bipolar
neurons, eccentrically bulging bipolar neurons (Fig. 2a),
bell-shaped bipolar neurons (Fig. 2b), short-stem pseudo-unipolar neurons and long-stem pseudo-unipolar neurons (Fig. 2c). A few multipolar neurons, which were
possibly less mature than the other types of neurons,
were also observed. The proportions of different types of
chick neurons observed at different developmental ages
are presented in Table 1.
Although 31% of spindle-shaped bipolar neurons
were seen on incubation day E6, they abruptly declined
in number on day E8. By day El0, eccentrically bulging
bipolar neurons were most numerous, ranging from 66%
to 73%. Thereafter, they decreased markedly in number,
while long-stem pseudo-unipolar neurons dramatically
increased in number to 53% on day El4, 72% on day
E l 8 and 82% at 2 days after hatching. The percentages
of bell-shaped bipolar neurons and short-stem pseudounipolar neurons remained relatively constant during
prenatal life, but diminished two days after hatching. On
postnatal day P2, 13% of the chick ganglion neurons
were immature.
Table 1 Percentages (rounded to the nearest integer) of different
neuron types at successive stages of development in chick dorsal
root ganglia determined from 500 dissociated neurons at each
stage
Embryonic day
6
8
10
14
18
P2
Neuron type
Multipolar
Spindle-shaped
bipolar
Eccentrically
bulging bipolar
Bell-shaped
bipolar
Short-stem
unipolar
Long-stem
unipolar
Total unipolar
3
5
0
1
0
0
31
12
5
2
1
1
66
69
73
27
10
6
0
14
12
13
11
6
0
0
5
4
6
5
0
0
5
53
72
82
0
0
10
57
78
87
Table 2 Percentages (rounded to the nearest integer) of different
neuron types at successive stages of development in rat dorsal root
ganglia determined from 500 dissociated neurons at each stage
Embryonic day
14
15
16
17
18
19
P1
Neuron type
2
Multipolar
Spindle-shaped
4
bipolar
Eccentrically
56
bulging bipolar
Bell-shaped
31
bipolar
Short-stem
4
unipolar
Long-stem
3
unipolar
Total unipolar
7
0
1
0
0
0
0
3
4
0
1
0
0
30
18
13
2
1
1
46
47
24
8
5
2
6
9
8
7
12
5
15
21
55
82
82
92
21
30
63
89
94
97
~ lO0
~ffi
8o]
"4
2
40
t1#
20
Scanning electron microscopy of dissociated
rat ganglion neurons
The proportions of different types of rat neurons observed at different developmental ages are presented in
Table 2. Several marked differences were noted between
the dissociated chick and rat ganglion neurons during development. First, pseudo-unipolarization began to take
place during E 8 - E 1 0 in the chick and around E l 4 in the
rat (Fig. 3). Thus, 50% of the neurons were unipolar at
about day E13.5 and 16.5 in chick and rat respectively
(Fig. 3). Second, the percentage of long-stem pseudounipolar neurons increased abruptly in the prenatal rat
rat
chick
0<
6
7
7
8
8
9
9
10 11 12 13 14 15 16 17 18 19 20 21 P1
10 11 12 13 14 15 16 117 18 19 20 PI P2(days)
Fig. 3 Percentages (rounded to the nearest integer) of unipolar
neurons at successive stage of development in chick ( 9 and rat
(0) dorsal root ganglia, determined from 500 dissociated neurons
at each stage
after day El6, whereas in the chick it increased relatively
gradually after day E l 0 (Fig. 3). Third, the percentage of
bell-shaped bipolar neurons was higher in the rat during
E14-E17 than in the chick throughout the embryonic period (Tables 1, 2). Last, very few immature neurons (3%
479
not unipolar) were found in the rat ganglia 1 day after
birth, whereas in 2-day-old chicks many immature neurons (13% not unipolar) were still present.
Discussion
Scanning electron microscopy revealed that pseudo-unipolarization takes place in a similar way in chick and rat
dorsal root ganglia; in both species, the cytoplasm between the central and peripheral processes of a bellshaped bipolar neuron elongates at certain developmental
stages to form a primitive stem process (Matsuda and
Uehara 1984). This appears to be a key event in the transformation of a bell-shaped bipolar neuron into a shortstem pseudo-unipolar neuron. No sign of close apposition
of the central process to the peripheral one was noted in
any of the neurons examined, and the primitive stem processes were always thicker than the sum of the diameters
of the two individual processes. From these findings, one
may exclude the possibility that the two processes contact
and fuse with each other to yield a common stem process.
A further elongation of the primitive stem process without fusion of the central and peripheral processes appears
to result in the formation of a long-stem pseudo-unipolar
neuron (Cajal 1883; His 1887; Murnaghan 1941; Matsuda and Uehara 1984; Ninomiya 1985). The molecular
basis for pseudo-unipolarization remains to be determined, although Schwann (satellite) cells are likely to
play a key role in the transforming event (Pannese I98I;
Mudge 1984; De Koninck et al. 1993).
Comparison of normal tables of chick and rat show
that chicks develop earlier than rats in terms of their general body development. In agreement with this, we found
that the rapid rise in the percentage of mature unipolar
neurons occurred approximately 3 days earlier in the
chick than in the rat. On the other hand, we found that
the development of mature neurons reached a plateau
more slowly in chicks. Thus, at birth, immature neurons
in the chick outnumbered those in the rat. Another reflection of this was that in their late stages of development, dissociated ganglia had a greater percentage of eccentrically bulging bipolar neurons and a smaller percentage of bell-shaped bipolar neurons in chick than in
rat embryos. In other words, although DRG neurons matured earlier in the chick than in the rat, they did so more
gradually.
These findings suggest that some neurons remain in a
more immature condition for longer during prenatal development in chicks than in rats. A possible explanation
for this difference may be that the extent and intimacy of
satellite cell investment is more developed in the rat than
in the chick (Pannese 1974, 1981). It is likely that in the
chick, the estimated 13% immature neurons at posthatching day 2 continue developing into mature forms, because such bipolar neurons comprise only about 2% of
the total in adult chick sensory ganglia (Smith 1913;
Truex 1939).
In general, the results of the present study on dissociated rat neurons agree with the previous study on intact
ganglia (Matsuda and Uehara 1984). However, the percentages of eccentrically bulged bipolar neurons in the
dissociated rat DRG on days El5 and El6 (30% and
18%) were somewhat lower than in the superficial layers
of intact ganglia at the same times (66% and 31%; Matsuda and Uehara 1984). In contrast, bell-shaped bipolar
neurons in the dissociated DRG on days El5 and El6
were more numerous than those in the corresponding intact DRG. These findings suggest that transition from eccentrically bulging bipolar neurons to bell-shaped bipolar neurons takes place frequently in the central portion
of the embryonic ganglion, confirming earlier observations (Cajal 1904). However, long-stem pseudo-unipolar
neurons seem to be generated predominantly in the superficial portion, since a higher incidence of long-stem
pseudo-unipolar neurons was noted in intact ganglia than
in dissociated neurons. Taken together, these observations suggest that individual steps in ganglion cell transformation may depend on the location from which the
neurons are derived or to which they migrate during embryogenesis.
In conclusion, the present study showed that the development of pseudo-unipolar neurons is similar in chick
and rat embryonic dorsal root ganglia, except that immature neurons type are seen more frequently in the chick
than in rat during the perinatal period.
Acknowledgements The authors are grateful to Prof. Yasuo Uehara, Department of Anatomy, Kumamoto University School of
Medicine for his excellent advice. They also thank Miss M. Fujimoto for secretarial assistance.
References
Cajal SR y (1883) Neue Darstellung vom histologischen Bau des
Centralnervensystems. Arch Anat Physiol Anat Abt 319-428
Cajal SR y (1890) Sin l'origin et les ramification des fibres nerveuses de la molle embryonnaire. Anat Anz 5:85-95
Cajal SR y (1904) Association del rnetodo del nitrato de plata con
el embyonario. Para el estudio de los focos motores y sensitivos. Trab Lab Invest Biol Univ Madrid 3:65-96
De Koninck P, Carbonetto S, Cooper E (1993) NGF induces neonatal rat sensory neurons to extend dendrites in culture after
removal of satellite cells. J Neurosci 13: 577-585
Hamburger V, Hamilton HL (1951) A series of normal stages in
the development of the chick embryo. J Morphol 88:49-92
His W (1886) Zur Geschichte des menschlichen Rtickenmarks und
der Nervenwurzeln. Abh Math-Phys K1 Kgl Ges Wiss 13:
479-513
His W (1887) Die Entwicklung der ersten Nervenbahnen beim
menschlichen Embryo. Arch Anat Physiol Anat Abt 369-378
Lenhossek M (1892) Beobachtungen an der Spinalganglien und
dem Rtickenmark yon Pristiurusembryonen. Anat Anz 7:
519-539
Lieberman AR (1976) Sensory ganglia. In: Landon D (ed) The peripheral nerve. Chapman & Hall, London, pp 188-278
Matsuda S, Uehara Y (1981) Cytoarchitecture of the rat dorsal
root ganglia as revealed by scanning electron microscopy. J
Electron Microsc 30:136-140
Matsuda S, Uehara Y (1984) Prenatal development of the rat dorsal root ganglia. A scanning electron-microscopic study. Cell
Tissue Res 235:13-18
Mudge AW (1984) Schwann cells induce morphological transformation of sensory neurones in vitro. Nature 309:367-369
Murnaghan DP (1941) Studies on living spinal ganglion cells.
Anat Rec 81:183-203
480
Ninomiya T (1985) Morphological changes of rat dorsal root ganglion cells in the process forming period. Sapporo Med J 54:
1-10
Pannese E (1968) Developmental changes of the endoplasmic reticulum and ribosomes in nerve ceils of the spinal ganlglia of
the domestic fowl. J Comp Neurol 132:331-364
Pannese E (1974) The histogenesis of the spinal ganglia. (Advances in anatomy, embryology and cell biology, vol 47) Springer,
Berlin Heidelberg New York
Pannese E (1981) The satellite cells of the sensory ganglia. (Advances in anatomy, embryology and cell biology, vol 65)
Springer, Berlin Heidelberg New York
Riederer BM, Barakat-Walter I (1992) Differential distribution of
microtubule-associated proteins, MAP 2 and MAP 5, during
chick dorsal root ganglion development in situ and in culture.
Brain Res Dev Brain Res 68:111-123
Smith EV (1913) Histology of the sensory ganglia of birds. Am J
Anat 14:251-297
Tennyson V (1965) Developing sensory neuroblasts. J Comp Neurol 124:267-318
Truex R (1939) Observations on the chicken Gasserian ganglion
with special reference to the bipolar neurons. J Comp Neurol
71: 473-486