(Stroke. 1997;28:1049-1059.)
© 1997 American Heart Association, Inc.
Articles |
Short- and Long-term Changes in Striatal Neurons and Astroglia After Transient Forebrain Ischemia in Rats
From the Department of Biomedical Sciences, Section of Physiology, University of Modena, and Interuniversity Center for the Study of Aging, Milan, Italy.
Correspondence to Dr Michele Zoli, Dipartimento di Scienze Biomediche, Sezione di Fisiologia, Università di Modena, via Campi 287, 41100, Modena, Italy. E-mail agnati{at}c220.unimo.it
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Background and Purpose The striatum is one of the regions most sensitive to transient forebrain ischemia. After 30-minute ischemia, areas of massive neuronal degeneration are clearly detectable a few hours after the insult and attain their maximal extension 24 hours after the insult. However, for most cellular and neurochemical parameters it is not known whether some recovery occurs at later times. We examined certain cell populations in the caudate putamen at different times after transient ischemia.
Methods Adult male Sprague-Dawley rats were subjected to 30-minute forebrain ischemia (four-vessel occlusion model). Six experimental groups were considered: control animals and ischemic animals killed 4 hours, 1 day, 7 days, 40 days, and 8 months after reperfusion. Three striatal cell populations were examined by means of immunocytochemistry coupled to computer-assisted image analysis: vulnerable medium spiny neurons, resistant aspiny neurons, and reactive astrocytes, labeled for their content of dopamine- and cAMP-regulated phosphoprotein mr32 (DARPP-32), somatostatin and neuropeptide Y, and glial fibrillary acidic protein, respectively.
Results (1) The area containing DARPP-32 immunoreactive neurons was markedly decreased (15% to 20% of control caudate putamen area) at 1 day after reperfusion and partially recovered at the following times (40% to 50% at 7 days and 50% to 60% at 40 days and 8 months after reperfusion). (2) The appearance of reactive astrocytes was precocious (4 hours to 1 day after ischemia) in the medial caudate putamen, the region in which DARPP-32 recovered within 40 days after ischemia, and late (7 to 40 days after ischemia) in the lateral caudate putamen, where no DARPP-32 recovery was detected. (3) Neuropeptide Y/somatostatin–containing neurons resisted the ischemic insult and could be detected in areas devoid of DARPP-32 immunoreactive neurons as long as 8 months after reperfusion.
Conclusions The present results show a marked recovery of DARPP-32–positive neurons within 40 days after 30-minute forebrain ischemia in the medial, but not the lateral, caudate putamen. Medial caudate putamen also contains a high density of reactive astrocytes on the first day after ischemia, suggesting that astrocytic support has an important role in the spontaneous recovery of ischemic neurons.
Key Words: astrocytes • cerebral ischemia, global • neuronal death • immunohistochemistry • rats
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The striatum is one of the regions most sensitive to transient forebrain ischemia induced by the 4VO method.1 2 3 After ischemia of long duration (30 minutes), massive degeneration and loss of Nissl-stained neurons are clearly detectable a few hours after the insult and attain maximal extension 24 hours after the insult.2 Neuronal loss is mainly restricted to the lateral part of the caudate putamen.2 3 Cellular alterations include loss of medium spiny projection neurons,2 4 largely corresponding to dopaminoceptive neurons,5 6 and increase in reactive astrocytes2 7 and microglia.8 On the other hand, large cholinergic4 and medium aspiny SS/NPY-containing interneurons are resistant to the ischemic insult.2 7 These changes in cell populations are accompanied by a number of neurochemical alterations. For instance, dopamine D1 receptors, dopamine D2 receptor mRNA (but not D2 bind- ing5 9 10 ), [3H]forskolin binding,5 and µ and
opiate binding11
show a patchy disappearance from the lateral caudate putamen, and
glutamate and dopamine neurotransmissions are deeply
altered.12 13 Most of the aforementioned alterations have been detected between 1 and 7 days after the insult. In a few instances, such as in the case of SS and NPY IR, full recovery is attained in 7 (SS) or 40 (NPY) days after ischemia.7 However, for most cellular and neurochemical parameters, it is not known whether some recovery is present after the first days after ischemia.
In the present study we investigated the development of ischemic lesions in the caudate putamen of the male rat at early (4 hours to 7 days) and late (40 days to 8 months) times after 30 minutes of forebrain ischemia induced by 4VO.3 14 Three striatal cellular populations were examined by means of immunocytochemistry coupled to computer-assisted image analysis: vulnerable medium spiny neurons, labeled for their content of DARPP-32; resistant aspiny neurons, labeled for their content of SS and NPY; and reactive astrocytes, labeled for their increased content of GFAP.
DARPP-32 is present in approximately half of the striatal medium
spiny neurons,15 16 which constitute more than 90% of the
striatal neuronal population. This phosphoprotein plays a role in
upregulating the gain of the transduction mechanism associated with the
dopamine D1 transmission.17 NPY and SS are colocalized in
a small population of striatal interneurons (
1% of neostriatal
neurons), which also contain high levels of nitric oxide
synthase.18 19 Striatal nitric oxide synthase–containing
neurons are also spared in other models of brain injury20
and human brain pathology.21 GFAP is a specific class of
intermediate filament that is present in the cytoplasm of
differentiated astrocytes.22 23 It has been shown that
GFAP IR increase is a marker of astroglial reaction to various types of
brain injury, either mechanical,24 25 26 27
toxic,28 or ischemic.7
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Animals
Male specific pathogen–free Sprague-Dawley rats (Charles River, Calco, Italy; body weight, 250 to 300 g) were used. Rats were kept one per cage under standardized humidity, temperature, and lighting conditions (light on at 8 AM and off at 8 PM) and given food pellets and water ad libitum.
Induction of Transient Forebrain Ischemia
All the procedures used were in accordance with institutional
(Italian Ministero della Sanità) guidelines for animal care. The
rats were deeply anesthetized with halothane (Fluothane,
Zeneca; 4% during the induction phase and 1.5% during the
maintenance phase), and surgery was performed according to the
4VO model as previously described.3 14 29 Briefly, both
common carotid arteries were exposed, an atraumatic silk thread
(EP15ex4/0) was loosely placed around each of them, and the incision
was closed with a single suture. Subsequently, both vertebral arteries
were electrocauterized and permanently occluded. The rats were allowed
to recover from anesthesia for 24 hours. At this time the
spontaneous behavior and the electroencephalographic recording
of these animals were similar to those of nonoperated rats. The rats
were then anesthetized with halothane (1%), the ventral neck
suture was removed, and the rats were allowed to recover from
anesthesia for 1 minute. Then both carotid arteries were
occluded with stainless steel clips (Biemer-Clip 0.29-0.39,
Aesculap-Werke). The body temperature was monitored with a rectal
thermometer, and animals were kept under a heating lamp until thermal
homeostasis was restored. The carotid clips were removed 30 minutes
later, and restoration of blood flow through these arteries was
verified by direct inspection.
During the ischemic period, the animals were tested for their level of consciousness, the presence or absence of righting and corneal reflexes, ability to walk and to climb, and electroencephalographic activity. Only those animals immediately losing their righting reflex, unresponsive for 20 to 30 minutes after 4VO, with an isoelectric electroencephalogram within 2 to 3 minutes after 4VO, and without recovery throughout the ischemic period were studied (for a full description of the animal model, see Zini et al3 and Grimaldi et al7 ).
Six experimental groups were considered: control animals (n=6) and ischemic animals killed 4 hours (n=8), 1 day (n=11), 7 days (n=20), 40 days (n=8), and 8 months (n=5) after 4VO.
When long-term effects of a lesion are evaluated, selective mortality within the animal population (ie, only less lesioned animals survive) may bias the results obtained. In this experiment, mortality after 4VO was restricted to the first day after reperfusion. Therefore, changes in the parameters investigated (ie, striatal cell populations) after 1 day following ischemia cannot be ascribed to selection of the animal population. Beginning 1 day after ischemia, the animals were randomly assigned to the different groups.
In our experience (see, for example, References 3, 7, 30, and 313 7 30 31 ), approximately 20% of the rats satisfying the criteria described above do not develop typical histological lesions in selectively vulnerable regions. Accordingly, 2 of 8 rats at 4 hours, 2 of 11 rats at 1 day, 3 of 20 rats at 7 days, 2 of 8 rats at 40 days, and 1 of 5 rats at 8 months did not show signs of neuronal insult in the striatum. Only the animals with signs of neuronal insult were considered in the study.
Immunocytochemistry
Immunocytochemistry was performed as previously
described.32 Six sections per animal for each antiserum
were taken at various coronal levels of the precommissural striatum
(regularly spaced from bregma 1.7 to 0.2 mm according to Paxinos
and Watson33 ). The following primary antisera were used:
mouse monoclonal antibody against DARPP-32 (16), rabbit polyclonal
antiserum against GFAP (Dako, lot No. 015), rabbit polyclonal antiserum
against SS,34 and rabbit polyclonal antiserum against NPY
(Peninsula, lot No. 006802-3), which have been previously
characterized. The antisera were used in a dilution of 1:2000 for
DARPP-32, 1:300 for GFAP, and 1:1500 for NPY and SS.
Image Analysis Procedures
Morphometric and microdensitometric analyses were
performed by means of an automatic image analyzer (IBAS I-II,
Zeiss Kontron).32 35
In the analysis of the extent of DARPP-32 IR-positive and -negative areas, the section was acquired by the television camera from the microscope at low magnification so that an entire striatum was analyzed at each time (size of field, 23.0 mm2). The caudate putamen was manually encircled by means of a mouse, and its area was measured. The decrease in total caudate putamen area per section in ischemic animals with respect to the total caudate putamen area per section in control animals was considered an index of regional atrophy. Subsequently, a densitometric discrimination35 was performed to separate DARPP-32 IR-positive from DARPP-32 IR-negative areas. An area of nonspecific labeling was selected, and the mean±SD of gray values was measured. The mean gray value of the nonspecific labeling -3 SD was considered a threshold for specific labeling. This procedure was repeated for each section analyzed (for further details, see References 32 and 3532 35 ). The parameters considered were size and optical density of the DARPP-32 IR-positive areas.
Manual cell count of DARPP-32 ir cells was performed in rectangular
fields (153x117 µm) acquired by the television camera from the
microscope (x40 magnification) at bregma level 0.8 mm. For each
caudate putamen, four fields were selected in the medial, centromedial,
centrolateral, and lateral parts (Fig 1
).
|
The analysis of GFAP ir cells was performed at bregma level
0.8 mm. Rectangular fields (153x117 µm) were acquired by
the television camera from the microscope (x40 magnification). For
each caudate putamen, five fields were selected. In control, 4-hour,
and 1-day postischemic animals, the sampled areas were
located in the medial, centromedial, centrolateral, lateral, and dorsal
parts of the caudate putamen (Fig 1
). In the 7- and 40-day groups,
sampling of medial, centromedial, and dorsal parts was maintained,
whereas in the lateral region (which at these time intervals contains
lesioned tissue, ie, is devoid of DARPP-32 ir neurons) two fields were
analyzed, one located in the core and the other on the medial
boundary of the DARPP-32 IR-negative area.
The number of positive cells in the field (n) was directly counted by the experimenter, who was unaware of treatment groups. Manual cell count was preferred to automatic count because it was difficult for the image analyzer to safely attribute the spiderlike GFAP ir processes of the astrocyte to a single positive cell. Then a densitometric discrimination was performed as described above to select the GFAP ir structures from the nonspecific background. Two parameters were obtained from this analysis: the field area (FA), ie, the overall GFAP ir area in the field, and the specific optical density (spOD) of GFAP ir structures. We then obtained a global parameter, the integrated optical density (intOD=FA*spOD), which gives an index of the overall amount of ir material in the field. Finally, the intOD was divided by n to obtain the average amount of GFAP ir material per positive cell in the sampled field.
Statistical analysis of a given parameter at different time intervals was performed by means of one-way ANOVA, followed by Bonferroni's correction for multiple comparisons, with one value per field per animal. The presence of a significant rostrocaudal trend in the size of the DARPP-32 IR-negative area was tested by means of the Jonckheere-Terpstra test for ordered alternatives.36
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Time Course of Changes in DARPP-32 IR
At 4 hours, a marked decrease in DARPP-32 IR was present in the caudate putamen. A decrease in DARPP-32 IR consisted of either reduction (Fig 2
) or complete loss (as observed at later
postreperfusion times) of positive cells and neuropil. The area of
DARPP-32 IR decrease was rather variable in different animals,
ranging from 10% to more than 90% of total caudate putamen area
(mean±SD=57.9±29.7%).
|
Beginning 1 day after ischemia, areas of complete DARPP-32 IR
disappearance (both cell bodies and neuropil) were consistently
observed in the caudate putamen (Fig 3). At 1 day after
ischemia, the lesion was very large (75% to 95% of the entire
caudate putamen area) in 8 of 9 animals. The remaining DARPP-32 ir
cells were usually located in a small strip close to the ventricle
(only its more ventral part in the largest lesions). In relatively
smaller lesions, islands of DARPP ir cells were also located close to
the corpus callosum. In the DARPP-32 IR-negative area, a nonspecific
deposition of diaminobenzidine was often observed (Fig 3A).
|
At later times the DARPP-32 IR-negative area became patchy and
progressively smaller. A patchy disappearance of DARPP-32 ir cells was
observed mainly in the lateral part of the caudate putamen (Figs 3B and 4A through 4C). The medial half of the caudate putamen
and the nucleus accumbens were spared in the vast majority of animals.
In most animals, small islands of DARPP-32 ir cells, particularly those
located in the subcallosal region, were still visible in the lateral
caudate putamen.
|
The lesion was more patchy and smaller at rostral than at caudal levels
(Fig 4A through 4C). The size of the DARPP-32 IR-negative area was
evaluated at five coronal levels from 1.4 to 0.2 mm from bregma. A
rostrocaudal trend toward an increase of DARPP-32 IR-negative area was
observed (Fig 5).
|
At 40 and 240 days after ischemia, the DARPP-32 IR-negative
areas appeared progressively smaller (Fig 3C and 3D). Loss of DARPP-32
IR was mostly confined to a single patch located in the lateral part of
the caudate putamen.
A morphometric analysis of DARPP-32 IR was performed at two
coronal levels (1.1 and 0.2 mm from bregma) of the caudate putamen
at several time intervals (from 1 day to 8 months) after reperfusion.
This analysis showed two phenomena (Figs 3 and 6): (1) the total caudate putamen area progressively
decreased up to 8 months after ischemia, when the size of this
region became approximately 60% that of control animals (see also
References 37 and 3837 38 ); (2) the DARPP-32 IR-positive area, expressed in
percentage of the total caudate putamen area in control animals (Fig 6), was approximately 15% to 20% of total control area at 1 day, 40%
to 50% of total control area at 7 days, 50% to 60% of total control
area at 40 days, and remained stable thereafter. When the rate of
recovery between 1 and 7 days is considered, it is likely that maximal
recovery of the DARPP-32 IR-positive area was, in fact, already
attained approximately 10 days after ischemia.
|
The microdensitometric analysis of the intensity of DARPP-32 IR
in the positive areas did not show any significant difference between
ischemic and control animals (Table).
|
Counting of DARPP-32 ir cells was performed in several fields of the
caudate putamen (Fig 1) from 4 hours to 40 days after reperfusion. In
the medial field, no significant change in DARPP-32 ir cell density was
observed at any time after ischemia (Fig 7). In
the centromedial, centrolateral, and lateral fields, a progressive
decline in the number of DARPP-32 ir cells was present from 4 hours
to 1 day after reperfusion. However, in the centromedial field full
recovery of DARPP-32 ir cell density was attained within 40 days,
whereas in centrolateral and lateral fields no recovery was observed
(Fig 7).
|
These findings indicate that a population of DARPP-32 ir neurons (an
irregular strip of cells mainly located in the medial portion of the
caudate putamen; Fig 3A) constituting 15% to 20% of the total caudate
putamen neurons does not lose DARPP-32 IR 1 day after ischemia.
A second population (a strip of cells located more laterally than the
previous strip of cells) constituting approximately 30% to 40% of the
total caudate putamen neurons is DARPP-32 IR-negative at 1 day,
recovers DARPP-32 IR between 1 and 40 days after ischemia (Fig 3B), and remains positive thereafter. Since this process is accompanied
by a progressive shrinkage of the part of the caudate putamen devoid of
DARPP-32 ir cells, the ratio between DARPP-32 IR-positive and -negative
areas gradually increases.
Time Course of Changes in GFAP IR
In control animals, a strip of GFAP ir astrocytes was present in
the subventricular and subcallosal regions of the caudate
putamen, whereas only scattered GFAP ir astrocytes were detected in the
rest of the caudate putamen (Fig 8).
|
An increase in GFAP IR was observed at all post-reperfusion times analyzed. At both 4 and 24 hours after reperfusion, a slight increase in GFAP IR was observed. Overall, three mediolateral subregions could be recognized, as follows.
(1) A medial subregion (20% of caudate putamen). This
subregion (with normal DARPP-32 ir neurons; Fig 7) showed normal GFAP
ir astrocytes in terms of number and staining intensity. It must be
noted, however, that in this subregion the number of GFAP ir astrocytes
in control animals was three times higher than in the rest of the
caudate putamen (Fig 9).
|
) and integrated optical
density (intOD) of GFAP IR per cell (an index of the amount of GFAP ir
material per cell) (
). For further details, see text.
*P<.05 vs respective control according to one-way ANOVA
followed by Bonferroni's correction for multiple comparisons.
(2) A centromedial and dorsal subregion (40% of caudate
putamen). This subregion (with a slight [at 4 hours] or total
[at 24 hours] disappearance of DARPP-32 ir neurons; Fig 7) showed
marked and significant increases in the number and ir content of
GFAP-positive astrocytes at both 4 and 24 hours (Figs 8C and 9).
(3) A centrolateral and lateral subregion (40% of caudate
putamen). This subregion (with a marked [at 4 hours] or total
[at 24 hours] disappearance of DARPP-32 ir neurons; Fig 7) showed no
change in number and a limited increase in ir content of GFAP-positive
cells with respect to control animals at 4 hours (Figs 8D and 10). No GFAP ir astrocyte could be detected in these
areas at 24 hours (Fig 10). As in the case of DARPP-32, a nonspecific
deposition of diaminobenzidine was often detected (see above).
|
Note that to speak of a mediolateral gradient is an oversimplification because the areas of DARPP-32 IR loss and GFAP IR changes were patchy, and islands of resistant neurons were often observed in the dorsal and lateral subcallosal portions of caudate putamen at all time intervals analyzed (see above).
At 7 days after ischemia, GFAP IR still had an
inhomogeneous distribution within the caudate putamen but
overall was much higher than at 4 and 24 hours. Again, three
mediolateral regions could be detected: (1) a medial subregion with
GFAP ir astrocytes not significantly different from control animals and
4- to 24-hour postischemic animals (Fig 9); (2) a
centromedial and dorsal subregion (corresponding to the region where
DARPP-32 IR recovers; Figs 3B and 7) with an increased number of GFAP
ir astrocytes showing increased GFAP IR compared with control but not
significantly different from the 24-hour group (Figs 8E and 9); and (3)
a centrolateral and lateral subregion (corresponding to the region
devoid of DARPP-32 ir neurons) with a huge increase of GFAP IR compared
with both control and 4- to 24-hour postischemic animals
(Figs 8F and 10). GFAP IR increase was not homogeneous
inside the subregion. In small lesions (less than 1 mm of
minimal diameter) the entire area was filled with strongly stained
astrocytes. In large lesions (more than 1 mm of minimal
diameter) the core was practically devoid of GFAP ir cells, while the
periphery contained strongly stained cells (Fig 4D through 4F). Since
the area of neuronal loss was larger at caudal levels, areas devoid of
GFAP IR were also larger at caudal levels (Fig 4D through 4F). On
average, approximately 20% of the caudate putamen area was devoid of
both DARPP-32 and GFAP ir cells (area of pannecrosis). Both small and
large lesions were encircled by a strip of less strongly stained cells.
The cellular analysis showed that the increase in GFAP IR was
due to a marked increase in GFAP IR cell content compared with both
control and previous postischemic time intervals. The
number of GFAP ir cells was significantly increased versus control and
previous postischemic time intervals in the centrolateral
area but not in the lateral area, where, however, GFAP ir cell number
was already high in control animals.
At later postischemic times (40 days and 8 months; Figs 4H and 11, respectively) the areas of DARPP-32 IR
disappearance (which were progressively smaller; see above) rather
uniformly contained high levels of GFAP IR (increased cell number and
ir content) comparable to those observed at 7 days (Fig 10). Instead,
the parts of caudate putamen containing DARPP-32 ir neurons (medial,
centromedial, and dorsal parts) had control or slightly increased GFAP
IR (Fig 9).
|
Summary of Region-Specific Changes in DARPP-32 and GFAP IR
Based on the patterns of change in DARPP-32 and GFAP IR described
above, the caudate putamen can be subdivided into four subregions with
different fates after 30-minute 4VO:
(1) A medial subregion. This subregion (comprising on average 20% of control caudate putamen area) showed no significant change in either DARPP-32 or GFAP IR at any postischemic time. Note that GFAP IR in this region was much higher than in the rest of the striatum in control animals.
(2) A centromedial and dorsal subregion. In this subregion (comprising on average 40% of control caudate putamen area), DARPP-32 ir neurons were not detected at 1 day but recovered their IR within 40 days (likely the recovery was already complete within 10 days). In this subregion GFAP IR was already increased at 4 hours to 1 day, remained elevated at 7 days, and returned to control levels at later time intervals studied.
(3) A centrolateral subregion. In this subregion (comprising on average 20% of control caudate putamen area), DARPP-32 ir neurons were not detected at 1 day and did not recover within 40 days. GFAP ir astrocytes were not detected at 1 day but were moderately to highly concentrated from 7 days to 8 months after ischemia. It is likely that astroglia remained activated until the subregion was completely reabsorbed. Accordingly, GFAP IR was elevated as long as 21 months after 4VO in the small DARPP-32 IR-negative areas still present in some of these animals (M. Zoli and R. Grimaldi, unpublished data, 1994).
(4) A lateral and caudal subregion. In this subregion (comprising on average 20% of control caudate putamen area), corresponding to the core of large ischemic lesions, DARPP-32 ir neurons and GFAP ir astrocytes were not detected from 4 hours to 7 days after ischemia (area of pannecrosis). This subregion was no longer detected at 40 or more days after ischemia.
SS and NPY IRs
As previously reported by Grimaldi et al,7 the number
of NPY and SS ir nerve cell profiles is markedly decreased at early
time intervals (4 hours and 1 day after the ischemic injury).
Subsequently (within 40 days after reperfusion), a complete recovery of
these ir cell populations takes place.
NPY and SS IR underwent further changes at 8 months after
ischemia. DARPP-32 IR-negative areas contained a markedly
higher density of NPY and SS IR neurons than DARPP-32 IR-positive
areas. In addition, these neurons were markedly more stained than the
neurons present in DARPP-32 IR-positive areas (Fig 11). The
apparent increase in cell density is likely due, on the one hand, to
the shrinkage of DARPP-32 IR-negative striatal tissue with an
accumulation of resistant SS and NPY ir neurons and, on the
other hand, to the detection of normally unstained neurons caused by
their increased antigen content.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Based on present morphological results, the caudate putamen can be subdivided into four mediolateral subregions with different fates after 30-minute 4VO (see above).
Region 1 is substantially unmodified by the ischemic insult.
Region 2 shows an initial disappearance and a subsequent progressive recovery of DARPP-32 ir neurons. At 40 days DARPP-32 ir neuron density is not significantly different from that of control animals. DARPP-32 IR disappearance at 1 day is likely to be a consequence of the depression of protein synthesis, which is known to occur in the early phase after ischemia.39 40 41 The presence of sparse DARPP-32 ir neuron loss (ie, fewer than one neuron per sampled field) occurring at late times after ischemia (such as in slowly progressive neuronal damage31 42 ) in this region cannot be excluded, however. Maximal increase in the number of GFAP ir astrocytes is already observed in this area 4 to 24 hours after ischemia. Since previous studies have shown that astrocyte hyperplasia is not present at early postischemic times,43 44 the increased number of GFAP ir astrocytes in region 2 likely corresponds to hypertrophic resident glia.
Region 3 shows an almost complete loss of DARPP-32 ir neurons at all
times after ischemia. In this region, GFAP IR is low or absent
at 4 to 24 hours after ischemia but markedly increases at later
(
7 days) postischemic times. Thus, during the maturation
of the ischemic lesion2 in neurons of region 3,
astrocytes are degenerated and/or biosynthetically impaired. Long-term
persistence of reactive astrocytes in areas of DARPP-32 ir neuron loss
correlates well with previous evidence obtained in the hippocampal
formation, where GFAP IR remains elevated for more than 2 months in
areas where neuronal loss has occurred (ie, CA1 field) but returns
toward basal levels within 40 days in areas where no neuronal loss has
occurred (ie, CA3 field).7 38 The present
analysis cannot determine whether the strongly GFAP ir
astrocytes detected at 7 days and later are hypertrophic resident
astroglia or proliferated astroglia. This latter hypothesis is favored
by previous evidence that mitotic figures or
[3H]thymidine accumulation is observed in the astroglia
located near the core of the infarct area starting at 2 to 3 days after
ischemia.43 44
Region 4 is devoid of vulnerable neurons and reactive astrocytes at early postischemic times (pannecrosis) and may be reabsorbed within 40 days.
Several anatomic and neurochemical features of striatal subregions may account for the gradient for increased vulnerability from the ventromedial to the dorsolateral part and from the rostral to the caudal part of the caudate putamen. For instance, the dorsolateral and caudal parts of the caudate putamen receive the distal portion of the neostriatal vascular supply.45 Somatomotor and limbic components of glutamatergic and dopaminergic inputs project to dorsolaterally and ventromedially located neostriatal cells, respectively.46 47 Indeed, the metabolism/flow uncoupling phenomenon, which predisposes to neuronal degeneration48 and is caused at least in part by dopamine input, was mainly evident in the dorsolateral striatum.12
The fate of the resistant NPY/SS ir interneurons is different from that of DARPP-32 ir neurons. Previous data7 showed a decrease in NPY and SS ir cell bodies at early time intervals (4 hours and 1 day after reperfusion) followed by a recovery between 7 and 40 days after reperfusion. In this report we show that as long as 8 months after ischemia NPY and SS ir neurons are detected in areas of DARPP-32 ir neuron loss and contain markedly higher levels of the two neuropeptides. Increased levels of these peptides may derive from altered trans-synaptic regulation of their biosynthesis and release after ischemia (for example, see Reference 4949 ). As a result of the anatomic polarization of their dendritic and axonal arborization, NPY- and SS-containing neurons may represent an intrinsic system capable of connecting the patch and matrix striatal compartments.19 50 However, it seems unlikely that these neurons, surviving in areas devoid of the vast majority of the other neuronal types, have any functional role in striatal circuitry, although their volume transmission inputs and outputs may still be maintained.51 In any case, their long-term survival indicates that they are capable of maintaining trophic links with surviving neuronal afferents to striatum and local glia, thus representing an example of regrowth (neuronal survival after lesion) without regeneration (reformation of a network with functional significance52 ).
DARPP-32 ir neurons in regions 1 and 2 are either resistant or only transiently impaired after 30-minute 4VO. In region 1, astroglia expressing high GFAP levels are already present in control animals. In region 2, an increased number of astroglia with increased GFAP content is already attained within 4 hours. The high density of reactive astroglia present during the first postischemic day in regions 1 and 2 may contribute to the protection or rescue of neurons of these areas. Astroglia are known to exert a neuroprotective action caused by such factors as neurotrophic factor release as well as maintenance of extracellular glutamate, K+ concentration, and pH.53 54 55 56 Interestingly, it has recently been shown that spreading depolarization waves, which are thought to be responsible for damage of neurons lying close to infarcted areas in the postischemic period,57 do not injure neurons unless glial cell function is impaired.58
Regions 3 and 4 are devoid of DARPP-32 ir neurons at all time intervals
studied beginning 1 day after ischemia and are progressively
reabsorbed, so that 8 months after ischemia the caudate putamen
is almost exclusively constituted by regions 1 and 2. GFAP ir astroglia
are either detected at low density or totally absent in regions 3 and 4
at early times after ischemia when they might exert some
neuroprotective action (see above). They are instead highly
concentrated at later times (7 days) in the parts of regions 3 and 4
that have not yet been reabsorbed. It has been shown that reactive
astrocytes appearing in lesioned tissue can have a phagocytic function
complementary to that of microglia.8 59 60 61 Therefore, the
high density of reactive astrocytes in areas devoid of neurons may
contribute to the progressive reabsorption of the lesioned tissue. In
agreement with this hypothesis, it has been shown that high density of
reactive astrocytes and microglia was detected in the hippocampus 2
months after ischemia, when neuronal debris is still
present and the reabsorption process is ongoing. In contrast,
reactive astrocytes and microglia were no longer detected 6 months
after ischemia, when neuronal debris is absent and the tissue
reabsorption process is completed.38
Region 2, and probably region 3 as well, corresponds to the ischemic penumbra.57 62 63 This term refers to the brain tissue surrounding the ischemic core that is transiently inactivated but remains viable and can undergo spontaneous and/or pharmacologically induced recovery. In region 2 there is spontaneous recovery in present conditions. It can be hypothesized that therapeutic interventions administered a few hours to several days after ischemia can rescue neurons in region 3, whereas those administered before or immediately after the ischemic insult can rescue neurons in region 3 and perhaps region 4. However, based on present results caution must be exerted when the efficacy of therapeutic interventions on striatal neurons after transient ischemia is evaluated. When evaluated before 10 days after ischemia, an apparent protective effect might consist of an accelerated recovery of neurons of region 2. On the other hand, treatments evaluated at later time intervals must take into account the progressive shrinkage of the caudate putamen. In fact, an artifactual neuronal recovery occurs at late postischemic times when data are expressed as lesioned area in relation to total area or as overall neuronal density.
|
Selected Abbreviations and Acronyms |
|---|
|
|
Acknowledgments |
|---|
This study was supported by grants from Ministero dell'Università e della Ricerca Scientifica e Technologica and Consiglio Nazionale delle Ricerche. We thank Giuseppe Mancinelli for his technical contribution.
Received October 2, 1996; revision received January 21, 1997; accepted January 21, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1. Brierley J. Cerebral hypoxia. In: Blackwood W, Corsellis J, eds. Grenfield's Neuropathology. London, England: Edward Arnold;1976:43-85.
2. Pulsinelli WA, Brierley JB, Plum F. Temporal profile of neuronal damage in a model of transient forebrain ischemia. Ann Neurol. 1982;11:491-498.[Medline] [Order article via Infotrieve]
3. Zini I, Grimaldi R, Merlo Pich E, Zoli M, Fuxe K, Agnati LF. Aspects of neural plasticity in the central nervous system, V: studies on a model of a transient forebrain ischemia model in male Sprague Dawley rats. Neurochem Int. 1990;16:451-468.
4. Francis A, Pulsinelli WA. Response of GABAergic and cholinergic neurons to transient cerebral ischemia. Brain Res. 1982;243:271-278.[Medline] [Order article via Infotrieve]
5. Benfenati F, Merlo Pich E, Grimaldi R, Zoli M, Fuxe K, Toffano G, Agnati LF. Transient forebrain ischemia produces multiple deficits in dopamine D1 transmission in the lateral neostriatum of the rat. Brain Res. 1989;498:376-380.[Medline] [Order article via Infotrieve]
6. Zoli M, Grimaldi R, Agnati LF, Zini I, Merlo Pich E, Toffano G, Fuxe K. Neurohistochemical studies on striatal lesions induced by transient forebrain ischemia: evidence for protective effects of the ganglioside analogue AGF2. Neurosci Res Comm. 1989;4:153-158.
7. Grimaldi R, Zoli M, Agnati LF, Ferraguti F, Fuxe K, Toffano G, Zini I. Effects of transient forebrain ischemia on peptidergic neurons and astroglial cells: evidence for recovery of peptide immunoreactivities in neocortex and striatum but not hippocampal formation. Exp Brain Res. 1990;82:123-136.[Medline] [Order article via Infotrieve]
8. Gehrmann J, Bonnekoh P, Miyazawa T, Hossmann K-A, Kreutzberg GW. Immunocytochemical study of an early microglial activation in ischemia. J Cereb Blood Flow Metab. 1982;12:257-269.
9. Benfenati F, Cimino M, Zoli M, Grimaldi R, Zini I, Agnati LF. Decrease in messenger RNA levels but not in the density of D2 dopamine receptors in rat striatum after transient forebrain ischemia. Neurosci Lett. 1991;126:6-8.[Medline] [Order article via Infotrieve]
10.
Globus MY-T. Role of dopamine in
ischemic neuronal damage. Stroke. 1989;20:827-828.
11.
Benfenati F, Merlo Pich E, Zoli M, Grimaldi R, Fuxe K,
Agnati LF. Changes in striatal µ and opiod receptors after
transient forebrain ischemia: a quantitative
autoradiographic study. Brain Res. 1991;546:171-175.[Medline]
[Order article via Infotrieve]
12.
Globus MY-T, Ginsberg MD, Harik SI, Busto R, Dietrich
WD. Role of dopamine in ischemic striatal injury:
metabolic evidence. Neurology. 1987;37:1712-1719.
13.
Globus MY-T, Busto R, Dietrich WD, Martinez E, Valdes
I, Ginsberg MD. Effect of ischemia on the in vivo
release of striatal dopamine, glutamate, and 
-aminobutyric acid
studied by intracerebral microdialysis.
J Neurochem. 1988;51:1455-1464.[Medline]
[Order article via Infotrieve]
14.
Pulsinelli WA, Brierley JB. A new model of
bilateral hemispheric ischemia in the unanesthetized
rat. Stroke. 1979;10:267-272.
15. Zoli M, Ferraguti F, Toffano G, Fuxe K, Agnati LF. Neurochemical alterations but not nerve cell loss in aged rat neostriatum. J Chem Neuroanat. 1993;6:131-145.[Medline] [Order article via Infotrieve]
16. Ouimet CC, Miller P, Hemmings HC Jr, Walaas SI, Greengard P. DARPP-32, a dopamine- and adenosine 3':5'-monophosphate-regulated phosphoprotein enriched in dopamine innervated brain regions, III: immunocytochemical localization. J Neurosci. 1984;4:111-124.[Abstract]
17. Hemmings H Jr, Nestler EJ, Walaas IS, Ouimet CC, Greengard P. Protein phosphorylation and neuronal function: DARPP-32, an illustrative example. In: Edelman GM, Gall WE, Cowan WM, eds. Synaptic Function. New York, NY: John Wiley; 1987:213-240.
18. Vincent SR, Kimura H. Histochemical mapping of nitric oxide synthase in the rat brain. Neuroscience. 1992;46:755-784.[Medline] [Order article via Infotrieve]
19. Kawaguchi Y, Wilson CJ, Augood SJ, Emson PC. Striatal interneurones: chemical, physiological and morphological characterization. Trends Neurosci. 1995;18:527-535.[Medline] [Order article via Infotrieve]
20. Beal MF, Kowall NW, Ellison DW, Mazureck MF, Swartz KJ, Martin JB. Replication of the neurochemical characteristics of Huntington's disease by quinolinic acid. Nature. 1986;321:168-171.[Medline] [Order article via Infotrieve]
21. Ferrante RJ, Kowall LW, Beal MF, Richardson EP, Bir ED, Martin JB. Selective sparing of a class of striatal neurons in Huntington's disease. Science. 1986;230:561-563.
22. Bignami A, Dahl D. Differentiation of astrocytes in the cerebellar cortex and the pyramidal tracts of the newborn rat: an immunofluorescence study with antibodies to a protein specific to astrocytes. Brain Res. 1973;49:393-402.[Medline] [Order article via Infotrieve]
23. Eng LF. Glial fibrillary acidic protein (GFAP): the major protein of glial intermediate filaments in differentiated astrocytes. J Neuroimmunol. 1985;8:203-214.[Medline] [Order article via Infotrieve]
24. Eng LF, De Armond SJ. Immunocytochemical study of astrocytes, in normal development and disease. In: Federoff S, Hertz L, eds. Advances in Cellular Neurobiology. New York, NY: Academic Press, Inc; 1982;3:145-171.
25. Ludwin SK. Reaction of oligodendrocytes to trauma and implantation: a combined autoradiographic and immunohistochemical study. Lab Invest. 1985;52:20-30.[Medline] [Order article via Infotrieve]
26. Bignami A, Dahl D, Rueger DG. Glial fibrillary acidic (GFA) protein in normal neuronal cells and in pathological conditions. In: Federoff S, Hertz L, eds. Advances in Cellular Neurobiology. New York, NY: Academic Press, Inc; 1980;1:285-310.
27. Mathewson AJ, Berry M. Observations on astrocyte response to a cerebral stab wound in adult rats. Brain Res. 1985;327:61-69.[Medline] [Order article via Infotrieve]
28. Brock TO, O'Callaghan JP. Quantitative changes in the synaptic vesicle proteins, synapsin I and p38, and the astrocyte specific protein, glial fibrillary acidic protein, are associated with chemical-induced injury to the rat central nervous system. J Neurosci. 1987;7:931-942.[Abstract]
29.
Schmidt-Kastner R, Paschen W, Grosse-Ophoff B, Hossmann
K-A. A modified four-vessel occlusion model for inducing
incomplete forebrain ischemia in rats.
Stroke. 1989;20:938-946.
30.
Merlo Pich E, Grimaldi R, Zoli M, Biagini G, Solfrini
V, Toffano G, Fuxe K, Agnati LF. Siagoside selectively
attenuates morphological and functional striatal impairments induced by
transient forebrain ischemia in rats.
Stroke. 1992;23:234-241.
31. Zoli M, Ferraguti F, Zini I, Bettuzzi S, Agnati LF. Increases in sulphated glycoprotein-2 mRNA levels in the rat brain after transient forebrain ischemia or partial mesodiencephalic hemitransection. Mol Brain Res. 1992;18:163-177.
32. Agnati LF, Fuxe K, Zoli M, Zini I, Härfstrand A, Toffano G, Goldstein M. Morphometrical and microdensitometrical studies on phenylethanolamine-N-methyltransferase and neuropeptide Y immunoreactive neurons in the rostral medulla oblongata of the adult and old male rats. Neuroscience. 1988;26:461-478.[Medline] [Order article via Infotrieve]
33. Paxinos G, Watson C. The Rat Brain in Stereotaxic Coordinates. London, England: Academic Press, Inc; 1982.
34. Johansson O, Hökfelt T, Elde RP. Immunohistochemical distribution of somatostatin-like immunoreactivity in the central nervous system of the adult rat. Neuroscience. 1984;13:265-339.[Medline] [Order article via Infotrieve]
35. Zoli M, Zini I, Agnati LF, Guidolin D, Ferraguti F, Fuxe K. Aspects of neural plasticity in the central nervous system, I: computer-assisted image analysis methods. Neurochem Int. 1990;16:383-418.
36. Hollander M, Wolfe DA. Nonparametric Statistical Methods. New York, NY: John Wiley & Sons, Inc; 1973.
37. Hara H, Harada K, Sukamoto T. Chronological atrophy after transient middle cerebral artery occlusion in rats. Brain Res. 1993;618:251-260.[Medline] [Order article via Infotrieve]
38. Kondo Y, Ogawa N, Asanuma M, Ota Z, Mori A. Regional differences in late-onset iron deposition, ferritin, transferrin, astrocyte proliferation and microglial activation after transient forebrain ischemia. J Cereb Blood Flow Metab. 1995;15:216-226.[Medline] [Order article via Infotrieve]
39. Hossmann KA. Disturbances of cerebral protein synthesis and ischemic cell death. Prog Brain Res. 1993;96:161-177.[Medline] [Order article via Infotrieve]
40.
Krause GS, Tiffany BR. Suppression of protein
synthesis in the reperfused brain. Stroke. 1993;24:747-755.
41.
Kogure K, Kato H. Altered gene expression in
cerebral ischemia. Stroke. 1993;24:2121-2127.
42. Nakano S, Kogure K, Fujikura H. Ischemia-induced slowly progressive neuronal damage in the rat brain. Neuroscience. 1990;38:115-124.[Medline] [Order article via Infotrieve]
43. du Bois M, Bowman PD, Goldsten GW. Cell proliferation after ischemic injury in gerbil brain: an immunocytochemical and autoradiographic study. Cell Tissue Res. 1985;242:17-23.[Medline] [Order article via Infotrieve]
44. Petito CK, Morgello S, Felix JC, Lesser ML. The two patterns of reactive astrocytosis in postischemic rat brain. J Cereb Blood Flow Metab. 1990;10:850-859.[Medline] [Order article via Infotrieve]
45. Welsh FA. Role of vascular factors in regional ischemic injury. In: Kogure K, Hossman A, Siesjo BK, Welsh FA, eds. Molecular Mechanisms of Ischemic Brain Damage, Prog Brain Res 63. Amsterdam, Netherlands: Elsevier Science Publishing Co; 1985:19-27.
46. Kelley AE, Domesick VB. The distribution of the projection from the hippocampal formation to the nucleus accumbens in the rat: an anterograde- and retrograde-horseradish peroxidase study. Neuroscience. 1982;7:2321-2335.[Medline] [Order article via Infotrieve]
47. Björklund A, Lindvall O. Dopamine-containing systems in the CNS. In: Björklund A, Hökfelt T, eds. Handbook of Chemical Neuroanatomy: Classical Transmitters in the CNS, Part 1. Amsterdam, Netherlands: Elsevier Science Publishing Co; 1984:55-122.
48. Tanaka K, Greenberg JH, Gonatas NK, Reivich M. Regional flow-metabolism couple following middle cerebral artery occlusion in cats. J Cereb Blood Flow Metab. 1985;5:241-252.[Medline] [Order article via Infotrieve]
49. Kerkerian L, Salin P, Nieoullon A. Cortical regulation of striatal neuropeptide Y (NPY)-containing neurons in the rat. Eur J Neurosci. 1989;2:181-189.
50. Gerfen CR. The neostriatal mosaic: compartmentalization of corticostriatal input and striatonigral output systems. Nature. 1984;311:461-464.[Medline] [Order article via Infotrieve]
51. Agnati LF, Zoli M, Strömberg I, Fuxe K. Intercellular communication in the brain: wiring versus volume transmission. Neuroscience. 1995;69:711-726.[Medline] [Order article via Infotrieve]
52. Agnati LF, Zini I, Zoli M, Fuxe K, Merlo Pich E, Grimaldi R, Toffano G, Goldstein M. Regeneration in the central nervous system: concepts and facts. In: Symon L, ed. Advances and Technical Standards in Neurosurgery, Vol 16. Vienna, Austria: Springer-Verlag; 1988:4-50.
53. Giulian D, Vaca K, Corpuz M. Brain glia release factors with opposing actions upon neuronal survival. J Neurosci. 1993;13:29-37.[Abstract]
54. Manthorpe M, Rudge JS, Varon S. Astroglial cell contributions to neuronal survival and neuritic growth. In: Federoff S, Vernadakis A, eds. Astrocytes. New York, NY: Academic Press, Inc; 1986:315-376.
55. Needels DL, Nieto-Sampedro M, Cotman CW. Induction of a neurite-promoting factor in rat brain following injury or deafferentation. Neuroscience. 1986;18:517-526.[Medline] [Order article via Infotrieve]
56.
Nieto-Sampedro M, Lewis ER, Cotman CW, Manthorpe M,
Skaper SD, Barbin G, Longo FM, Varon S. Brain injury causes a
time-dependent increase in neuronotrophic activity at the lesion
site. Science. 1982;217:860-861.
57. Hossmann KA. Viability thresholds and the penumbra of focal ischemia. Ann Neurol. 1994;36:557-565.[Medline] [Order article via Infotrieve]
58.
Largo C, Cuevas P, Somjen GG, Martín del
Río R, Herreras O. The effect of depressing glial
function in rat brain in situ on ion homeostasis,
synaptic transmission, and neuron survival. J Neurosci. 1996;16:1219-1229.
59. Raisman G, Field PM. A quantitative investigation of the development of collateral reinnervation after partial deafferentation of the septal nuclei. Brain Res. 1973;50:241-264.[Medline] [Order article via Infotrieve]
60. Matthews DA, Cotman CW, Lynch G. An electron microscopic study of lesion-induced synaptogenesis in the dentate gyrus of the adult rat, I: magnitude and time course of degeneration. Brain Res. 1976;115:1-21.[Medline] [Order article via Infotrieve]
61. Cheng H-W, Jiang T, Brown SA, Pasinetti GM, Finch CE, McNeill H. Response of striatal astrocytes to neuronal deafferentation: an immunocytochemical and ultrastructural study. Neuroscience. 1994;62:425-439.[Medline] [Order article via Infotrieve]
62.
Astrup J, Siesjo BK, Symon L. Threshold in
cerebral ischemia: the ischemic penumbra.
Stroke. 1981;12:723-725.
63. Heiss W-D, Graf R. The ischemic penumbra. Curr Opin Neurol. 1994;7:11-19.[Medline] [Order article via Infotrieve]
This article has been cited by other articles:
 |
|
 |
|
  Y. Kuge, K. Hikosaka, K.-i. Seki, K. Ohkura, K.-i. Nishijima, T. Kaji, S. Ueno, E. Tsukamoto, and N. Tamaki Characteristic Brain Distribution of 1-14C-Octanoate in a Rat Model of Focal Cerebral Ischemia in Comparison with Those of 123I-IMP and 123I-Iomazenil J. Nucl. Med., July 1, 2003; 44(7): 1168 - 1175. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. V. BORLONGAN, M. YAMAMOTO, N. TAKEI, M. KUMAZAKI, C. UNGSUPARKORN, H. HIDA, P. R. SANBERG, and H. NISHINO Glial cell survival is enhanced during melatonin-induced neuroprotection against cerebral ischemia FASEB J, July 1, 2000; 14(10): 1307 - 1317. [Abstract] [Full Text]
|
|
|
|
 |
|
 M. Fujioka, T. Taoka, K.-I. Hiramatsu, S. Sakaguchi, and T. Sakaki Delayed Ischemic Hyperintensity on T1-Weighted MRI in the Caudoputamen and Cerebral Cortex of Humans After Spectacular Shrinking Deficit Stroke, May 1, 1999; 30(5): 1038 - 1042. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||