Keywords
blast injury - traumatic brain injury - acetylcholinesterase - rat model - tau - visuospatial
memory
Introduction
Blasts and blast waves are a common cause of traumatic brain injury in the military,
both during training as well as during wartime. Civilians may also get exposed to
blasts during terrorist attacks and domestic or industrial accidents.
The pathophysiology, outcomes, and management options for blast-induced neurotrauma
(BINT) differs from the more commonly seen impact neurotrauma.[1] The outcomes of blast injury depend on the severity of injury with moderate and
severe injuries having fatal or highly morbid outcomes.[2] On the other hand, mild blast injury has a more indolent course without any visible
features on imaging, but with the risk of long-term cognitive sequelae. Multiple exposures
produce a syndrome quite similar to chronic traumatic encephalopathy[3] and usually occur in military personnel, while single exposure blasts are more common
in the civilian population.
This study utilizes a rat model of BINT using a novel, cost-effective method to study
the cognitive effects of mild blast injuries at different pressures. The apparatus
used to generate the blast shockwave is a novel tube of Indian origin known as the
modified Reddy's tube which has been validated by the investigators in an earlier
study.[4]
This study considers mild BINT as a unique entity with a distinct pathology. It has
some features of diffuse axonal injury and is associated with electrophysiological
changes, genetic changes, and systemic changes of autonomic, endocrine, and immunological
dysfunction.
Using this model, we have studied the effects that mild BINT has on the retrograde
spatial memory of a rat. We have also evaluated the hippocampus for change in tau
protein content and acetylcholine esterase (AChE) activity, and the structural damage
caused by the blast shockwaves as seen on light microscopy.
Materials and Methods
Forty 2-month-old male Sprague-Dawley rats weighing 200 to 250 g were procured from
the Central Animal Research Facility, National Institute of Mental Health and Neuro-Sciences
(NIMHANS), Bengaluru, Karnataka, India after ethical clearance from the Institutional
Animal Ethics Committee. The rats were housed in 12 hours light and 12 hours dark
cycles with access to standard rat chow and water ad libitum. They were randomly divided
into 4 groups of 10 rats each. Then, the rats were placed in 8 cages, with 5 rats
in each cage. After the division, they were labeled and marked with numbers 1 to 5.
Following placement into the 8 cages, the entire cage of rats was exposed to the same
pressure of blast shockwave. The pressure subjected to each cage was separately recorded
on a key Excel sheet that was unavailable to the research team until the analysis
was completed. On day 28 postblast exposure, half the rats were sacrificed. There
were two rats randomly chosen by their marked numbers from each cage first followed
by one rat from the cages labeled 1 to 4. The remaining rats were sacrificed on postblast
exposure day 84. Samples collected were labeled by the cage and rat number. Statistical
analysis of data generated was done using cage numbers. After the entire analysis
was completed, the key and cage numbers were tallied to determine which rat belonged
to which group and then the results were interpreted.
Barnes Maze Trials
All four groups were subjected to trials on a Barnes maze. The 40 rats underwent acquisition
training with three trials per day of the Barnes maze for four consecutive days. In
this protocol, the rat was placed in the center of the maze and covered with a black
cylinder. The cylinder was suddenly removed, while at the same time, an overhanging
lamp was switched on. The rats rushed away from the light source and explored the
holes in the Barnes maze to escape. Each look into a wrong hole was counted as an
error. The entry in the correct hole and consequently the chamber was considered as completion. The time taken from the removal of the covering cylinder to the completion was recorded
along with the number of errors made by the rat. After acquisition training, on the
fifth day, 30 of the rats were exposed to blast shockwaves from the modified Reddy's
tube. The rats were allowed to recover and 2 days after the blast or sham exposure
the rats were reevaluated with three trials of the Barnes maze task. They were again
evaluated on day 10 and day 21 postblast exposure.
Blast Exposure
Three groups of 10 rats each were randomly divided and exposed to three different
blast pressures on the 5th day after procurement. The exposure blast pressures were
81, 160, and 210 kPa as noted at 1 cm from opening of the driven section, measured
using a piezoelectric sensor.
The rats were exposed in the prone position with the fore- and hind-limbs restrained
using a cellophane sheet, such that only the head of the rat was exposed. The neck
was restrained with the help of a rubber collar and metal prongs which held the collar
in place. The snout of the rat was at 1 cm from the opening of the modified Reddy's
tube. After exposure to the blast, the rat was freed and observed for any external
injuries. All the rats were stunned for 5 to 15 minutes after exposure to the blast
shockwave.
The remaining 10 rats were used as controls, and they underwent sham blasts. These rats were restrained in a similar manner on the 5th day after the commencement
of acquisition and exposed to the sound of the blast without being exposed to the
shockwave ([Figs. 1] and [2]).
Fig. 1 Rat restrained in front of the open end of the modified Reddy's tube.
Fig. 2 Graphic representation of the blast wave showing the Friedlander curve with the overpressure
and underpressure together lasting 2.5 ms for a peak pressure of 210 kPa.[4]
Sacrifice
The experimental animals (those exposed to blast waves) were equally and randomly
divided into two groups which were sacrificed on day 28 and day 84 postblast exposure.
The rats were sacrificed by manual cervical dislocation, following which the skull
caps were dissected out and the brain was removed. The cerebral hemispheres were placed
on an ice-cold surface and divided in the mid sagittal plane using a scalpel blade.
One half of the brain was preserved in 10% neutral-buffered formalin. From the other
hemisphere, the hippocampus was dissected out and placed in an Eppendorf tube and
frozen at –20°C. The remaining portion of the brain was preserved in another container
at –20°C. The hippocampi were homogenized in 50 mmol tris-buffered saline, pH 7.4
containing 0.5% TritonX-100 with pestle and sonicated with 3 pulses of 10 seconds
each, and centrifuged at 10,000 revolutions per minute for 15 minutes at 4°C. The
supernatants were collected and subjected to tau enzyme-linked immunosorbent assay
(ELISA) and acetylcholinesterase assay.
The other half of the brain was fixed in 10% neutral-buffered formalin for a minimum
period of 4 weeks prior to processing. The tissue was then processed for paraffin
embedding. The brainstem and cerebellum were disconnected at the level of crus cerebri
and embedded in the mid-sagittal plane. The cerebral hemisphere was sliced at the
level of the olfactory bulb, infundibulum, and mammillary bodies.
Five-micron thick serial sections were taken and stained with hematoxylin and eosin
(H&E), Luxol fast blue (LFB) for myelin, and Cresyl violet fast stain for neurons.
Each case was evaluated for alterations in neurons (neuronal shrinkage and dark neurons
on H&E, Cresyl violet stains: reflecting ischemic changes), perivascular inflammation,
microglial nodules, and ventricular dilatation. All changes were semiquantitatively
graded as mild, moderate, and severe.
Rat Tau Protein ELISA
Total tau in the hippocampal homogenate was estimated using commercially available
rat tau quantitative ELISA kit (My Bio Source, United States). The assay sample (after
being appropriately diluted) was incubated with the buffer and tau-horseradish peroxidase
(HRP) conjugate in a precoated plate for 1 hour at 37°C. After the incubation the
wells were decanted and washed five times. The wells were then incubated with a substrate
for HRP enzyme. The product of the enzyme substrate complex was blue. The substrate
used was 3,3′, 5,5′ tetramethylbenzidine. A stop solution was added which arrests
the reaction and turns the solution yellow. The intensity of the color was measured
spectrophotometrically at 450 nm in a microplate reader and was inversely proportional
to the tau concentration, as both the tau from the sample and the tau-HRP conjugate
compete for binding sites on the anti-tau antibody immobilized in the well. A standard
curve was plotted between the optical density and concentration of standards from
which the concentration of tau was extrapolated.
Acetylcholine esterase Assay
The principle of this assay is based on the measurement of the activity of the enzyme
AChE on the substrate acetylthiocholine iodide (ACTI). The enzyme cleaves the substrate
into thiocholine and acetate. The thiocholine reacts with a coloring agent (in this
case dithiobisnitrobenzoate) to form a yellow-colored product. The esterase activity
in the homogenized hippocampus of the rat was measured by following the earlier described
procedure. The buffer used was 0.1 M phosphate buffer at a pH of 8.0. The standard
value for different concentrations of AChE from electric eel (Sigma Aldrich, Bengaluru,
Karnataka, India) in hydrolyzing ACTI (Sigma Aldrich) was recorded as optical density
at 412 nm. The readings were then plotted on a graph. Amount of enzyme activity in
the hippocampal homogenates was extrapolated from the standard curve.
Statistical Analysis
The completion time and errors were analyzed using a nonparametric analog to two-way
repeated measures analysis of variance using a package called nparLD. It tested the
effects of groups (sham, low blast pressure, medium blast pressure, and high blast
pressure) and the effects of time (postinjury day 3, postinjury day 7, postinjury
day 21) on the outcome variables (time to completion and errors). The Kruskal–Wallis
test was used to compare the tau protein levels and acetylcholinesterase activity
levels at various time points ([Fig. 3]).
Fig. 3 Flowchart depicting the experimental study. BOP, blast overpressure; PBI, postblast
injury.
Results
All 40 rats underwent acquisition and recall using Barnes maze trials on day 7, day
14, and day 21. A rat each from the sham exposure group, low pressure blast group,
and high pressure blast group were excluded due to fall from the Barnes maze. Three
rats in the sham group were sacrificed on day 28 and three rats on day 84 postinjury.
In the low blast pressure group, four rats were sacrificed on day 28 and five rats
on day 84 postinjury. Seven rats were sacrificed on day 28 and three rats on day 84
postinjury in the medium blast pressure group. Six rats were sacrificed on day 28
postinjury and three rats on day 84 postinjury in the high blast pressure group.
The rats and samples were labeled randomly after blast exposure and a key was generated
to record the assignment. The results were calculated and interpreted while blinded
to the key. The final analysis was then done after decoding from the key.
Retrograde Memory Assessment on Barnes Maze
The number of pokes into the wrong hole and the time taken to reach the correct hole
after being exposed to the bright light were both evaluated and compared between the
sham exposure animals and blast shockwave exposed animals.
The time taken prior to blast exposure and following blast exposure was compared between
different groups. The results are shown in [Tables 1] and [2].
Table 1
Comparison of mean time (seconds) to completion in different groups at various time
points
Pressure group
|
Preblast time
|
Postblast 3rd day recall
|
Postblast 7th day recall
|
Postblast 21st day recall
|
p-Value
|
Sham
|
40.44 (11–102)
|
34.778 (10–67)
|
34.22 (10–65)
|
33.67 (12–55)
|
|
Low pressure group (81 kPa BOP)
|
26.13 (9.33–58.67)
|
24.9 (10–52.33)
|
29.07 (19.67–38.33)
|
22.43 (8.33–55.67)
|
0.53
|
Medium pressure group (160 kPa BOP)
|
41.33 (16.67–87)
|
44.93 (14.67–117)
|
41.99 (16–97)
|
38.1 (12.33–92)
|
0.766
|
High pressure group (210 kPa BOP)
|
54.963 (21.67–74.33)
|
27.702 (14.67–47)
|
35.89 (18.67–56.33)
|
35.89 (24.67–48)
|
0.315
|
Abbreviation: BOP, blast overpressure.
Table 2
Comparison of mean errors in different groups at various time points
Pressure group
|
Preblast errors
|
Postblast 3rd day errors
|
Postblast 7th day errors
|
Postblast 21st day errors
|
p-Value
|
Sham
|
3.111 (1–5.67)
|
2.67 (1–5)
|
3.67 (1–8)
|
3.62 (0.3–6.33)
|
|
Low pressure group (81 kPa BOP)
|
2.03 (0.67–4.33)
|
2.43 (1–7.43)
|
3.79 (1.33–8)
|
2.9 (0.67–7)
|
0.598
|
Medium pressure group (160 kPa BOP)
|
4.13 (1.67–8.67)
|
5.46 (1–15.33)
|
4.73 (1–13.3)
|
5.33 (1.33–15)
|
0.43
|
High pressure group (210 kPa BOP)
|
4.48 (0.67–9)
|
4.44 (1.33–11.33)
|
5.073 (2–10)
|
4.41 (2.33–7.33)
|
0.97
|
Abbreviation: BOP, blast overpressure.
The results show that the completion time had an improvement in performance on postblast
exposure day 3 with shorter times taken to complete the maze in the low and high blast
pressure groups. Following this the subsequent time taken to complete the Barnes maze
trial continued to decrease with an improvement in performance in the low blast pressure
blast groups. The rats exposed to high pressure blast wave have a significantly better
performance as compared to the other rats on postblast injury day 3. But after the
initial improvement seen on the 3rd postexposure day (likely some error as there is
no possible explanation for this phenomenon, it is likely due to increased familiarity
with the maze after continued learning or increased vigilance after exposure to trauma),
performance was static over time (performance declined from the very low value of
postblast 3rd day test, but was still better than the preblast value), this likely
indicates impaired retention over time.
The change in time taken to completion is statistically significant among the rats
exposed to high blast pressures when compared with rats exposed to sham blasts or
low blast pressures.
The errors made by the rats did not show any significant differences when comparing
the blast-exposed groups with the control groups ([Figs. 4]
[5]
[6]) ([Table 3]).
Fig. 4 Comparison of mean time to completion in different groups at various time points.
Fig. 5 Comparison of mean errors in different groups at various time points.
Fig. 6 Comparison of total tau protein levels (pg/100 mg tissue) in different groups at
various time points.
Table 3
Comparison of total tau protein levels (pg/100 mg tissue) in different groups at various
time points
Pressure group
|
Day 28
|
Day 84
|
Sham
|
2050 (1250–3400) [n = 3]
|
2150 (1400–2900) [n = 3]
|
Low pressure group (81 kPa BOP)
|
880 (550–1500) [n = 4]
|
2240 (1300–3450) [n = 5]
|
Medium pressure group (160 kPa BOP)
|
536.5 (420–700) [n = 7]
|
2133.3 (550–3300) [n = 3]
|
High pressure group (210 kPa BOP)
|
754.2 (830–1400) [n = 6]
|
2100 (1650–3000) [n = 3]
|
Abbreviation: BOP, blast overpressure.
The hippocampal homogenates of the blast-exposed rats show a transient reduction in
the levels of total tau protein on postblast exposure day 28 compared to the controls.
Following this, the total tau protein levels on postexposure day 84 were similar in
all the groups. This could reflect a repair phenomenon leading to a rise in total
tau levels by day 84. Alternatively, the transient reduction at day 28 postinjury
could be due to a wash out of tau from the brain into the circulation[5] ([Table 4]).
Table 4
Comparison of acetylcholinesterase activity (U/mg tissue) in different groups at various
time points
Pressure group
|
Day 28
|
Day 84
|
Sham
|
1.408 [n = 3]
|
1.526 [n = 3]
|
Low pressure group (81 kPa BOP)
|
1.372 [n = 4]
|
1.455 [n = 5]
|
Medium pressure group (160 kPa BOP)
|
1.754 [n = 7]
|
1.365 [n = 3]
|
High pressure group (210 kPa BOP)
|
1.892 [n = 6]
|
1.445 [n = 3]
|
Abbreviation: BOP, blast overpressure.
The acetylcholinesterase activity did not show any statistically significant change
at the different blast pressure exposed groups and the control groups at both postexposure
day 28 and post exposure day 84.
Histopathology
The rats exposed to different blast pressures and sham blasts were sacrificed at two
time points—4 and 12 weeks postinjury. Changes in gray and white matter in all neuroanatomical
areas were assessed. The white matter showed a greater degree of involvement than
the gray matter. The brain sections at 4 weeks postinjury showed evidence of variable
edema and vacuolation in white fiber tracts predominantly involving the cerebellar
white matter and corpus callosum with demyelination highlighted by LFB staining. In
contrast, the brainstem white matter, internal capsule, anterior commissure, optic
tract, and crus cerebri were spared. The rats sacrificed at 12 weeks postblast exposure
demonstrated more prominent demyelination on LFB in cerebellar white matter and corpus
callosum. In addition, there was focal gliosis seen in the lower portions of the corpus
callosum in midline, at the point of insertion of the septum. Distortion of white
matter fibrillar pattern with interspersed oligodendroglial cells was seen in the
corpus callosum, external capsule, internal capsule, and crus cerebri. The control
group also showed white matter vacuolation and edema suggesting that these changes
are likely secondary to hypoxia as a terminal event. Gray matter changes were limited
to the cingulate/frontal cortex and hippocampus with conspicuous sparing of caudate,
putamen, thalamic nuclei, and brainstem nuclei. The hippocampus showed the most characteristic
damage pattern that correlated well with the blast pressures. The dentate gyrus, CA1,
and CA3 demonstrated dark neurons, with pyknotic nuclei, and intensely stained cytoplasm.
This change was most noticeable in the basal layers and the angle of the dentate gyrus
and prominent at 4 weeks postblast injury. The rats sacrificed at 12 weeks postblast
injury showed recovery from injury, with normal neuronal cells. However, there was
evidence of scarring and distortion of architecture. The basal ganglia, thalamus.
and sensorimotor cortex showed minimal damage which did not seem to increase with
increasing pressures. The sensorimotor and pyriform cortices showed mainly focal neuronal
damage in middle and deeper layer foci of damage in the blast-exposed rats when compared
with the control rats.
In the cerebellum, the Purkinje cells revealed varying degrees of hypoxic/ischemic
change that did not correlate with the blast pressure exposure. The control animals
showed similar changes, suggesting that these changes are likely secondary to hypoxic
injury occurring as a terminal event ([Figs. 4]
[5]
[6]
[7]
[8]).
Fig. 7 (A–C) Hippocampus from rats sacrificed at day 28 postinjury (Nissl's stain). (A) Control hippocampus showing dentate gyrus, and Ammon's horn with CA3, part of CA2,
and CA1. (B) Few dark, hyperchromatic, pyknotic neurons in basal layers of dentate gyrus in rats
exposed to 210 kPa blast pressure (arrows). (C) Dark, hyperchromatic and pyknotic neurons in CA3 interspersed between normal neurons
in rats exposed to 210 kPa blast pressure (arrows). Nissl stain; magnification = scale
bar.
Fig. 8 (A–D) White matter changes (Luxol fast blue stain). (A, B) Controls showing well preserved myelinated fiber tracts in whole mount section of
brain at level of mammillary body with closeup view showing compact myelinated tracts
(B). (IC, internal capsule; EC, external capsule; F, fornix; T, thalamus; H, hippocampus;
Hy, hypothalamus). (C) Disarray of fiber tracts, with patchy demyelination and scarring at junction of
internal and external capsule day 84 postinjury in rats exposed to medium blast pressure
of 160 kPa. (D) Severe demyelination in IC 28 days postinjury in rats exposed to high blast pressure
of 210 kPa. (magnification = scale bar).
Discussion
In this study, a rodent model of blast injury was studied, and the cognitive, behavioral,
biochemical, and pathological changes were evaluated following blast pressures using
an indigenously developed modified Reddy's tube which can deliver predetermined blast
pressures in a controlled manner.
A small laboratory-based simulation for BINT is possible due to the availability of
reliable and portable shockwave tubes. Although these tubes mimic the effects of primary
blast wave well, their utility in replicating real-life situations of blast explosion-induced
trauma is limited. However, for the study of the pathobiology and physics of the blast
exposure, these studies are adequate, especially due to minute control that can be
exerted over the physical attributes, the environment, and the nature of the subject
as well.
The blast pressure was chosen based on previous pilot study done by one of the senior
authors. The blast pressures chosen were the pressures at which blast/pressure-stratified
white matter and gray matter injury was seen.[4]
The rat is a commonly used laboratory model and a sturdy animal; with earlier studies
establishing its analogous and homologous anatomical structures. Due to its small
size, maintenance is easy, and the environment can be carefully controlled to avoid
confounders. The parameters of the experiment can be easily varied with controlled
conditions. For this study, Sprague-Dawley rats were chosen.
Male rats aged 2 months were used as this age group corresponds with young adults
in humans, and most of the central nervous system development is complete at this
age.
The rats were exposed to the blast pressure at a distance of 1 cm from the driven
end of the shockwave tube while the rat was in a prone position. The prone position
of the rat lets the injury be limited to the brain and cranial structures due to the
blast wave pressure decaying before it reaches the thorax. The effects of primary
blast injury on the brain alone without the effects of systemic exposure can thus
be studied.
Human exposure to blasts leads to the chest or the back bearing the main brunt of
the blast shockwave.
The Barnes maze is a tool with diverse applications, and it is a physiological and
sensitive test of visuospatial memory and learning. The results noted in the current
study corroborate with those seen in other studies that show that there is no significant
change in retrograde recall of previous path and visual cues, resulting in similar
completion time across different blast pressures and control animals. The initial
improvement with a faster response and completion of the Barnes maze test seen in
rats exposed to 210 kPa shockwave blast overpressure could be a consequence of increased vigilance following blast exposure as other studies
show normalization of retrograde memory function after the initial 24 hours.[6]
The rats exposed to higher blast pressures showed a deterioration of completion time
at later recalls which was significant when compared to animals exposed to low blast
pressures and sham control animals which do indicate perhaps an impairment in anterograde
learning as seen in multiple studies.[7] It may also indicate ongoing secondary effects of the primary blast injury. This
is supported by the demonstration of parenchymal injury mainly seen in the hippocampus.
The dentate gyrus shows maximal effect followed by CA3 and CA1, while CA2 is spared.
The hippocampal injury is maximal in acute phase but persistent at 4 weeks, in the
high blast pressure-exposed rats; however, these did not have significant effects
on the spatial memory as demonstrated by relatively similar times to completion among
the different groups on recall day 21.
Apart from evidence of neuronal death, the white fibers show a disorganized pattern
in multiple areas which may contribute to the cognitive dysfunction and delays in
impulse transmission.
Although there appeared to be no evidence of gross motor deficits, there was evidence
of demyelination and misalignment of white matter fibers of the internal capsule and
crus cerebri. The motor and sensory areas showed relative preservation of neurons
with no evidence of overt neurodegenerative changes.
There was no evidence of shear injury along blood vessels even on day 84 postinjury
as seen in other studies.[7] The acetylcholinesterase activity in the hippocampal homogenates was found to be
similar on day 28 postinjury and day 84 postinjury indicating either the changes occurred
before the first sacrifice, or there was no significant change in activity in the
hippocampus after exposure to blast shockwave. The blast shockwave overpressure has
been found in other studies to cause injury mainly in frontal and basifrontal areas
which gradually improves before 4 weeks. The current study indicates that the hippocampus
acetylcholinesterase activity remains normal even at 12 weeks postblast. Acetylcholinesterase
activity is seen to be decreased in several other areas, while the hippocampus was
spared.[6]
[8]
The levels of total tau protein appear to transiently decrease in blast-exposed rats
at day 28 postinjury when compared to control animals. The fall in total tau may indicate
structural damage to cells with elution of tau protein into cerebrospinal fluid and
venous system. The fall in total tau protein levels may also be due to decreased binding
of the tau ELISA antibody due to posttranslational modification of the tau proteins
because of the blast exposure.[7] The delayed sacrifice at 84 days postblast injury shows normalization of levels,
comparable between blast-exposed and control animals likely reflecting reparative
response.
These minor changes seen in the cellular and fiber architecture on histopathology
without any major gross demyelination or severe neuronal injury may be a cause of
the delayed cognitive deficits seen in humans. The drastic fall in tau protein levels
may lead to ultrastructural reorganization at a later point of time which may be a
cause of impaired function due to aberrant recovery over time.
There are few studies that assess the structural or biochemical status of the rat
brain after the first month postinjury.[9] The delayed sacrifice at 84 days postinjury reveals evidence of recovery with no
persistent deficits.
Summary
A single blast exposure did not significantly affect the visuospatial memory and its
recall. Although functionally, cognitive deficits were not prominent, there was histopathological
and biochemical evidence of decreased tau protein levels that reflected ongoing neuronal
damage ([Figs. 7]
[8]
[9]
[10]
[11]).
Fig. 9 (A–D) Neuronal morphology as seen in rats exposed to 210 kPa blast overpressure (Nissl
stain). (A) Well-preserved neurons in laminar arrangement in sensorimotor cortex. (B) Several dark, pyknotic neurons in superficial and middle layers of the cingulate
cortex, day 28 postinjury. (C) Neurons in thalamic nuclei well preserved. (D) Neurons in caudate nuclei well preserved. (magnification = scale bar).
Fig. 10 (A–C) White matter changes (Luxol fast blue stain). (A) Control—Whole mount section at level of caudate-putamen (CaP) shows well-preserved
myelin in corpus callosum (CC). (B) Closeup view of corpus callosum showing compact myelinated fiber tracts. (C) Focal demyelination and scarring of corpus callosum – day 28 postinjury. (CC, corpus
callosum; CaP, caudate-putamen; ON, optic nerve). (magnification = scale bar).
Fig. 11 Changes in cerebellum. (A, B) Cerebellar folia showing preserved myelination in white matter (*) on Luxol fast
blue stain (A) with normal Purkinje neurons (B) highlighted by Nissl stain. (C, D) 28 days postinjury showing focal demyelination in white matter (C, *) and several Purkinje neurons are shrunken, dark, and pyknotic (D, arrows) on Nissl stain in rats exposed to high blast pressures of 210 kPa. (magnification = scale
bar).