вход по аккаунту



код для вставкиСкачать
Remodeling of Lesioned Kitten Visual
Cortex After Xenotransplantation
of Fetal Mouse Neopallium
1Department of Psychology, Life Sciences Center, Dalhousie University,
Halifax, Nova Scotia, B3H 4J1 Canada
2Department of Anatomy and Neurobiology, Dalhousie University,
Halifax, Nova Scotia, B3H 4H7 Canada
Remodeling of the mechanically injured cerebral cortex of kittens was studied in the
presence of a neural xenograft taken from mouse fetuses. Solid neural tissue from the
neopallium of a 14-day-old fetus was transferred into a cavity prepared in visual cortical area
18 of 33-day-old kittens. Injections of bromodeoxyuridine (BrdU) were used to monitor
postoperative cell proliferation. Two months after transplantation, the presence of graft tissue
in the recipient brain was assessed by Thy-1 immunohistochemistry. Antibodies specific for
neurons, astrocytes, and oligodendrocytes and hematoxylin staining for endothelial cells were
used for the characterization of proliferating (BrdU1) cells.
The following were the major observations: 1) Of ten transplanted kittens, four had the
cavity completely filled with neural tissue that resembled the intact cerebral cortex in its
cytoarchitecture, whereas, in four other kittens, the cavity was partially closed. In two
kittens, the cavity remained or became larger, which was also the case with all four
sham-operated (lesioned, without graft) animals. 2) A substantial part of the remodeled tissue
was of host origin. Only a few donor cells survived and dispersed widely in the host
parenchyme. 3) In the remodeled region of transplanted animals, the densities of nerve, glial,
and endothelial cells were similar to those in intact animals. 4) Cell proliferation increased
after transplantation but only within a limited time, because, 2 months after the operation,
the number of mitotic cells in the grafted cerebral cortex did not differ from that in intact
Our data suggest that the xenograft evokes repair processes in the kitten visual cortex
that lead to structural recovery from a mechanical insult. The regeneration seems to rely on a
complex interplay of many different mechanisms, including attenuation of necrosis, cell
proliferation, and immigration of host cells into the wounded area. J. Comp. Neurol.
395:91–111, 1998. r 1998 Wiley-Liss, Inc.
Indexing terms: neural transplantation; regeneration; plasticity; cerebral cortex; cat
The regenerative capacity of the central nervous system
(CNS) in mammals has become a major focus of much
contemporary research in basic as well as in applied
neurobiology. The long-held assumption that repair of the
injured CNS is virtually impossible has been called into
question by investigations that explore the extent of repair
in diverse animal species at different developmental stages
and also by studies that attempt to induce regeneration
through application of external agents (for reviews, see
ffrench-Constant and Mathews, 1994; Logan et al., 1994;
Varon and Conner, 1994).
Susceptibility of the CNS to injury and its subsequent
capacity to heal and regenerate may vary considerably
through an animal’s lifetime, particularly during the various stages of development. Notably, in this respect, Kolb
and coworkers (Kolb, 1987; Kolb and Gibb, 1991; Kolb et
al., 1994) reported that in 1–6-day-old rats large lesions of
the frontal or parietal neocortex had much more deleteri-
Grant sponsor: Natural Sciences and Engineering Research Council of
Canada; Grant number: 7660.
*Correspondence to: Dr. J. Ourednik, Swiss Federal Institute of Technology, Department of Neurobiology-HPM, Hönggerberg, 8093 Zürich, Switzerland. E-mail:
Received 30 April 1997; Revised 9 January 1998; Accepted 20 January
ous anatomical and functional consequences than similar
insults performed in 7–12-day-old animals, in which remarkable functional and structural reorganization was
observed. To date, the reasons for the different vulnerability and capacity for recovery in the CNS at different
phylogenetic and ontogenetic stages of development are
largely unknown. Important factors that may dictate the
regenerative capability of the CNS at a defined developmental period include the actual rate of cell proliferation
(especially of neuronal precursors) and the degree of cell
differentiation that has been attained. To a large extent,
these events are under the control of growth-promoting
and inhibitory substances, the presence and ratio of which
have been shown to vary during development. A recent
striking example of the role of growth factors was the
demonstration that cells dissociated from the subventricular zone of the adult mouse brain could be stimulated to
differentiate in vitro when epidermal growth factor (EGF)
or basic fibroblast growth factor (bFGF) were added to the
culture medium (Reynolds and Weiss, 1992; Reynolds et
al., 1992; Richards et al., 1992; Vescovi et al., 1993; Gritti
et al., 1995, 1996).
A promising line of research pursued during the last
decade has examined the potential of neural grafts to
reduce or even eliminate the negative sequelae of injury in
the CNS. The transfer of fetal neural tissue into a wounded
brain region, under certain circumstances, may lead to at
least partial repair and compensation for the induced
structural and functional defects (for reviews, see Berry,
1989; Fischer and Gage, 1993; Dunnett and Björklund,
1994). This beneficial effect is virtually always attributed
to the proliferation of graft-derived tissue and to its
integration into the host’s CNS. Explorations of the contributions from an alternative mechanism, namely, that the
host tissue may be stimulated by the presence of the graft
to participate actively in the regenerative and proliferative
processes, have been attempted only rarely (Lundberg and
Møllgård, 1979; Polezhaev et al., 1988). More recently, our
own studies have shown that neural tissue isolated from
neuroepithelium of the embryonic CNS of the mouse can
effectively stimulate repair mechanisms within the lesioned brain of a juvenile host from a genetically closely
related mouse strain (Ourednik et al., 1993a,b; Ourednik
and Ourednik, 1994). Similar regeneration-inducing and
remote effects of grafts were also reported by Valoušková
and Gálik (1995), who performed bilateral lesions in the
frontal neocortex of adult rats but placed fetal neural
grafts into only one of the two cavities. Not only was the
lesion on the grafted side refilled with neural tissue, but
the contralateral, nongrafted cavity was also significantly
reduced in size.
The purpose of the present study was to investigate
whether the stimulating influence of a fetal neural transplant on CNS regeneration that had been observed previously with two congenic mouse strains as donor and host
would also be apparent with grafts between two species
that were evolutionary more remote. We chose to produce
lesions in the kitten primary visual cortex, a well-known
model system for studies of developmental plasticity of the
central visual pathways, and to implant the lesion with
neural grafts prepared from the fetal mouse telencephalon. The chosen age of the host animals (4–5 weeks) was at
a time when neurogenesis in the cat visual system was
complete (Luskin and Shatz, 1985) but still before the time
of anatomical and functional maturation in the visual
TABLE 1. Degree of Cavity Remodeling in Transplanted
and Sham-Operated Animals1
day of death
of lesion area
cells per five
(level I–V)
1Five degrees of remodeling of the cavity were defined as follows: complete filling with
exuberant tissue, as seen in Figure 5 (11); complete filling without exuberant tissue
(1); partial filling (1/2); persistent cavity of about the same size as originally prepared
(2) and cavity evidently bigger than the original one (—). Numbers of Thy-1.21 cells
reflect the absolute numbers of cells observed at levels I–V (see Fig. 1). m, Male; f,
female; n.a., not applicable; level I–V, frontal planes of cutting (see Fig. 1).
cortex (Cragg, 1975; Mitchell and Timney, 1984; Shatz and
Luskin, 1986; Dyck et al., 1993; Price et al., 1994; Daw,
1995). The age coincided with the time of maximal activitydependent plasticity, when the kitten visual cortex is most
susceptible to modification by experiential manipulations
of the visual input (Mitchell and Timney, 1984; Daw, 1995).
Use of the mouse as donor permitted a more simple and
plentiful supply of neural graft tissue than that supplied
from cat fetuses, but, more importantly, it allowed the use
of anti-Thy-1 immunohistochemistry for later identification of grafted and host tissue (Charlton et al., 1983; Zhou
et al., 1985; McKeon et al., 1989; Ourednik et al., 1993a;
Ourednik and Ourednik, 1994), because the antigen’s
isoforms differ in cat and mouse (Morris and Grosveld,
1989; Saleh and Bartlett, 1989). Although Thy-1 is not
fully expressed in all brain areas during early development, in the adult CNS, it is present on the surface of most
neuronal cell bodies and their processes and on astroglial
cells (for review, see Morris and Grosveld, 1989).
To allow a direct comparison with our previous allografting experiments (Ourednik et al., 1993a,b), in which no
immunosuppressive drugs were used, we performed the
present xenotransplantation in the absence of these agents
as well. Pilot experiments on three kittens (Table 1,
kittens 1–3) showed that 2-month-old grafted lesions were
partially (two kittens) or completely (one kitten) closed
and that the repair could occur without immunosuppression.
Transplantation of fetal tissue into a cat brain has been
reported only rarely (Müller and Best, 1989; Villablanca et
al., 1990; Yinon and Gelerstein, 1991; Yinon et al., 1993;
Targett and Blakemore, 1994), and these reports have
provided only scant descriptions of the operative techniques that are more demanding than those required for
grafting into rodent CNS. Therefore, an important secondary aim of the present study was to establish and describe
extensively a suitable surgical and transplantation procedure that could be employed in kittens.
Two sets of data were collected and analyzed. The first set
of data, which is documented below in Part A, describes
the remodeling of the grafted wound and the extent to
which the graft participated in the repair. The second set of
data, which is described below in Part B, provides information concerning the contribution of specific cell types
(neurons, astrocytes, oligodendrocytes, and endothelial
cells) to the remodeling of the lesion (evaluations of
changes in related subcortical structures were not included in this study). To this end, densities of these cell
populations were determined by using the appropriate cell
type-specific immunohistochemical markers, namely, neuron-specific nuclear protein A60 (NeuN) for neurons, glial
fibrillary acidic protein (GFAP) for astrocytes, cyclic nucleotide phosphohydrolase (CNP) for oligodendrocytes, and
two different isoforms of the Thy-1 antigen for the detection of donor and host cells. Endothelial cells were identified in hematoxylin-stained sections on the basis of their
morphology. In combination with these cell type-specific
markers, we used postoperative bromodeoxyuridine (BrdU)
labeling of dividing cells and a BrdU-specific antibody to
monitor the postoperative proliferation of the various cell
types of interest. Collectively, these data permitted the
evaluation of the relative contribution of two possible
processes to the filling of the cavity, namely, rearrangement of existing cells in the vicinity of the lesion vs.
postoperative proliferation among either donor or specific
host cell populations.
Donors. Adult C57BL/6J mice were mated overnight,
and the day of vaginal plug detection was defined as
embryonic day 0. Transplants were obtained from 14.5-dayold (60.5 day) fetuses that were examined carefully under
a dissecting microscope for their overall condition and
precise stage of development according to the criteria
specified by Kaufman (1992). Only healthy and not developmentally retarded fetuses were used.
Recipients. The ten host animals were 33-day-old
kittens (61 day) from an outbred, nonspecific, domestic cat
lineage. The day of birth was considered to be day 0. At the
time of operation, the average weight of the kittens was
470 6 80 g for males and 360 6 50 g for females.
Controls. Two types of age-matched control animals
were prepared in parallel with the animals that received a
graft: 1) sham-operated kittens (n 5 7) that received the
lesion but no graft and 2) intact kittens (n 5 3).
A total of 20 kittens from seven litters of different queens
were used (see Table 1). The experiments conformed to the
strict guidelines regarding the care and use of animals
specified by the Society for Neuroscience (1991) and the
Canadian Council on Animal Care (1993/4). Cat and mouse
colonies were held on a standard diet, maintained on a
12-hour light-dark regime, and inspected regularly several
times each day.
Surgical procedures
Operations were performed under sterile conditions
with the help of a dissecting microscope. Kittens were
injected with xylazine (Rompun; Bayer, Etobicoke, Ontario, Canada; 2.2 mg/kg i.m.) and anesthetized with
gaseous halothane (Fluothane-Halothane B.P.; WyethAyerst, Montreal, Quebec, Canada) mixed with oxygen
(flow 1,500 mm3/minute) in a Fluotec 3 container (CDMV;
Fluotec, St. Hyacinthe, Quebec, Canada). Inhalation
started with the gas containing 5% of halothane for about 2
minutes to effect, then the halothane concentration was
reduced to 1.5–2.0%, which was maintained during the
rest of the operation. Once the animal was sedated, the
hair covering the right side of the neurocranium was
shaved, and the skin was swabbed with an antiseptic/
antibacterial solution (1:30 dilution of Zephiran chloride;
Winthrop Laboratories, Aurora, Ontario, Canada) in distilled water (dH2O) containing 0.08% chlorhexidine gluconate (Hibitan; Ayerst Laboratories, Montreal, Quebec,
Canada) and with the surgical microbicidal skin cleanser
Betadine (Purdue Frederick Inc., Toronto, Ontario,
Canada). Finally, three drops (approximately 150 µl in
total) of the local anesthetic Marcaine (Sanofi Winthrop,
Markham, Ontario, Canada) were applied to the skin of
the scalp before surgery.
A 2-cm-long parasagittal incision of skin and subcutaneous tissue was performed 0.5 cm to the right of the midline
of the scalp, extending approximately from the external
occipital protuberance to the anterior edge of the pinna.
The skin was held back with a small retractor, and the
pericranium above the chosen brain region was cut in
anteroposterior direction, scraped laterally from the bone
with a spatula, and then cut away. With a high-speed
microdrill, four small holes were drilled (0.9 mm drill bit
diameter) that defined the corners of a small rectangular
window in the bone (6 3 5 mm) centered 8 mm rostral to
lambda and 4 mm to the right of the midline and overlying
the area of the lateral sulcus where part of the visual
cortical area 18 is represented (see Fig. 1). The bone
between the microdrill perforations was cut with a small
circular saw (16 mm diameter; FST, North Vancouver,
British Columbia, Canada), while the region was moistened constantly and cooled with drops of sterile physiological solution. The removed piece of bone was flipped over
and kept beside the prepared cranial opening until it was
Preparation of the lesion. The exposed meninges
were gently cut off, and, after the bleeding stopped spontaneously (generally within 5–10 minutes), the area was
cleaned with small pieces of sterile gauze and sterile
saline. A small rectangular block of brain tissue measuring
3 mm in the mediolateral direction and 4 mm in the
anteroposterior direction and measuring 1.3–1.5 mm in
depth was delineated by four perpendicular cuts with a
microsurgical, hook-shaped knife. The block of tissue
(18.0 6 1.5 mm3) was then undercut and withdrawn with a
custom-made, tiny, spatula-shaped knife. The cavity was
carefully cleaned of blood by using dental absorbent paper
points and with very gentle aspiration by using Pasteur
pipettes drawn out to an outer tip diameter of ca. 0.3 mm
and connected to an electrical vacuum pump (negative
pressure set to a maximum of 100 mm Hg). Utmost care
was taken not to damage the walls of the cavity by the
suction. Finally, the cavity was rinsed with Dulbecco’s
modified Eagle medium (D-MEM; catalog no. 11965–043;
Gibco BRL, Grand Island, NY), and the same was used
later during the dissection of the mouse embryos and the
preparation of the graft tissue. The position and extent of
the cavity in the kitten visual area 18 are shown in Figure 1.
Grafting. In parallel with cleaning of the cavity in the
host kitten brain, fresh graft tissue was prepared as
follows. Mouse dams were deeply anesthetized with an i.p.
Fig. 1. Position and extent of the lesion in the kitten visual cortex.
Top: Dorsal view of the occipital region of the cerebral hemispheres of
a 3-month-old kitten. A schematic representation of visual areas 17,
18, 19, and 21a is projected onto the left hemisphere, whereas the
location of the lesion is shown to the right (box). Roman numerals I–V
indicate coronal levels from which sections were chosen for analysis.
Bottom: Hematoxylin-stained, 20-µm frontal section from level III
with the position and extent of the lesion indicated in the right
hemisphere (box). Asterisk indicates the Sylvian fissure. Scale bar 5
6 mm.
injection (10 µl/g body weight) of a mixture of sodium
pentobarbital (11% Somnotol; MTC Standard Pharmaceuticals, Cambridge, Ontario, Canada) diluted with 38%
propylene glycol in dH2O, giving an anesthetic dose of 75
mg of sodium pentobarbital per kg body weight. They were
then placed on their backs, and their abdomens were
swabbed with Zephiran chloride (see above). A 1-cm-long,
transverse incision was made in their lower abdomen to
allow access to the uterus.
Fetuses were removed one by one (starting from the
uppermost uterine corner), as needed. Only freshly prepared grafts were used for transplantation, and each fetus
served to prepare only one transplant. After a fetus was
removed, the incisions of uterus and abdomen were clamped
with hemostats, and the abdomen was covered with a piece
of sterile gauze. The total time during which the dam was
kept under narcosis was about 3–6 hours, depending on
the number of transplanted animals operated that day.
Additional small doses of the Somnotol mixture were
injected as required.
The removed fetus was rinsed in D-MEM (see above),
liberated from its embryonic envelopes, and killed by a cut
through the cervical spinal cord. The subsequent dissection and preparation of the graft was performed with the
aid of tiny forceps and microscissors in the same medium
in a small, concave, black dish placed under the dissecting
microscope. The cartilaginous fetal skull was opened to-
gether with the overlying skin by one cut in its midsagittal
plane and two lateral cuts extending from each eye back to
the cervical spinal cord. All meninges covering the cerebral
hemispheres were peeled off, and the brain was transferred into fresh medium. A small block of tissue (1.5 6
0.25 mm3), including the whole thickness of the cortical
wall, was cut from the right neopallium. The graft was
rinsed in fresh medium, placed on the round tip of a glass
rod (0.5 mm diameter), and transferred immediately,
without any particular orientation, to the bottom of the
cleaned recipient cavity. Only one block of graft tissue was
placed in each cavity.
Conclusion of the operation. The cranial window was
closed with the corresponding piece of bone, and the area
was cleaned of blood and rinsed with physiological saline.
For suturing of the overlying connective tissues and skin,
resorbable threads (4–0 Dexon II; Cyanamid Inc., Montreal Quebec, Canada; four to six stitches) and nonresorbable threads (4–0 Novafil; Cyanamid Inc.; six to nine
stitches) were used. The sutured skin was cleaned with
pieces of gauze soaked in sterile saline and scrubbed with
Betadine. One operation lasted approximately 80 6 30
minutes, after which, the kitten was disconnected from the
anesthetic apparatus and placed in a plastic cage maintained at 36.5°C. Once it was awake, after about 1.0–1.5
hours, the kitten was returned to its mother. Neither
prophylactic antibiotics nor immunosuppressive drugs were
ever administered to the animals.
Monitoring of postoperatively dividing cells
To monitor postoperatively proliferating cells, kittens
were injected i.p. with 58-bromodeoxyuridine (BrdU; 50
mg/kg body weight; Sigma, Oakville, Ontario, Canada)
administered at a concentration of 10 mg/ml in physiological saline containing 0.007 N NaOH (Takahashi et al.,
1992). Because the availability of the injected BrdU is time
limited (20–40 minutes; Boswald et al., 1990), the animals
were injected repeatedly in order to increase the probability of its incorporation by a high number of dividing cells.
The first injection occurred 48 hours after operation, and
the following ones were administered every other day, for a
total of six injections per animal. Age-matched, shamoperated and intact kittens were injected in parallel.
Perfusion and histology
All animals were killed 61 6 2 days after operation,
when the average weight of the kittens was 1,860 6 450 g
for males and 1,510 6 180 g for females. Three additional
sham-operated kittens were perfused immediately after
the operation in order to observe the extent and variability
of the fresh lesions (Table 1, kittens 15–17). The animals
were injected intrahepatically with an overdose of sodium
pentobarbital (Euthanyl; Canada Packers Ltd., MTC Pharmaceuticals Division, Mississauga, Ontario, Canada; 2
ml/4.5 kg body weight) and, once they were in deep
anesthesia, they were perfused through the left heart
ventricle for 7 minutes with 0.1 M phosphate-buffered
saline (PBS), pH 7.4 (100 ml/minute flow rate maintained
with a Masterflex Pump; Cole-Palmer Instrument Co.,
Chicago, IL).
The skull was opened with an electrical circular saw (16
mm diameter) and a rongeur, and the forebrain anterior to
the tentorium was removed. Because of the risk of mechanical damage, about 1.5 cm2 of bone were left attached
temporarily above the operated area. The frontal lobes
were cut off, and the rest of the brain was incised midsagittally and immersed into acidic alcohol (5% of acetic acid in
70% ethanol) for 2 days. Following fixation, the remaining
bone piece was removed very carefully by using a rongeur,
microsurgical scissors, forceps, and a magnifying glass.
The brain tissue was dehydrated in successively higher
concentrations of ethanol, beginning with an 80% solution
(30 hours; two or three baths), followed by immersion in a
95% solution (4 hours; three or four baths), and, finally,
absolute ethanol (4 hours; three or four baths). After
dehydration, the tissue was embedded in polyester wax
(BDH; Poole, Dorset, United Kingdom; 2.5 days; three
baths at 38°C).
Coronal 20-µm sections were cut on a microtome from
anterior level I to posterior level V, as indicated in Figure
1, with a distance of 3–4 mm between each level. These
levels correspond approximately to the Horsley-Clarke
coronal planes P5.4 to A4.2 of the adult cat brain (Tusa et
al., 1981). With the additional help of the atlas by Snider
and Niemer (1961), the block of tissue was oriented
carefully in the coronal plane to ensure equivalent cutting
levels for both hemispheres. Complete series of about
150–300 sections per animal were collected on slides
coated with 1% gelatine containing 1% formol or with
Vectabond (Vector Laboratories, Burlingame, CA) and
were stored at room temperature (RT) until further use.
Just before immunostaining, sections were dewaxed by
successive immersion for 3 minutes in two baths of absolute ethanol (they were air dried for about 30 seconds after
the first bath to improve their attachment to the glass
slides), rehydrated, and washed twice for five minutes
each in PBS. The sections were used immediately for
immunohistochemical staining without being dried.
Cell identification
and immunohistochemistry
To analyze cortical cytoarchitecture, immunostaining
with cell type-specific antibodies (Abs) was conducted on
sections cut from coronal levels I–V (see Fig. 1). To follow
the extent to which the postoperative cell proliferation
varied among the different cell types, double immunostaining was conducted on sections cut from coronal level III.
The first immunoreaction was always against BrdU, followed by intensive washing in PBS (1–2 hours), and then
by incubation with a cell type-specific Ab.
The Abs and procedures for the various immunoreactions are described separately below. Negative controls
(the primary Ab replaced by the diluent) were always
processed in parallel with the immunoreactions for each
Cells labeled with BrdU were detected according to a
modified version of the protocol of Takahashi et al. (1992).
Sections were hydrolyzed at RT in 2 N hydrochloric acid
(HCl). In a test using acridine orange (Von Bertalanffy and
Nagy, 1962), we established an optimal denaturing time of
25–35 minutes that provided a balance between maximal
production of single-stranded DNA (to improve accessibility of the incorporated BrdU to the Ab) and maintenance of
good cytoarchitecture. After hydrolysis, sections were rinsed
for 1 minute in PBS (0.1 M), pH 6; blocked for 30 minutes
in a solution of 10% normal horse serum (NHS) containing
1% H2O2; and then incubated for 45 minutes at RT with
the monoclonal rat anti-BrdU Ab (clone BU1/75; ICR1;
Sera-Lab, Crawley Down, United Kingdom) diluted 1:75 in
PBS (0.05 M), pH 7.4, containing 0.5% Tween 20. Detection
of the bound primary Ab with a biotinylated rabbit anti-rat
immunoglobulin (Ig) was conducted in accordance with the
instructions of the Vectastain kit (Elite Vectastain; Vector
Laboratories). The final dark-blue reaction product
was obtained in a nickel/cobalt-3,38-diaminobenzidinetetrahydrochloride (Ni/Co-DAB) solution: 15 ml PBS
with 10 mg DAB, 3 mg nickel-ammonium sulphate
(NiSO4[NH4]2SO4.6H2O), 3.5 mg cobalt chloride
(CoCl2.6H2O), and 1 µl 30% hydrogen peroxide (H202).
Some of the immunostained sections were counterstained
with Mayer’s hematoxylin (Stevens, 1990).
Murine Thy-1. The following protocol was used to
detect the Thy-1.2 antigen expressed on the cells xenografted from C57BL/6J mice. After inactivation of endogenous peroxidase in 3% H2O2 for 15 minutes and washing
in PBS (0.1 M), pH 7.2, sections were blocked for 30
minutes in 1% fetal calf serum (FCS) and 0.5% bovine
serum albumin (BSA). The monoclonal rat anti-mouse
Thy-1.2 Ab was diluted 1:100 (clone 30-H12; Becton Dickinson, Mountain View, CA) or 1:50 (Pharmingen, San
Diego, CA) in 20% FCS in PBS, and it was left to react
overnight at 4°C. After washing, the sections were incubated for 2.5 hours at RT with a horseradish peroxidase
(HRP)-conjugated rabbit anti-rat Ig (Dako, Glostrup, Denmark) diluted 1:40 in 20% FCS in PBS. The final color
reaction in the presence of DAB was controlled under a
microscope and intensified by replacing PBS with an
imidazole buffer (Morris, 1991). The reaction was usually
stopped within 5–10 minutes in running H2O. To assure
specificity of the Thy-1.2 immunostaining, sections from
nontransplanted (sham-operated and intact) kitten brains
and sections from 2-month-old C57BL/6J mouse brains
(Thy-1.21) were assayed as well.
Feline Thy-1. To assess the presence of host cells in the
remodeled area, after blocking, sections from transplanted
animals were incubated with a monoclonal mouse anti-cat
Thy-1 Ab diluted 1:1 (Saleh and Bartlett, 1989) in 20%
FCS in PBS and left overnight at 4°C. After washing, the
sections were incubated at RT with a biotinylated rabbit
anti-mouse Ig (dilution 1:150) for 2.5 hours, and visualization occurred according to the protocol provided by the
Vectastain kit. To control the specificity of this immunostaining, sections from C57BL/6J mouse brains were left
to react with the cat-specific Ab in parallel.
NeuN. To detect NeuN1 cells, we used our modification
of a protocol by Mullen et al. (1992; personal communication). Sections were incubated overnight at 4°C with the
monoclonal mouse Ab A60 (Mullen et al., 1992) diluted
1:100 in 10% NHS in PBS (0.1 M), pH 7.2. Specific binding
was detected with a biotinylated horse anti-mouse Ig
(dilution 1:150; 60 minutes at RT) by using Vectastain and
GFAP. Sections were reacted with a polyclonal rabbit
anti-cow GFAP Ab (Dako) diluted 1:1,000 in 10% NHS
supplemented with 0.05 M BSA and 0.1% Triton X-100.
After an overnight incubation at 4°C and washing in PBS,
an HRP-conjugated Ab against rabbit Ig (Dako; dilution
1:40 in 10% NHS/0.1% Triton X-100 in PBS) was applied
for 2.5 hours at RT. After washing, the enzymatic product
was visualized by using a routine DAB stain.
CNP. Sections were incubated overnight at 4°C with
the mouse monoclonal Ab against CNP (Promega, Madi-
son, WI) diluted 1:300 in 10% NHS. An HRP-conjugated
secondary Ab against mouse Ig was applied for 2.5 hours at
RT, and the enzymatic product was visualized with DAB.
Endothelial cells. Endothelial cells could be identified
with ease in sections stained with Mayer’s hematoxylin
(Stevens, 1990) on the basis of their nuclear characteristics (dense staining, elongated and irregular shape) and
their frequent presence in close vicinity to the lumen of a
blood vessel (Fig. 7).
Microscopy and photomicrography
Sections were mounted in Permount (Fisher Scientific,
Halifax, Nova Scotia, Canada) or Entellan (Merck, Darmstadt, Germany) and were analyzed with a Zeiss Axiophot
microscope (Thornwood, NY) or with a Nikon Labophot
microscope (Tokyo, Japan) equipped with a camera lucida
drawing tube. The photomicrographs in Figures 1A, 5A,
and 6 were prepared with a Nikon F-601M camera mounted
on either a dissecting microscope (Wild M650; Leica,
Heidelberg, Germany) or a Nikon Labophot. The photomicrographs in Figures 1B, 2, and 5B,C were taken with a
Leica MPS52 camera mounted on a Leica Wild MZ8
microscope equipped with a Leica EB transmission-light
stand. Figures 3 and 7 were taken with a Zeiss Axiophot
microscope. Photomicrographs were taken on Kodak Technical Pan film (Eastman Kodak, Rochester, NY) at 50 ASA
or on Kodak film EPJ 64T.
Qualitative evaluation. The position and extent of
the lesions in transplanted and sham-operated kittens
were compared for each animal in hematoxylin-stained
sections at all five coronal levels I–V (see Fig. 1). The
camera lucida drawings in Figure 4 illustrate examples of
both fully and partially filled cavities as well as persisting
lesions in the transplanted kittens. For quantitative measurements, these drawings were magnified (approximately 23), equivalent regions of the right and left hemispheres were cut out and weighed, and their weight ratios
(operated/intact; O/I) were determined. The cuts followed
the cortical circonvolutions and included the representations of areas 17, 18, 19, and 21a on the lateral gyrus and
part of the suprasylvian gyrus (compare with Fig. 1; see
also Figs. 1.10 and 1.11 in Tusa et al., 1981). To minimize
errors caused by possible distortion of the operated cortex,
the choice of corresponding coronal levels in the left and
right hemispheres was verified according to subcortical
structures by using the atlas of Snider and Niemer (1961).
Drawings were also made from sections cut through other
areas of the transplanted brains to illustrate the dispersion of the Thy-1.21 graft cells (Fig. 6).
Quantitative evaluation
Thy-1.21 (donor) cells. For every animal, the absolute
number of Thy-1.2-labeled cells was determined from five
brain sections that were taken from all evaluated levels
I–V (Table 1; compare also with Fig. 1A). At each level, the
total section area was evaluated, and the average number
of donor cells was calculated separately (Fig. 6).
Relative cell densities in the operated area. The relative
cell density (numbers of NeuN1, GFAP1, Thy-1.21, and
CNP1 cells as well as endothelial cells per 1 mm2) was
determined in the four animals with completely closed
cavities (kittens 3, 7, 9, and 10; see Table 1), and the
results were compared with data obtained from intact
brains (kittens 18, 19, and 20) and sham-operated brains
(kittens 11, 12, 13, and 14). Transplanted animals with
only partially closed lesions and sham-operated kittens
killed immediately after lesioning were not included in the
Cells were counted in area 18 across the entire thickness
of the right sham-operated and transplanted visual cortices, and the results were compared with those obtained in
an intact cortex (see schematic drawing in Fig. 8). Two
regions were evaluated in sham-operated brains: An area
proximal to the cavity (within 100–200 µm) and a distal
region located 200–1,000 µm from the cavity walls. For
each brain, animal, and cell category, one section from
level III (Fig. 1) was analyzed at 4003 or 6303 magnification (Fig. 8).
Postoperative cell proliferation. Relative cell densities
of double-stained cells (BrdU/NeuN1 BrdU/GFAP1, BrdU/
CNP1, BrdU/Thy-1.21, and BrdU1 endothelial cells) were
calculated in a manner similar to that described above
(Fig. 9).
For each quantitative evaluation, arithmetic means
with standard deviations were calculated, and the statistical significance of the counts with respect to those found in
intact animals was assessed by using Student’s t-test (P #
Part A: Morphological aspects of remodeling
in transplanted and/or sham-operated
kittens and fate of the graft
Lesion immediately after operation. Upon opening
the skull of sham-operated kittens that had been perfused
immediately after preparation of the lesion (kittens 15–17;
Table 1, Fig. 2A), an extensive hematoma was observed in
the meninges covering the operated hemisphere, accompanied by edema of the neural tissue within 5–10 mm of the
wound margin. During subsequent fixation and dehydration of the tissue, the edema deflated and was invisible in
mounted sections. Although the position of the lesions
varied slightly in anteroposterior and/or mediolateral extent (about 1–3 mm), the size of the lesion was rather
uniform in all operated animals, and the cavity was
localized, as intended, in the gray matter of the lateral
gyrus of the occipital cortex, in visual area 18.
Operated region 2 months after surgery
General features. All kittens destined for long-term
survival recovered quickly and without complication from
surgery. After several hours, their behavior was comparable to that of their nonoperated litter mates. No changes
of visual perception and/or orientation behavior were
evident in either the sham-operated or the transplanted
animals. Two months after the operation, the bone piece
that had been removed temporarily during surgery and
the surrounding skull were well fused together. The lesioned and/or lesioned and transplanted brain regions
were covered with an unusually thick layer of meninges
(Fig. 3) that firmly attached the brain’s surface to the inner
surface of the neurocranium.
Sham-operated kittens. Lesions without a transplant
always remained present as a cavity filled with cerebrospinal fluid and roofed by heavily vascularized meninges
(kittens 11–14; Table 1, Fig. 2B). In most sham-operated
Fig. 2. Photomicrographs of hematoxylin-stained, 20-µm frontal
sections at level III (see Fig. 1) through operated area 18 of three
kitten brains. A: The cavity immediately after operation, before the
transfer of the graft (animal 15 in Table 1); B,C: The cavity 2 months
after operation in a sham-operated (B; kitten 11) and a transplanted
(C; kitten 3) animal. Note, that in the grafted cortex in C, the lateral
gyrus and the suprasylvian gyrus are fused and that the posterolateral sulcus (arrowhead in A and B), which normally separates them, is
missing. Asterisks indicate the Sylvian fissure. w, White matter; g,
gray matter. Scale bar 5 3 mm.
animals, the wound’s position and dimensions (adjusted
for growth) were approximately the same as originally
defined. However, in two cases (kittens 13 and 14; Table 1),
the lesion was considerably enlarged. The gray matter
underneath the wound had apparently degenerated, so
that the bottom of the cavity was in close contact with the
white matter (Fig. 2B). The walls were lined with a thin
layer of nonneuronal tissue that merged gradually with
the meninges. A prominent glial scar never developed,
although some accumulation of glial cells was observed
close to the bottom and the most superficial edges of the
cavity. At the time of perfusion, necrosis still appeared to
be in progress in the vicinity of the lesion (within 100–200
µm), as evidenced by a reduced cell density and cell
Transplanted kittens. Remodeling of cavities in transplanted brains differed from animal to animal, without
any linkage to sex. The most obvious variation was in the
degree of closure of the cavities. In four of ten grafted
kittens (kittens 3, 7, 9, and 10; Table 1), the cavity was
completely filled with neural tissue (Figs. 2C, 3, 5C). In
two of these (kittens 9 and 10), exuberant tissue connected
lateral and suprasylvian gyri in such a way that the sulcus
between the gyri was missing (Fig. 2C; Fig. 4, left column,
level IV; Fig. 5A,C). In four other cases (kittens 1, 2, 5, and
6), the cavities were closed only partially, persisting as a
small hollow covered with meninges at level III and
extending variably in the anteroposterior direction but
never invading the adjacent gyrus (Fig. 4, middle column).
Only two transplanted cavities remained open (kittens 4
and 8; Fig. 4, right column). In the case of kitten 4, the
wound extended at least to level II anteriorly and to level
IV posteriorly; at level I, the lateral gyrus was diminished
compared with the intact contralateral hemisphere, and
the sulcus between the lateral and suprasylvian gyri at
level V was deeper (compare right column of Fig. 4 with the
original position of the cavity around level III in Fig. 1).
The remodeled area was covered by meninges which, in
this region, appeared thicker than normal (Fig. 3).
These qualitative observations were accompanied by the
calculations of the O/I indices (see Materials and Methods)
in coronal sections of both hemispheres cut at levels I–V
(Fig. 4). In the four sham-operated animals, the mean O/I
index (0.83) was significantly less than 1.0, reflecting the
much smaller area of the lesioned side compared with the
contralateral intact side. By contrast, in the four transplanted kittens with completely filled cavities, the hemispheres were of comparable size, as reflected by an O/I
index close to 1 (1.074), which was also close to the index
observed in intact animals (1.026). Interestingly, the O/I
index in animals with partially closed lesions (kittens 1, 2,
5, and 6) was also in this range (1.020), a result that could
be attributed to expansion of the tissue surrounding the
cavity. In animals in which the grafted lesion failed to close
(kittens 1 and 8), the lesioned hemisphere was smaller
Figure 3
(O/I index, 0.94), which was also the case in the shamoperated kittens (O/I index, 0.83).
In the transplanted animals with fully filled cavities, the
original borders of the lesion were no longer visible, and
the organization of the cerebral cortex, including lamination and presence of both neuronal and nonneuronal cell
populations, resembled that of intact kittens (Fig. 3). Only
two unusual structures were evident. First, a narrow sheet
of nerve fibers lined the remodeled area medially (Figs. 2C,
3; Fig. 4, left and middle columns, levels II–IV). Here,
intermingled with glial and endothelial cells, nerve fibers
projected in an aberrant way from the white matter up to
the pial surface. At the upper cortical layers (I and II), this
fiber bundle fanned out slightly and was penetrated by a
dense meshwork of vessels originating in the meninges.
Second, in the two animals (kittens 9 and 10) with neural
tissue protruding above the gyral surface, the tissue was
penetrated by a thin fiber tract that also originated in the
underlying white matter. In coronal sections, the protruding tissue was reminiscent of a small extra gyrus (Fig.
Mitotic figures (Fig. 3) were observed in all animals at
the time of death, indicating a weak level of proliferative
activity in the neocortical tissue of 3-month-old kitten
brains. All mitoses were BrdU2. At this age, the number of
mitoses was similar in intact (72.2 6 69.9 mitotic cells per
mm2) and transplanted (85.1 6 82.5 mitotic cells per mm2)
animals. In most cases, the dividing nuclei appeared to be
those of endothelial and glial cells. These observations not
only indicate ongoing vasculogenesis in the juvenile cat
brain but also suggest that the graft does not influence the
rate of cell proliferation at this particular stage (3-monthold kittens) of tissue remodeling.
Fate of graft tissue. Immunohistochemical analysis of
the tissue replacing the cavity revealed that the vast
majority of the constituent cells of this tissue did not
express the mouse-specific Thy-1.2 antigen diagnostic of
the xenograft (Fig. 5C) but were positive for the feline
Thy-1 antigen (Fig. 5D). Although a solid piece of neural
tissue had been transplanted initially, no compact graft
was observed 2 months later, and just solitary Thy-1.21
cells, which were visible only at high magnification, were
found dispersed in the recipient’s brain (Figs. 6, 7). These
Thy-1.21 cells were observed not only in the transplanted
area but also in subcortical regions and, occasionally, even
in the contralateral hemisphere. The majority were situated in or close to the operated cortical region, mostly in
upper cortical layers I and II (at cutting level III), where
they intermingled with fibers of the aberrant tract deFig. 3. Photomicrographs of 20-µm-thick, hematoxylin-stained sections that show a comparison of the transplanted visual cortical area
18 (Tr; animal 7 in Table 1) 2 months after operation with the same
region in an age-matched intact (In) kitten cortex (cortical layers I–VI;
animal 18). The schematic drawing in the lower left corner shows the
position at cutting level III (compare with Fig. 1) of the photomicrograph shown above. d, Dorsal; l, lateral. The lower middle photomicrograph is a reduced version of the picture from the grafted area and
serves for the identification of the relevant structures. The black line
indicates the approximate size and shape of the original cavity
(compare with Fig. 2A). A narrow sheet of nerve fibers (asterisk)
extends from the white matter (wm) to the meninges (m). It fans out as
it reaches the upper cortical layers (I and II) before it merges with the
meninges. The photomicrograph in the lower right corner shows
examples of a neuron (N), a glial cell (G), an endothelial cell (E), and a
mitotic cell (M) in the transplanted area. Scale bars 5 1 mm in Tr and
ln photomicrographs, 10 µm in lower right corner.
scribed above. Graft cells also populated other areas, like
the superior colliculus, the lateral geniculate body, and the
white matter (optic radiation, optic tract, and corpus
callosum); occasionally, they were observed in other regions (e.g., in the mesencephalon) that do not belong to the
central visual system.
The frequency of Thy-1.21 cells varied from animal to
animal (Table 1), and there was no obvious correlation
between the degree of tissue remodeling in the lesion area
and the number of graft cells present. Thus, for example, a
large difference was observed between the numbers of
Thy-1.21 cells in animals 9 and 10, even though, in both
cases, the cavities were completely filled. On the other
hand, in the grafted cavities that persisted and that
resembled the cavities observed in sham-operated animals, donor cells could still be detected (kittens 4 and 8;
Table 1).
Quantitative measurement of the number of donor cells
proved their average number to be very low, i.e., only 5–40
cells per brain section. Sections from the center of the
transplanted area at level III contained the highest number of Thy-1.21 cells, whereas their numbers decreased
with increasing distance from the grafted region (Fig. 6).
The Thy-1.21 cells often showed certain abnormalities
characteristic of degenerating cells. Such cells either possessed swollen somata, so that they appeared larger than
surrounding host cells, or, at the other extreme, they had
dark, shrunken nuclei (not shown). Cells of the latter type
were often observed in the vicinity of the operated area
and close to the meninges, whereas swollen donor cells
were found outside of this region.
The specificity of the Thy-1.2 immunostaining was verified in several ways. First, in nontransplanted (shamoperated and intact) control animals, no Thy-1.21 cells
were detected (Table 1). Second, sections from transplanted animals that were processed without the primary
Thy-1.2-specific Ab were negative. Third, brain sections
taken from 2-month-old C57BL/6J mice (positive for the
Thy-1.2 antigen) were stained heavily, as expected. Finally, immunostaining for the cat-specific Thy-1 antigen
indeed showed that the vast majority of the remodeled
tissue belonged to the kitten host, whereas control sections
from mouse tissue remained negative.
Part B: Cell types contributing
to the remodeling of the lesion
Cell density. Photomicrographs that show examples
of staining for different cell types (neurons, astrocytes,
oligodendrocytes, endothelial cells, and donor cells) are
provided in Figure 7, whereas quantitative data concerning the contribution of these cell types to the remodeling of
the lesions are plotted in Figure 8.
Intact kitten brains. In the individual graphs of Figure
8, the relative cell densities calculated for the different cell
types in intact visual cortical area 18 of 3-month-old
kittens are represented by white bars. CNP1 oligodendrocytes (670 6 80 per mm2) and NeuN1 nerve cells (640 6 70
per mm2) were the most numerous. The number of endothelial cells was 470 6 110 per mm2, whereas that of the
GFAP1 astrocytes was 170 6 65 per mm2. Together, the
overall cell density was 1,950 6 100 cells per mm2. The
glial-neuronal index (the ratio of glial [GFAP1 and CNP1]
to neuronal [NeuN1] cells) in this area was 0.76.
Sham-operated brains. The density of cells around the
persisting cavity varied with the distance from the cavity
Fig. 4. Top: Camera lucida drawings of the dorsal portions of
coronal sections from three transplanted kitten brains taken at
anteroposterior levels I–V (see Fig. 1). Each column represents a
kitten showing one of the three defined degrees of tissue remodeling
observed within the operated region: Fully filled (1), partially
filled (1/-) and persisting lesion (-). Arrowheads indicate the most
evident differences in relation to the contralateral hemisphere.
Bottom: Weight ratios (operated/intact [O/I] indices) of equivalent
regions of the right (operated) and left (intact) hemispheres determined by weighing of camera lucida drawings (similar to those shown
above). Bars represent means with their standard deviations in
sham-operated (Sham) and transplanted (Tr) kitten brains with fully,
partially, and unfilled cavities. Statistical significance at P # 0.05
(asterisk) was calculated with respect to the mean value of the O/I
index of intact animals (thick horizontal line).
walls at which the counts were made (see schematic
drawing in Fig. 8). Whereas the density distal to the cavity
(200–1,000 µm from the cavity walls) was not significantly
different from the cell density in the intact animals (Fig. 8,
horizontally hatched bars), at locations proximal to the
cavity walls (within 100–200 µm), the density was generally lower compared with that of the remaining cortex. The
only exception to this general picture was the density of
endothelial cells, which was higher by almost 100% (830 6
170 cells per mm2 vs. 480 6 110 cells per mm2) proximal to
the cavity walls. This increase in the number of endothelial cells near the cavity is in agreement with the presence
of a dense meshwork of blood vessels originating in the
meninges that formed around the cavity.
Proximal to the cavity (Fig 8., vertically hatched bars),
neurons were the sparsest cell type, with a density of only
25 6 10 cells per mm2. The densities of astrocytes and
oligodendrocytes were also reduced to 48.5 6 49 cells per
mm2 and 450 6 270 cells per mm2, respectively. The
glial-neuronal index in this area was very high (19.9).
However, because the distribution of the oligodendrocytes
was irregular, with the cells often forming clusters of
varying sizes, their density fluctuated along the border of
the lesion from values as low as 180 cells per mm2 to values
approaching those in intact controls. The overall consequence of these changes in the densities of neurons,
astrocytes, and oligodendrocytes proximal to the persisting cavity was that their combined density dropped to
around 524 cells per mm2, a value that was 35% of that
found in the intact visual cortex of age-matched kittens.
Transplanted brains. In the transplanted animals with
fully closed cavities, the relative numbers of NeuN1,
GFAP1, and CNP1 cells as well as of endothelial cells were
comparable to those in the intact visual cortex (Fig. 8,
cross-hatched bars). The measured values were (in cells
per mm2) 650 6 120 for neurons, 160 6 50 for astrocytes,
700 6 90 for oligodendrocytes, and 530 6 50 for endothelial cells, which gave a total of 2,040 6 120 cells per mm2.
The glial-neuronal index in the remodeled cortices of
transplanted kittens with fully filled cavities was 0.76, a
value identical to that calculated in area 18 of intact
Two months after transplantation, as described above,
the number of Thy-1.21 donor cells in the kitten brain was
very low (see Fig. 6, counts per section). Quantitative
counts made only from the center of the remodeled cortex
at level III (see Fig. 8, schematic drawing) revealed that
the number of Thy-1.21 cells fluctuated around a mean of
10 6 15 cells per mm2 (Fig. 8). The majority of these cells
were dispersed near the cortical surface in the vicinity of
Fig. 5. Detection of Thy-1 antigens 2 months after operation in
grafted visual area 18 of kitten 9 (see Table 1), in which cortical
remodeling led to complete filling of the lesion with neural tissue.
A: Dorsal view of the brain showing the area of the initial lesion with a
‘‘plug’’ of nervous tissue now linking the lateral (top) and suprasylvian
(bottom) gyri. Vertical bars indicate the level at which the sections in
B–D were taken. Medial is up, lateral is down. B–D: Details of three
adjacent frontal 20-µm sections from the level indicated in A.
B: Hematoxylin-stained section. C: Section immunostained for the
Thy-1.21 mouse graft tissue indicating that no compact mass of graft
tissue forming a distinct boundary with the host brain was found. The
relatively few donor cells dispersed in the kitten brain were visible
only under high-power magnification, as shown in Figure 6. D: Section
immunostained for the cat Thy-1 antigen showing that the vast
majority of the tissue filling the cavity is of host origin. For B–D,
medial is to the left, and lateral is to the right. Scale bar 5 1.5 mm.
Fig. 6. Illustration of the main regions of dispersion of Thy-1.21
mouse cells throughout the cat brain 2 months after grafting. Coronal
sections were taken at levels II–IV. Left: Camera lucida drawings
showing brain regions where the majority of the immunostained graft
cells were localized (solid triangles). Right: Photomicrographs of
Thy-1.21 mouse cells observed in transplanted area 18 of the visual
cortex (VC; counterstained with hematoxylin) and in subcortical
regions, such as the superior colliculus (CS; without counterstaining)
or the lateral geniculate body (CGL; without counterstaining). Bottom:
Counts of the number of Thy-1.21 cells observed in transplanted kitten
brains. Bars depict the means of the absolute numbers of cells (with
standard deviations) counted in one section from each of the evaluated
levels I–V. Scale bar 5 50 µm.
Fig. 7. Photomicrographs showing examples of each of the evaluated cell types as revealed either by immunostaining (Thy-1.2, donor
cells; neuron-specific nuclear protein A60 [NeuN], neurons; glial
fibrillary acidic protein [GFAP], astrocytes; cyclic nucleotide phosphohydrolase [CNP], oligodendrocytes) or from hematoxylin staining
(Hem; endothelial cells). Arrows indicate bromodeoxyuridine-positive
(BrdU1) cells of a specific type that incorporated the DNA marker at
the time of their last cell cycle close to the time of the BrdU injection.
Scale bar 5 100 µm.
Fig. 8. Densities of neurons (NeuN1), astrocytes (GFAP1), oligodendrocytes (CNP1), donor cells (Thy-1.21), and endothelial cells (E) per
mm2 in visual area 18 of 3-month-old kitten brains. Schematic
drawing at top left illustrates the position and size of the region that
was analyzed quantitatively in coronal sections at level III (d, dorsal;
l, lateral; see also Fig. 1). Two regions were evaluated in shamoperated brains: An area proximal to the cavity (cross-hatched area in
the schematic drawing located 100–200 µm of the cavity walls) and a
distal area (hatched area located 200–1,000 µm of the cavity walls). In,
intact animals; Sh-p, sham-operated animals (proximal to cavity);
Sh-d, sham-operated animals (distal to cavity); Tr, transplanted
kittens. Statistical significance (asterisk) at P # 0.05 was determined
with respect to intact controls.
the broadened, unusual fiber tract described above. In
other regions of the remodeled cortex, the distribution of
the Thy-1.21 donor cells was more random.
Postoperative proliferation of cells as revealed by
incorporation of BrdU
BrdU/Thy-1.21 cells. The density of Thy-1.21 donor
cells that were also BrdU1 (2.7 6 1.9 cells per mm2)
represented about 25% of their population (Figs. 7, 9).
Most of the double-labeled cells were found in the transplanted area, whereas those Thy-1.21 cells that migrated
farther into the host brain parenchyme were usually
BrdU/NeuN1 cells. In intact animals, the number of
BrdU/NeuN double-labeled cells was very low (about 1.3 6
0.5 cell per mm2; Figs. 7, 9). In sham controls, the density
of such cells varied from 0.8 6 0.8 cells per mm2 in areas
proximal to the persisting lesion to 1.9 6 2.1 per mm2 in
more distal regions, but neither value was significantly
different from that observed in intact kittens. In striking
contrast to these numbers, the density of BrdU/NeuN
double-labeled cells in transplanted animals was higher by
almost a factor of ten, at 9.5 6 4 cells per mm2.
BrdU/GFAP1 cells. In intact area 18, the density of
BrdU/GFAP1 cells was only 4.3 6 1.2 cells per mm2 (Figs.
7, 9). In sham-operated animals, proximal to the border of
the cavity, the density was very similar (4.2 6 3 cells per
mm2); but, at more distal locations (at 200–1,000 µm from
the cavity), the density of BrdU1 astrocytes was nearly five
times higher, at 19 6 8 cells per mm2. In transplanted
animals, their density (33.9 6 6 cells per mm2) was more
than eightfold higher than that observed in intact controls.
BrdU/CNP1 cells. The density of BrdU1 oligodendrocytes proximal to the cavity in sham-operated kittens
(75 6 55 cells per mm2) was less than half of the value
observed in intact controls (165 6 40); but, at more distal
locations, the density was similar (Figs. 7, 9). In transplanted kittens, the density of BrdU/CNP1 cells (195 6 20
cells per mm2) was similar to that observed in intact
BrdU-labeled endothelial cells. The density of endothelial cells that stained for BrdU in transplanted animals
(155 6 45 cells per mm2) was similar to the value observed
in the intact kitten neocortex (155 6 30 cells per mm2;
Figs. 7, 9). In sham-operated animals, the density dropped
to 40 6 30 cells per mm2 in regions proximal to the lesion
and to 85 6 55 mm2 in more distal locations.
The key observation of this study was a remarkable
host-mediated recovery of mechanically lesioned kitten
visual cortex under the influence of a fetal murine xenograft. In 40% of cases, the original cavity was replaced
completely with neural tissue, which, moreover, was not of
mouse origin but, instead, was composed almost exclusively of host cells. The cytoarchitecture of the remodeled
region resembled that of the intact cerebral cortex. No
repair was ever observed in sham-operated animals, in
which the lesioning invariably led to permanent damage
and to the formation of a necrotic zone that extended
100–200 µm from the cavity borders and within which the
densities of neuronal and glial cells were reduced. On the
other hand, this zone was characterized by an unusually
dense meshwork of blood vessels. The qualitative differences between sham-operated and transplanted cortices
were reflected by substantial quantitative differences in
cell densities of different cell populations. In sham controls, the densities of most cell types decreased in the
degenerating area, with neurons suffering the greatest
reduction. Consequently, the glial-neuronal index (the
ratio of glial [GFAP1 and CNP1] to neuronal cells) was
very high in these animals (19.9). In striking contrast, in
the remodeled cortices of transplanted kittens the index
was identical to the value in intact control animals (0.76).
These various observations suggest that neural transplantation may not rely solely on integration of the graft as a
‘‘patch’’ into the host brain, where it replaces more or less
completely the missing tissue in structural and functional
terms. Rather, it appears that the graft exerts at least two
different effects. First, it seems to reduce both necrotic
processes and scar formation, which normally would prevent or seriously impede any repair. In addition, our
results suggest that the graft may also serve as a trigger
and an initiator of regenerative processes in the recipient
tissue that could not occur without its initial presence. The
fate of the graft and the nature of the mechanisms of repair
it may initiate are considered separately below.
Fate of the grafted mouse cells
in the kitten CNS
Two months after transplantation, only very few donor
cells had survived in the host brain. Apparently, the
murine graft had been recognized as foreign and was
gradually destroyed by the kitten immune system. Although we were always aware of a possible immune
response of the host toward the xenograft (see, e.g., Finsen
et al., 1991; Pakzaban and Isacson, 1994), we avoided the
use of any immunosuppressive drugs, because we did not
want to disturb in any way the interaction between graft
and host. Such an interaction, although it was much
weaker, was also tolerated in our earlier allografts in mice,
in which graft tissue was also resorbed (Ourednik et al.,
1993a,b; Ourednik and Ourednik, 1994) without any significant detriment to the regenerative process. Finally,
experiments by Blakemore et al. (1995) showed that, when
administration of immunosuppressive drugs was stopped
in animals receiving demyelinating lesions and grafts of
myelinating oligodendrocytes, myelination continued by
host glia, whereas the graft cells were gradually resorbed.
The small number of graft cells did not survive as a
compact mass of tissue that could have integrated into the
host at the site of the cavity. Instead, the mouse cells
became dispersed in the cat brain, reaching regions at
significant distances from the graft’s initial placement.
Thus, although grafted cells were found mostly in the
operated region, additional scattered donor cells were
observed in ipsilateral subcortical areas and even in the
contralateral hemisphere. A similar migration to the contralateral side was observed with motoneurons grafted
into the striatum (Ruiz-Flandes et al., 1993) and with
astrocytes that were placed originally into deep neocortical
layers (Lund et al., 1993). The ability of grafted nerve and
glial cells to migrate and to integrate into the neural
network of the recipient has been described in the past for
nerve and glial cells (see, e.g., Finsen and Zimmer, 1986;
Privat et al., 1986; Goldberg and Bernstein, 1988; McConnell, 1988; Kawamura et al., 1988; Sieradzan and Vrbová,
1989; Demierre et al., 1990; Sotelo et al., 1990; Lund at al.,
1993; Warrington et al., 1993; Gumpel, 1994; Lois and
Alvarez-Buylla, 1994) for both short- and long-distance
Fig. 9. Densities of BrdU-stained cells per mm2 in doubleimmunostained sections: Neurons (BrdU/NeuN1), astrocytes (BrdU/
GFAP1), oligodendrocytes (BrdU/CNP1), and donor cells (BrdU/Thy1.21) and endothelial cells in hematoxylin-stained sections (BrdU/E).
BrdU was administered to the operated animals every other day
during the 12 days immediately following the operation, and the intact
animals were injected at corresponding ages. Statistical significance
(asterisk) at P # 0.05. For additional details, see the legend to
Figure 8.
migration. The latter seems to occur mainly after interspecies transplantations, whereas allogenic grafts tend to
remain in the vicinity of the region of implantation (Wells
et al., 1987). Björklund and coworkers (1982) suggested
that, in the case of a xenograft with its stronger immunogenic
effect, the capacity to migrate could be crucial for the donor
cells in order to escape recognition and attack by immunocompetent cells at the side of a ruptured blood-brain barrier.
About 25% of the Thy-1.21 (donor) cells were labeled by
BrdU. These cells apparently stopped division soon after
transplantation, around the time of the BrdU injections,
and they were observed mainly at the site of deposit. By
contrast, in more distant brain regions, most graft cells
were BrdU2, an observation that could mean either that
the cells ceased division before the BrdU injections or that
the label became diluted in subsequent divisions by newly
synthesized, nonlabeled DNA. Because BrdU is available
for only a relatively short time (20–40 minutes; Boswald et
al., 1990), dilution of the label is the most likely interpretation, because dividing cell precursors have better migratory capabilities than more differentiated cells and, hence,
a greater capacity to escape the host’s immune response
(mainly in the area where the hematoencephalic barrier
was damaged during the operation) and to infiltrate more
remote areas of the kitten brain.
The migration of the transplanted cells raises many
questions. For example, it is not clear how the transplanted cells migrate in the host CNS, what guiding clues
they utilize, and the degree to which they migrate to
specific targets. We did observe that the grafted cells
appeared not only in the operated region and in the
various neural structures that form the central visual
pathways but also at other sites throughout the brain. It is
conceivable that structures like fiber tracts and blood
vessels may facilitate cellular movement by serving as
scaffolds for mechanical support and by providing molecular sites for attachment and orientation (see, e.g., Zhou et
al., 1990; Jacque et al., 1991; for review, see Hynes and
Lander, 1992).
Cytoarchitecture and cell proliferation
in intact vs. sham-operated and
transplanted kitten brains
Intact brains. Although there have been a number of
previous measurements of the density of neurons in the cat
visual cortex (see, e.g., Cragg, 1975; Beaulieu and Colonnier, 1983, 1985, 1987; O’Kusky, 1985; Takacs et al., 1992),
it is difficult to compare these earlier values with ours
because of substantial differences in the ages of the
animals, differences in the preparation of the histological
specimens that could change the correction factor for
shrinkage of the tissue, and in our study the use of cell
type-specific markers that were not available in the earlier
studies. Indeed, the use of such markers allowed us to
provide data on the relative densities of cell types other
than neurons, which, to our knowledge, have not been
reported previously in the kitten visual cortex. Moreover,
because our aim was to compare the densities of various
cell types in the different experimental conditions and not
to provide definitive data on normal cell densities in the
manner of earlier studies, we neither prepared the tissue
nor conducted the measurements in a manner that makes
meaningful comparisons with earlier studies.
The intact kittens received BrdU injections at the same
age (between postnatal weeks 5 and 7) as the shamoperated and transplanted kittens. The presence of BrdUlabeled cells implies that, during that particular period of
time, many cells still were completing their last cell cycle.
Even some mitotic figures were present at the time of
perfusion (at the end of 3 months). Double labeling with
BrdU and cell type-specific markers or hematoxylin staining revealed that the dividing cells were mostly oligodendrocytes and endothelial cells, whereas, for the most part,
the formation of neurons had terminated (with some rare
exceptions). These observations were indicative of the fact
that, at the particular time of observation, cell proliferation and tissue differentiation linked to vasculogenesis
and myelination were still in progress.
To our knowledge, quantitative documentations of the
time table of nonneuronal cell formation in the cat brain
are not available. The genesis of nerve cells in the cat
primary visual cortex was described in detail by Luskin
and Shatz (1985), who employed [3H]thymidine autoradiography. In agreement with findings in other mammalian
brains, in which the period of nerve cell formation in the
cerebral cortex has been shown to finish before birth (for
reviews, see Korr, 1980; Jacobson, 1991), they demonstrated that proliferation of nerve cell precursors in the
kitten visual cortex ceased around the time of birth. On the
other hand, there are reports of rare production of single
nerve cells during adulthood (see, e.g., Altman, 1962;
Kaplan, 1981; Altman and Bayer, 1993). Indeed, we also
observed a few postnatally formed cells, which were double
labeled with NeuN- and BrdU-specific Abs (Figs. 7, 9). So
far, the source and function of postnatally formed neurons
are not clear. It was shown recently that the source of these
cells may be either the persisting subventricular zone in
the forebrain (Lois and Alvarez-Buylla, 1993; Luskin,
1993; Morshead et al., 1994; for reviews, see Gage et al.,
1995; Svendsen and Rosser, 1995; Weiss et al., 1996;
Stemple and Mahanthappa, 1997) or progenitor cells in
the brain parenchyme (Kaplan, 1981, 1985; Okano et al.,
1993). It is conceivable that this weak postnatal neurogenesis reflects a slow turn-over of the neuronal population
evoked by microinjuries or plastic events accompanying
learning and memory.
Sham-operated brains. An injury of the CNS is usually followed by a set of reactions that appear to attenuate
its structural and functional consequences and that begin
with a recruitment of immunocompetent cells (Kerr and
Bartlett, 1989; Sloan et al., 1991) and the production of
glial (Billingsley et al., 1982) and endothelial (Broadwell et
al., 1990) cells, which combine to form a glial scar (Berry et
al., 1983). In agreement with this general finding, in the
lesioned area, we observed a necrotic zone penetrated by a
dense meshwork of blood vessels, which probably guaranteed a continuous supply of macrophages and other immunocompetent cells.
From our data, it became evident that the vulnerability
to injury is not the same for all brain cell populations. The
number of cells proximal to the lesion decreased, with
neurons and astrocytes showing the greatest reduction. In
addition, the extent and nature of the observed proliferative activity depended on the distance from the walls of the
cavity. Proximally, the numbers of BrdU-labeled oligodendrocytes and endothelial cells decreased compared with
intact brains (Fig. 9). The substantial reduction of BrdU1
endothelial cells and the simultaneous increase of vascularization and, hence, the density of endothelial cells in this
region (compare the graphs for endothelial cells in Figs. 8, 9)
could well be attributed to a very active endothelial proliferation and a resulting dilution of the original amount of the
BrdU label by newly synthesized, nonlabeled DNA. At more
distal locations from the lesion, where cell density was similar
to that of the intact neocortex, the numbers of double-labeled
neurons and oligodendrocytes were not different from those
observed in intact controls, whereas the density of BrdU1
astrocytes was higher.
In harmony with recent reports (Huang and Lim, 1990;
Skup et al., 1993) describing the rare formation of single
neurons in the vicinity of a damaged brain area, we also
found individual BrdU1 neurons within 200–1,000 µm of
the lesion. However, their number was the same as that
found in intact controls, from which we conclude that the
degenerative process was overriding any possible attempts
to repair the damage.
Transplanted animals. The observed variability in
the degree of graft-induced closure of the original cavity
and in the number of surviving graft cells cannot be
explained readily. The regenerative process might be
influenced by methodological factors (e.g., variability of
the size of the graft at the moment of transplantation,
possible damage of the white matter reached during the
preparation of the cavity) as well as by unequal dispositions of the individual animals (e.g., degree of the immune
response). The vast majority of the tissue that replaced the
cavity in the successfully transplanted cortices, as we have
shown with Thy-2 immunohistochemistry, was of host
origin. The remodeling of the transplanted area was
accompanied by quite good preservation of the neocortical
cytoarchitecture. The cavity was filled with both neuronal
and nonneuronal cells, and morphometric analysis of the
representation of individual cell types and their packing
densities showed no significant quantitative changes with
respect to the tissue in equivalent brain regions of intact
controls. No hyperplastic increase in density, like that of
endothelial cells in the sham-operated animals, was observed for any specific cell population. The similarity in
cell densities of intact and transplanted animals suggested
a coordinated increase in all cell classes rather than
multiplication and/or immigration of selected cell populations, which occurs in the case of glial cells during postinjury gliosis (for review, see Norenberg, 1994) or in endothelial cells during revascularization (Lawrence et al., 1984).
In these animals, the formation of new tissue in the
operated region was accompanied by intensification of
postoperative cell proliferation. The present data are in
agreement with other findings (Lundberg and Møllgård,
1979; Polezhaev et al., 1988; Ourednik et al., 1993a,b;
Ourednik and Ourednik, 1994), which indicate that the
host brain can be stimulated to cell proliferation by an
embryonic neural graft. In the present case, the number of
BrdU1 cells (which divided shortly after the operation, i.e.,
at the time of the BrdU injections) was increased in
transplanted animals compared with that in the intact
controls. On the other hand, the number of mitotic figures
(reflecting actual proliferative activity that was occurring
at the time of perfusion, i.e., at 3 months of age) was
similar in both groups. Consequently, these data suggest
that the graft-dependent activation of cell division was
limited in duration and occurred shortly after transplantation.
The number of postoperatively labeled dividing cells was
variable, depending on the cell type. The density of BrdU/
NeuN1 cells increased nearly ten times, and the density of
BrdU/GFAP1 cells increased about eight times. The number of surviving Thy-1.21 (donor) cells in transplanted
kitten brains, as mentioned above, was very low, and only
25% of these cells were BrdU1. The overall number of
BrdU1 neurons (9.5 6 4 cells per mm2) was significantly
higher than the number of BrdU1 donor cells (2.7 6 1.9).
After subtraction of the latter from the former, the number
of postoperatively dividing neurons derived from the kit-
ten host can be estimated to have increased from about
three in the intact animals to six or seven nerve cells per
mm2 in the transplanted animals (see also Fig. 9).
Although the cytoarchitecture of transplanted and intact kitten brains and their cell densities were similar, the
number of double-labeled (BrdU1) cells per mm2 varied
among the different cell types. However, this variability
may not necessarily reflect only differences in the extent of
cell proliferation among the cell groups but could as well be
explained, in part or in whole, by differences in the
capacities of surrounding nondividing (BrdU2) cells to
immigrate into the transplanted region. Moreover, it should
be kept in mind that very intensive cell proliferation can
lead to the dilution of the label below the threshold of the
immunodetection. To distinguish between these possibilities, it will be necessary to document the total length of cell
formation and its exact time table. To accomplish the
former, animals would have to be injected repeatedly with
BrdU for a longer time; whereas, to address the latter
different groups of animals could be administered a single
BrdU injection at various postoperative intervals, and the
already differentiated tissue could be analyzed at a common, defined time point.
Possible mechanisms underlying the closure
of the transplanted cavity
It has been shown recently that a neural graft may exert
both protective and trophic influences on degenerative
processes in the host CNS (Bregman and Reier, 1986;
Kesslak et al., 1986a,b; Sofroniew et al., 1986; Cunningham et al., 1987; Sørensen et al., 1989, 1990; Dunnett,
1990; Gage and Fischer, 1991) and may reduce formation
of a glial scar, which normally would hinder or prevent
repair (Krüger et al., 1986; Houlé and Reier, 1988). Along
the same lines, Valoušková and Gálik (1995) observed a
positive influence of a graft on the closure of a nontransplanted cortical cavity prepared in the cortex contralateral
to the transplanted cavity. In an injured brain region,
reactive astrocytes and microglia are known to produce a
plethora of factors, some of which may induce pathological
changes, but others may also support recovery (Giulian,
1993; Giulian et al., 1993). It has been documented (Altar
et al., 1992) that administration of exogenous factors (in
this case, nerve growth factor) may enhance the neuroprotective effect of the endogenously elevated levels of the
same molecule. Therefore, it is conceivable that, in our
experiments, we counteracted injury-induced necrosis and
stimulated recovery by supplying the site of injury with a
mixture of trophic factors present in the fetal tissue. These
factors, together with the endogenous ones, would shift the
balance toward a host-based recovery and would override
the influence of molecules interfering with adult CNS
The possible presence of graft-derived trophic factors
that might assist the host to recover from brain injury
receives support from other studies of neurotransplantation (Stein et al., 1985; Bregman and Reier, 1986; Kesslak
et al., 1986a,b; Cunningham et al., 1987; Valoušková and
Gálik, 1995) as well as from important recent reports of
multipotent stem cells persisting in the subependymal
layer of the adult mammalian brain (see, e.g., Lois and
Alvarez-Buylla, 1993, 1994; Morshead et al., 1994; Weiss
et al., 1996), which can be stimulated to proliferation and
differentiation by neurotrophins, such as EGF (Reynolds
at al., 1992) or bFGF (Gritti et al., 1996). Along these lines,
Craig et al. (1996) showed that infusion of EGF into adult
mouse forebrain for 6 days resulted in a dramatic increase
in the proliferation and total numbers of subependymal
cells and induced their migration away from the walls of
the lateral ventricle into the adjacent cortical, striatal, and
septal parenchyme. Seven weeks after EGF withdrawal,
3% of the newly formed cells were of neuronal origin. This
last report is an example of how, in vivo, a single signaling
molecule can exert a dramatic effect on progenitor and
stem cells in the mammalian CNS.
Several mechanisms could account for successful remodeling of the transplanted cavity. At one extreme, the graft
could induce a reactivation of host cell proliferation, and
the lesion could be filled by resulting new cell populations.
At the other extreme, discrete cell populations within the
host tissue could be stimulated to migrate toward the area
of repair, or a morphogenic movement of whole blocks of
host tissue could lead to a convergence of the cavityforming walls toward each other in the parenchyme (a
‘‘zipping-up’’ effect). The absence of a sulcus between the
suprasylvian and medial gyri in the operated area provides partial evidence for the latter process. These unusual
gyral patterns clearly exceed the reported interindividual
and interhemispheric variability of gyrification described
earlier in the intact cat brain (Webster, 1981). However,
the hyperplastic thickening of the parenchyme in this
region, which, in two cases, led to the appearance of
protruding tissue rising above the surrounding brain
surface and having the appearance of a small extra gyrus,
could just as easily be reconciled with the graft’s activation
of reconstruction of the operated area through immigration and/or proliferation of cells originating elsewhere in
the host brain parenchyme. Both of these processes would
imply a communication of graft and host cells through both
direct contact and the release of soluble substances by the
graft cells during the early postoperative period, before
their rejection by the host’s immune system. Whether the
postoperative activation of cell proliferation was sufficient
by itself to compensate for the structural defects caused by
the injury is uncertain. However, it seems more likely that
additional mechanisms, such as immigration of cells from
the surrounding parenchyme, make major contributions to
the process of regeneration.
The presented data provide evidence that murine fetal
neural tissue has the potential to stimulate regeneration
by injured kitten neocortex. This unexplored role of a
neural graft in CNS repair will require further investigation. Thus, the description of early postoperative stages
and the analysis of the degree of damage and remodeling
in subcortical visual centers as well as experiments assessing possible functional recovery are the next logical steps
in the study. The resulting morphological and physiological data will have to be complemented by molecular and
biochemical analyses aimed at the characterization of the
factor(s) that mediate the graft/host interactions necessary for the observed regeneration.
We thank Drs. Mullen and Bartlett for kindly providing
the anti-NeuN and anti-cat Thy-1 antibodies; Drs. P.E.
Neumann and A. Fine for their constructive comments on
the paper; Dr. C.J. Beaver, P. Ritchie, and V. Baker for
assistance with the operations; and V.R. Fraser and R.
Burgess for their dedicated care of the mouse and cat
Altar, C.A., M. Armanini, M. Dugich-Djordjevic, G.L. Bennett, R. Williams,
S. Feinglass, V. Anicetti, D. Sinicropi, and C. Bakhit (1992) Recovery of
cholinergic phenotype in the injured rat neostriatum: Roles for endogenous and exogenous nerve growth factor. J. Neurochem. 59:2167–
Altman, J. (1962) Are new neurons formed in the brains of adult mammals?
Science 135:1127–1128.
Altman, J. and S. Bayer (1993) Are new neurons formed in the brains of
adult mammals? A progress report, 1962–1992. In A.C. Cuello (ed):
Neuronal Cell Death and Repair. New York: Elsevier, pp. 203–225.
Beaulieu, C. and M. Colonnier (1983) The number of neurons in the
different laminae of the binocular and monocular regions of area 17 in
the cat. J. Comp. Neurol. 217:337–344.
Beaulieu, C. and M. Colonnier (1985) A comparison of the number of area
17, 18 and posteromedial suprasylvian (PMLS) area in the cat. Brain
Res. 339:166–170.
Beaulieu, C. and M. Colonnier (1987) Effect of richness of the environment
on the cat visual cortex. J. Comp. Neurol. 226:478–494.
Berry, M. (1989) Transplantation and regeneration of neural tissue in the
central nervous system. Curr. Opin. Neurol. Neurosurg. 2:946–952.
Berry, M., W.L. Maxwell, A. Logan, A. Mathewson, P. McConnell, D.E.
Ashhurst, and G.H. Thomas (1983) Deposition of scar tissue in the
central nervous system. Trauma 3:31–53.
Billingsley, M.L., J.A. Straw, and H.G. Mandel (1982) Glial DNA synthesis
and cell proliferation in the lesioned frontal cortex. Brain Res. 247:325–
Björklund, A., U. Stenevi, S.B. Dunnett, and F.H. Gage (1982) Cross-species
neural grafting in a at model of Parkinson’s disease. Nature 298:652–
Blakemore, W.F., A.J. Crang, R.J. Franklin, K. Tang, and S. Ryder (1995)
Glial cell transplants that are subsequently rejected can be used to
influence regeneration of glial cell environments in the CNS. Glia
Boswald, M., S. Harasim, and B. Maurer-Schultze (1990) Tracer dose and
availability of thymidine and bromodeoxyuridine: Application of bromodeoxyuridine in cell kinetic studies. Cell Tissue Kinet. 23:169–181.
Bregman, B. and P.J. Reier (1986) Neural tissue transplants rescue
axotomized rubrospinal cells from retrograde death. J. Comp. Neurol.
Broadwell, R.D., H.M. Charlton, P. Ebert, W.F. Hickey, J.C. Villegas, and
A.A. Wolf (1990) Angiogenesis and the blood-brain barrier in solid and
dissociated cell grafts within the CNS. Progr. Brain Res. 82:95–101.
Canadian Council of Animal Care (1993/4) E.D. Olfert (ed): Guide to the
Care and Use of Experimental Animals, Vol. 2. Ottawa, Ontario: Brada
Printing Services, Inc.
Charlton, H.M., A. Barclay, and A.F. Williams (1983) Detection of neuronal
tissue from brain grafts with anti-Thy-1.1 antibody. Nature 305:825–
Cragg, B.G. (1975) The development of synapses in the visual system of the
cat. J. Comp. Neurol. 160:147–166.
Craig, C.G., V. Tropepe, C.M. Morshead, B.A. Reynolds, and S. Weiss (1996)
In vivo growth factor expansion of endogenous subependymal neural
precursor cell populations in the adult mouse brain. J. Neurosci.
Cunningham, T.J., C.B. Sutilla, and F. Haun (1987) Trophic effect of
transplants following damage to the cerebral cortex. Ann. NY Acad. Sci.
Daw, N.W. (1995) Visual Development. New York: Plenum Press.
Demierre, B., J.-C Martinou, and A.C. Kato (1990) Embryonic motoneurons
grafted into the adult CNS can differentiate and migrate. Brain Res.
Dunnett, S.B. (1990) Neural transplantation in animal models of dementia.
Eur. J. Neurosci. 2:657–587.
Dunnett, S. and A. Björklund (1994) Functional Neuronal Transplantation.
New York: Raven Press.
Dyck, R.H., L.J. Van-Eldik, and M.S. Cynader (1993) Immunohistochemical
localization of the S-100 beta protein in postnatal cat visual cortex:
Spatial and temporal patterns of expression in cortical and subcortical
glia. Dev. Brain Res. 72:181–192.
ffrench-Constant, C. and G.A. Mathews (1994) Brain repair lessons from
developmental biology. J. Neurol. 242:S29–S32.
Finsen, B. and J. Zimmer (1986) Timm staining of hippocampal nerve cell
bodies in the Koyoto rat. A cell marker in allo- and xenografting of rat
and mouse brain tissue, revealing neuronal migration. Dev. Brain Res.
Finsen, B., T. Sørensen, B. Castellano, E.B. Pedersen, and J. Zimmer (1991)
Leucocyte infiltration and glial reactions in xenografts of mouse brain
tissue undergoing rejection in the adult rat brain. J. Neuroimmunol.
Fischer, L.J. and F.H. Gage (1993) Grafting in the mammalian central
nervous system. Physiol. Rev. 73:583–616.
Gage, F.H. and L.J. Fisher (1991) Intracerebral grafting: A tool for the
neurobiologist. Neuron 6:1–12.
Gage, F.H., J. Ray, and L.J. Fisher (1995) Isolation, characterization, and
use of stem cells from the CNS. Annu. Rev. Neurosci. 18:159–192.
Giulian, D. (1993) Reactive glia as rivals in regulating neuronal survival.
Glia 7:102–110.
Giulian, D., K. Vaca, and M. Corpuz (1993) Brain glia release factors with
opposing actions upon neuronal survival. J. Neurosci. 13:29–37.
Goldberg, W.J. and J.J. Bernstein (1988) Fetal cortical astrocytes migrate
from cortical homografts throughout the host brain and over the glia
limitans. J. Neurosci. Res. 20:38–45.
Gritti, A., L. Cova, E.A. Parati, R. Galli, and A.L. Vescovi (1995) Basic
fibroblast growth factor supports the proliferation of epidermal growth
factor-generated neuronal precursor cells of the adult mouse CNS.
Neurosci. Lett. 185:151–154.
Gritti, A., E.A. Parati, L. Cova, P. Frolichsthal, R. Galli, E. Wanke, L.
Faravelli, D.J. Morassutti, F. Roisen, D.D. Nickel, and A.L. Vescovi
(1996) Multipotential stem cells from the adult mouse brain proliferate
and self–renew in response to basic fibroblast growth factor. J. Neurosci. 16:1091–1100.
Gumpel, M. (1994) Transplanted transgenically marked oligodendrocytes
survive, migrate and myelinate in the normal mouse brain as they do in
the shiverer mouse brain. Eur. J. Neurosci. 6:814–824.
Houlé, J.D. and P.J. Reier (1988) Transplantation of fetal spinal cord into
the chronically injured adult rat spinal cord. J. Comp. Neurol. 269:535–
Huang, L. and R. Lim (1990) Identification of injury-induced mitotic cells in
adult rat cerebral cortex by neuron-specific markers. Dev. Brain Res.
Hynes, R.O. and A.D. Lander (1992) Contact and adhesive specificities in
the associations, migrations, and targeting of cells and axons. Cell
Jacobson, M. (1991) Developmental Neurobiology, 3rd Ed. New York:
Plenum Press.
Jacque, C., I. Suard, P.V. Collins, and N. Baumann (1991) Migration
patterns of donor astrocytes after reciprocal striatum-cerebellum transplantation into newborn hosts. J. Neurosci. Res. 29:421–428.
Kaplan, M.S. (1981) Neurogenesis in the 3-month-old rat visual cortex. J.
Comp. Neurol. 195:323–338.
Kaplan, M.S. (1985) Formation and turnover of neurons in young and
senescent animals, an electron microscopic and morphometric analysis.
Ann. N.Y. Acad. Sci. 457:173–192.
Kaufman, M.H. (1992) The Atlas of Mouse Development. London: Academic
Press, Inc.
Kawamura, K., T. Nanami, Y. Kikuchi, and A. Kitakami (1988) Grafted
granule and Purkinje cells can migrate into the mature cerebellum of
normal adult rats. Exp. Brain Res. 70:477–484.
Kerr, R.S.C. and P.F. Bartlett (1989) The immune response to intraparenchymal fetal CNS transplants. Transplant. Proc. 21:3166–3168.
Kesslak, J.P., M. Nieto-Sampedro, J. Globus, and C.W. Cotman (1986a)
Transplants of purified astrocytes promote behavioral recovery after
frontal cortex ablation. Exp. Neurol. 92:377–390.
Kesslak, J.P., L. Brown, C. Steichen, and C.W. Cotman (1986b) Adult and
embryonic frontal cortex transplants after frontal cortex ablation
enhance recovery on a reinforced alternation task. Exp. Neurol. 94:615–
Kolb, B. (1987) Recovery from early cortical damage in rats. I. Differential
behavioral and anatomical effects of frontal lesions at different ages of
neural maturation. Behav. Brain Res. 25:205–220.
Kolb, B. and R. Gibb (1991) Environmental enrichment and cortical injury:
Behavioral and anatomical consequences of frontal cortex lesions.
Cerebral Cortex 1:189–198.
Kolb, B., R. Gibb, and D. Van der Kooy (1994) Neonatal frontal cortical
lesions in rats alter cortical structure and connectivity. Brain Res.
Korr, H. (1980) Proliferation of different cell types in the brain. Ado. Anat.
Embryol. Cell Biol. 61:1–72.
Krüger, S., J. Sievers, C. Hansen, M. Sadler, and M. Berry (1986) Three
morphologically distinct types of interface develop between adult host
and fetal brain transplants: Implications for scar formation in the adult
central nervous system. J. Comp. Neurol. 249:103–116.
Lawrence, J.M., S.K. Huang, and G. Raisman (1984) Vascular and astrocytic reactions during establishment of hippocampal transplants in
adult host brain. Neuroscience 12:745–760.
Logan, A., J.J. Oliver, and M. Berry (1994) Growth factors in CNS repair
and regeneration. Progr. Growth Factor Res. 5:379–405.
Lois, C. and A. Alvarez-Buylla (1993) Proliferating subventricular zone
cells in the adult mammalian forebrain can differentiate into neurons
and glia. Proc. Natl. Acad. Sci. USA 90:2074–2077.
Lois, C. and A. Alvarez-Buylla (1994) Long-distance neuronal migration in
the adult mammalian brain. Science 264:1145–1148.
Lund, R.D., H.-F. Zhou, and K.T. Yee (1993) The migration of astrocytes
after implantation to immature brains. Int. J. Neurosci. 11:595–601.
Lundberg, J.J. and K. Møllgård (1979) Mitotic activity in adult rat brain
induced by implantation of pieces of fetal rat brain and liver. Neurosci.
Lett. 13:265–270.
Luskin, M.B. (1993) Restricted proliferation and migration of postnatally
generated neurons derived from the forebrain subventricular zone.
Neuron 11:173–189.
Luskin, M.B. and C.J. Shatz (1985) Neurogenesis of the cat’s primary visual
cortex. J. Comp. Neurol. 242:611–631.
McConnell, S.K. (1988) Fates of visual cortical neurons in the ferret after
isochronic and heterochronic transplantation. J. Neurosci. 8:945–974.
McKeon, R.J., B.P. Vietje, and J. Wells (1989) Integrations between donor
and host tissue following cross-species septohippocampal transplants.
Exp. Neurol. 103:213–221.
Mitchell, D.E. and B. Timney (1984) Postnatal development of function in
the mammalian visual system. In I. Darian-Smith (ed): Handbook of
Physiology, Section I: The Nervous System, Vol. 3, Part 1. Sensory
Processes. Bethesda: American Physiological Society, pp. 507–555.
Morris, R. (1991) Immunoperoxidase staining of gene products in cultured
cells using monoclonal antibodies. Methods Mol. Biol. 7:339–359.
Morris, R. and P. Grosveld (1989) Expression of Thy-1 in the nervous
system of the rat and mouse. In A.E. Reif and M. Schlesinger (eds): Cell
Surface Antigens. New York: Marcel Dekker Inc., pp. 121–147.
Morshead, C.M., B.A. Reynolds, C.G. Craig, M.W. McBurney, W.A. Staines,
D. Morassutti, S. Weiss, and D. Van der Kooy (1994) Neural stem cells
in the adult mammalian forebrain: A relatively quiescent subpopulation
of subependymal cells. Neuron 13:1071–1082.
Mullen, R.J., C.R. Buck, and A.M. Smith (1992) NeuN, a neuronal specific
nuclear protein in vertebrates. Development 116:201–211.
Müller, C.M. and J. Best (1989) Ocular dominance plasticity in adult cat
visual cortex after transplantation of cultured astrocytes. Nature
Norenberg, M.D. (1994) Astrocyte responses to CNS injury. J. Neuropathol.
Exp. Neurol. 53:213–220.
Okano, H.J., D.W. Pfaff, and R.B. Gibbs (1993) RB and Cdc2 expression in
brain: Correlation with 3H-thymidine incorporation and neurogenesis.
J. Neurosci. 13:2930–2938.
O’Kusky J. (1985) Synapse elimination in the developing visual cortex: A
morphometric analysis in normal and dark-reared cats. Dev. Brain Res.
Ourednik, J., W. Ourednik, and H. Van der Loos (1993a) Do foetal neural
grafts induce repair by the injured juvenile neocortex? Neuroreport
Ourednik, J., W. Ourednik, H. Van der Loos, and B.M. Riederer (1993b) Do
embryonic neural grafts induce repair by the injured juvenile neocortex? Soc. Neurosci. Abstr. 19:623.9.
Ourednik, W. and J. Ourednik (1994) Newly formed host cells in grafted
juvenile neocortex express neuron-specific marker proteins. Neuroreport 5:1073–1076.
Pakzaban, P. and O. Isacson (1994) Neural xenotransplantation: Reconstruction of neuronal circuitry across species barriers. Neuroscience 62:989–
Polezhaev, L.V., M.A. Alexandrova, and V.N. Kleshchinov (1988) DNA
synthesis and mitotic division of cortical neurons in adult rats upon
intrabrain transplantation of embryonic nervous tissue. J. Hirnforsch
Price, D.J., J.M.R. Ferrer, C. Blakemore, and N. Kato (1994) Postnatal
development and plasticity of corticocortical projections from area 17 to
area 18 in the cat’s visual cortex. J. Neurosci. 14:2747–2762.
Privat, A., H. Mansour, A. Pavy, M. Geffard, and F. Sandillon (1986)
Transplantation of dissociated foetal serotonin neurons into the transected spinal cord of adult rats. Neurosci. Lett. 66:61–66.
Reynolds, B.A. and S. Weiss (1992) Generation of neurons and astrocytes
from isolated cells of the adult mammalian central nervous system.
Science 255:1707–1710.
Reynolds, B.A., W. Tetzlaff, and S. Weiss (1992) A multipotent EGFresponsive striatal embryonic progenitor cell produces neurons and
astrocytes. J. Neurosci. 12:4565–4574.
Richards, L.J., T.J. Kilpatrick, and P.F. Bartlett (1992) De novo generation
of neuronal cells from the adult mouse brain. Proc. Natl. Acad. Sci. USA
Ruiz-Flandes, P., B. Demierre, L. Mattenberger, and A.C. Kato (1993)
Migration of neurons after transplantation. Migration of purified
embryonic motoneurons grafted into adult mouse CNS. Int. J. Dev.
Neurosci. 11:525–533.
Saleh, M. and P.F. Bartlett (1989) Isolation of a Thy-1-like glycoprotein
from cat brain: Distribution in retina and brain defined by monoclonal
antibodies. J. Neurosci. 23:152–161.
Shatz, C.J. and M.B. Luskin (1986) The relationship between the geniculocortical afferents and their cortical target cells during development of
the cat’s primary visual cortex. J. Neurosci. 6:3655–3668.
Sieradzan, K. and G. Vrbová (1989) Replacement of missing motoneurons
by embryonic grafts in the rat spinal cord. Neuroscience 31:115–130.
Skup, M., D.G. Herrera, A. Ribeiro-da-Silva, and A.C. Cuello (1993)
Neurogenesis in the adult rat hippocampus: Lesion stimulated phenomenon? In A.C. Cuello (ed): Neuronal Cell Death and Repair. New York:
Elsevier, pp. 267–281.
Sloan, D.J., M.J. Wood, and H.M. Charlton (1991) The immune response to
intracerebral neural grafts. Trends Neurosci. 14:341–346.
Snider, R.S. and W.T. Niemer (1961) A Stereotaxic Atlas of the Cat Brain.
Chicago: The University Press.
Society for Neuroscience (1991) Handbook for the Use of Animals in
Neuroscience Research. Washington, DC: Society for Neuroscience.
Sofroniew, M.V., O. Isacson, and A. Björklund (1986) Cortical grafts prevent
atrophy of cholinergic basal nucleus neurons induced by excitotoxic
cortical damage. Brain Res. 378:409–415.
Srensen, J.C., J. Zimmer, and A.J. Castro (1989) Fetal cortical transplants reduce thalamic atrophy induced by frontal cortical lesions in
newborn rats. Neurosci. Lett. 98:33–38.
Srensen, J.C., A.J. Castro, E.J. Neafsey, and J. Zimmer (1990) Reduction
of postlesion atrophy by transplants of fetal cerebral tissue: Hosttransplant nerve connections and trophic factors. In A. Björklund, A.J.
Aguayo, and D. Ottoson (eds): Brain Repair, Vol. 56: Wenner-Gren
Center International Symposium Series. New York: M-Stockton Press,
pp. 122–144.
Sotelo, C., R.M. Alvarado-Mallart, R. Gardette, and F. Crapel (1990) Fate of
grafted embryonic Purkinje cells in the cerebellum of the adult ‘‘Purkinje cell degeneration’’ mutant mouse. I. Development of reciprocal
graft-host interactions. J. Comp. Neurol. 295:165–167.
Stein, D.G., R. Labbe, M.J. Attella, and H.A. Rakowsky (1985) Fetal brain
tissue transplants reduce visual deficits in adult rats with bilateral
lesions of the occipital cortex. Behav. Neural Biol. 44:266–277.
Stemple, D.L. and N.K. Mahanthappa (1997) Neural stem cells are blasting
off. Neuron 18:1–4.
Stevens, A. (1990) The hematoxylins. In J.D. Bancroft and A. Stevens (eds):
Theory and Practice of Histological Techniques, 3rd Ed. Edinburg:
Churchill Livingstone, pp. 107–118.
Svendsen, C.N. and A.E. Rosser (1995) Neurones from stem cells. TINS
Takacs, J., P. Saillour, M. Imbert, M. Bogner, and J. Hamori (1992) Effect of
dark rearing on the volume of visual cortex (areas 17 and 18) and
number of visual cortical cells in young kittens. J. Neurosci. Res.
Takahashi, T., R.S. Nowakowski, and V.S. Caviness, Jr. (1992) BUdR as an
S-phase marker for quantitative studies of cytokinetic behaviour in the
murine cerebral ventricular zone. J. Neurocytol. 21:185–197.
Targett, M.P. and W.F. Blakemore (1994) The use of xenografting to
evaluate the remyelinating potential of glial cell cultures. Eye 8:238–
Tusa, R.J., L.A. Palmer, and A.C. Rosenquis (1981) Multiple cortical visual
areas: Visual field topography in the cat. In C.N. Woolsey (ed): Cortical
Sensory Organisation. Multiple Visual Areas, Vol. 2. Clifton, NJ:
Humana Press, pp. 1–31.
Valoušková, V. and J. Gálik (1995) Unilateral grafting of fetal neocortex
into a cortical cavity improves healing of a symmetric lesion in the
contralateral cortex of adult rats. Neurosci. Lett. 186:103–106.
Varon, S. and J.M. Conner (1994) Nerve growth factor in CNS repair. J.
Neurotrauma 11:473–486.
Vescovi, A.L., B.A. Reynolds, D.D. Fraser, and S. Weiss (1993) bFGF
regulates the proliferative fate of unipotent (neuronal) and bipotent
(neuronal/astroglial) EGF-generated CNS progenitor cells. Neuron
Villablanca, J.R., B.L. Shook, D.A. Hovda, and R.L. Sutton (1990) Transplantation of fetal frontal cortex onto degenerating thalamus of cats and
kittens. Dev. Neurosci. 12:1–10.
Von Bertalanffy, F.D. and K.P. Nagy (1962) Fluorescence microscopy and
photomicrography with acridine orange. Med. Radiogr. Photogr. 38:
Warrington, A.E., E. Barbarese, and S.E. Pfeifer (1993) Differential myelinogenic capacity of specific developmental stages of the oligodendrocyte
lineage upon transplantation into hypomyelinating hosts. J. Neurosci.
Res. 34:1–13.
Webster, W.G. (1981) Morphological asymmetries of the cat brain. Brain
Behav. Evol. 18:72–79.
Weiss, S., B.A. Reynolds, A.L. Vescovi, C. Morshead, C. Craig, and D. Van
der Koy (1996) Is there a neural stem cell in the mammalian forebrain?
TINS 19:387–393.
Wells, J., B.P. Vietje, and D.G. Wells (1987) Migration and differentiation of
xenogeneic and homogeneic brain cells transplanted into the adult rat
hippocampus. Ann. N.Y. Acad. Sci. 495:804–806.
Yinon, U. and S. Gelerstein (1991) Isochronic transplantation of neonatal
grafts in the visual cortex of cats: Responsiveness, ocular dominance
and specificity of cortical cells to visual stimulation. Exp. Brain Res.
Yinon, U., R. Shemesh, H. Arda, G. Dobin, and P.P. Jaros (1993) Physiological studies in deafferented visual cortex cells of cats following transplantation of fetal xenografts from the rat’s cortex. Exp. Neurol. 122:335–
Zhou, C.-F., G. Raisman, and R.J. Morris (1985) Specific patterns of fibre
outgrowth from transplants to host mice hippocampi, shown immunohistochemically by the use of allelic forms of Thy-1. Neuroscience
Zhou, H.F., L.H.-C. Lee, and R.D. Lund (1990) Timing and pattern of
astrocytic migration from xenogeneic transplants of the cortex and
corpus callosum. J. Comp. Neurol. 292:320–330.
Без категории
Размер файла
1 275 Кб
Пожаловаться на содержимое документа