| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Dipartimento di Fisiologia Generale ed Ambientale, Università degli Studi di Bari, I-70126 Bari, Italy;
Istituto di Anatomia Umana Normale, Policlinico, I-70124 Bari, Italy; and
Zentrum Anatomie, Abteilung Histologie, Universitaet Goettingen, D-37075 Goettingen, Germany
1Correspondence: Dipartimento di Fisiologia Generale ed Ambientale, Università degli Studi di Bari, via Amendola 165/A, I-70126 Bari, Italy. E-mail: a.frigeri{at}biologia.uniba.it
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSImmunocytochemistryAQP4 immunoprecipitation from...Preparation of brain membrane...Western blot analysisQuantitative reverse...RESULTSDISCUSSIONREFERENCES |
|---|
85% in gold
particles (32±2/µm vs. 4.7±0.61/µm). Western blot experiments
conducted using membrane fractions from brain cortex revealed a strong
reduction (of 70%) in AQP4 protein in adult mdx mice, and RT-PCR
experiments demonstrated that the reduction was not at transcription
level. More interesting was the finding that AQP4 reduction was
associated with swelling of astrocytic perivascular processes whose
ultrastructural modifications are commonly indicated as an important
and early event in the development of brain edema. No apparent
reduction in AQP4 was found in mdx stomach and kidney. Our data provide
evidence that dystrophin deficiency in mdx mice leads to disturbances
in AQP4 assembly in the plasma membrane of fast skeletal muscle fibers
and brain astrocytic end-feet, suggesting that changes in the osmotic
equilibrium of the neuromuscular apparatus may be involved in the
pathology of muscular dystrophy.—Frigeri, A., Nicchia, G. P.,
Nico, B., Quondamatteo, F., Herken, R., Roncali, L., Svelto, M.
Aquaporin-4 deficiency in skeletal muscle and brain of dystrophic mdx
mice.
Key Words: muscular dystrophy • aquaporins • orthogonal arrays of particles • AQP4 • mdx mice • water channels
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSImmunocytochemistryAQP4 immunoprecipitation from...Preparation of brain membrane...Western blot analysisQuantitative reverse...RESULTSDISCUSSIONREFERENCES |
|---|
. The product of the gene in the
muscle is dystrophin, a 427 kDa protein consisting of four structural
domains: an amino-terminal actin binding domain, a central rod region,
a cysteine-rich domain, and a carboxyl-terminal domain
(2)
. In normal muscle fibers, dystrophin is associated
with the sarcolemma via a membrane-spanning glycoprotein complex of
dystrophin-associated proteins (DAPs; 3
). The
sarcolemma-spanning complex is proposed to link the extracellular
matrix components to the membrane cytoskeleton, thereby stabilizing the
fiber periphery during muscle contraction (4)
. Mutations
in any of the proteins within this complex result in muscle fiber
instability, leading to muscular dystrophy (5)
. The DMD
gene also codes for two nonmuscle isoforms of dystrophin: the
brain-type dystrophin and the Purkinje cell-type dystrophin transcribed
by different promoters (6)
. Moreover, internal promoters
in the huge DMD gene regulate the expression of four smaller proteins
differing in their amino-terminal domains (7)
.
Much of the information about the precise mechanism by which the
absence of dystrophin leads to muscle cell necrosis comes from the
dystrophin-deficient mdx mouse. This animal model has a genetic defect
in the homologous region to the human DMD gene (8)
and
similarly lacks the dystrophin protein, but manifests a less severe
phenotype.
In both DMD patients and mdx mice, a clear morphological feature of
dystrophic muscle fibers is the drastic reduction in orthogonal arrays
of particles (OAPs; 9
, 10
), now shown to be the
morphological equivalent of the AQP4 water channel (11
, 12)
.
In a recent study, we demonstrated the functional expression of a water
channel (i.e., AQP4) in the sarcolemma of fast-twitch fibers of
skeletal muscle for the first time (13)
. Moreover,
preliminary results showed a drastic reduction in AQP4
immunolocalization in mdx mouse muscle plasma membrane
(13)
, indicating a possible involvement of AQP4 in the
biochemical alteration of muscle fibers in DMD. It is not known which
skeletal muscle fibers are affected, nor whether the drastic reduction
in AQP4 water channels is due to a decrease in AQP4 synthesis or to
disturbances in AQP4 assembly into the muscle plasma membrane.
The main site of AQP4 expression is, however, the central nervous
system. In particular, AQP4 has been localized in astrocytic end-feet,
as well as in ependymal cells where OAPs have been found (14
, 15)
. It has been proposed that AQP4 plays an important role in
controlling water movement between blood and brain, thus maintaining
the osmotic balance of the brain (16)
.
Furthermore, in mdx brain, biochemical analysis has shown an increased
extracellular and decreased intracellular brain volume
(17)
, but no findings on OAP alterations (i.e., changes to
the AQP4 water channel) have been reported. Therefore, in the present
study we investigated whether the alteration in intracellular brain
volume reported in mdx mice may be due to an alteration in AQP4
expression and/or assembly.
Our results indicate that an important reduction in AQP4 protein content is found in the neuromuscular system of the dystrophic animal. This reduction, at least in the brain, is clearly associated with an altered water balance, as indicated by the presence of swollen glial processes.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSImmunocytochemistryAQP4 immunoprecipitation from...Preparation of brain membrane...Western blot analysisQuantitative reverse...RESULTSDISCUSSIONREFERENCES |
|---|
|
Immunocytochemistry |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSImmunocytochemistryAQP4 immunoprecipitation from...Preparation of brain membrane...Western blot analysisQuantitative reverse...RESULTSDISCUSSIONREFERENCES |
|---|
and with IIA or IIB
(18)
or I type (Sigma, St. Louis, Mo.) myosin heavy chain
(MHC) mouse monoclonal antibodies for 1 h at room temperature.
After washings, sections were incubated for 1 h with CY3-coupled
goat anti-rabbit and FITC-coupled goat anti-mouse antibodies. Sections
were examined with a Leica DMRXA photomicroscope equipped for
epifluorescence, and digital images were obtained with a cooled CCD
camera (Princeton Instruments, Princeton, N.J.). The specificity of the
AQP4 antibody used in this study was previously reported (13
, 14)
.
Immunoperoxidase
Brains were dehydrated in an ascending ethanol series and
embedded in paraffin. Five micrometer sagittal sections were collected
on polylysine-coated slides and deparaffinized. After blocking, the
sections were sequentially incubated with 1) primary rabbit
affinity-purified anti-AQP4 antibody (0.3–0.5 µg/ml) diluted in TBS
overnight at 4°C; 2) secondary antibody, goat anti-rabbit
(Dakopats, Hamburg, Germany) diluted 1:50 in TBS for 30 min at room
temperature, followed by the PAP complex, using commercial reagents
(Dakopats). Finally, the sections were treated with 0.06% 3,3'
diaminobenzidine in Tris-HCl buffer in the presence of
H2O2 and counterstained
with Mayer’s hematoxylin for 1 min. Control experiments performed
using immunodepleted antibodies showed no staining.
Immunogold electron microscopy
Small pieces of brain cortex were fixed in 1% glutaraldehyde
and embedded in the acrylic resin LR-Gold (Bio-Rad, Richmond, Calif.)
with 0.8% benzil. Thin sections (60 nm) cut with an LKB V
ultramicrotome were incubated for 10 min at room temperature with TBS
buffer; unspecific binding sites were blocked with 1% bovine serum
albumin-TBS, pH 7.4, for 10 min at room temperature. The sections were
incubated with the primary antibody (affinity-purified anti-AQP4,
0.3–0.5 µg/ml) at room temperature overnight, washed with TBS, and
incubated for 1 h at room temperature with the second antibody
(goat anti-rabbit) coupled with 6 nm gold particles (Chemicon Intern.
Inc., S.I.C., Rome, Italy). After washing with TBS, the grids were
stained with 1% uranyl acetate, followed by 1% lead citrate, and
examined with a Zeiss EM 109 electron microscope (Zeiss, Oberkochen,
Germany).
Quantification of AQP4 gold particle distribution
To quantify AQP4 gold particle distribution on the glial
end-feet facing the microvessels, 198 electron micrographs at the final
magnification of x30,000 were chosen for control and for mdx brain
mouse frontal cortex. The number of particles distributed along the
total length of the abluminal microvessels side (basement membrane and
glial end-feet) was counted with the use of an electronic pen connected
to a graphic tablet (Digicad Plus Kontron-Elektronic, GMBH, Germany)
and to a VIDAS 2.5 computerized image analyzer (Kontron Elektronic).
One hundred thirty-five vessel profiles were analyzed for four mdx mice
and 65 for three control mice. The results were expressed as the number
of gold particles per micrometer of microvessels abluminal front. The
mean value in each case, the final mean value for all controls and for
the mdx mouse brains, and the standard deviation were calculated for
each variable. The statistical significance of the difference between
the mean values of the number of gold particles and between the luminal
and abluminal surfaces of the endothelial cells was determined by the
Student’s t test for unpaired data.
|
AQP4 immunoprecipitation from mouse muscle |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSImmunocytochemistryAQP4 immunoprecipitation from...Preparation of brain membrane...Western blot analysisQuantitative reverse...RESULTSDISCUSSIONREFERENCES |
|---|
|
Preparation of brain membrane fraction |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSImmunocytochemistryAQP4 immunoprecipitation from...Preparation of brain membrane...Western blot analysisQuantitative reverse...RESULTSDISCUSSIONREFERENCES |
|---|
|
Western blot analysis |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSImmunocytochemistryAQP4 immunoprecipitation from...Preparation of brain membrane...Western blot analysisQuantitative reverse...RESULTSDISCUSSIONREFERENCES |
|---|
. AQP4 protein was revealed by enhanced
chemiluminescence (ECL+plus, Amersham, Little Chalfont, U.K.).
Coomassie-stained gels and ECL film were scanned using a UMAX SPEED IIC
scanner and Adobe Photoshop software. Densitometric analysis was
performed using Scion Image software, and AQP4 protein levels in mdx
mice were expressed as fractions of the respective levels in control
mice. The statistical analysis was conducted by the Student’s
t test for unpaired data. |
Quantitative reverse transcriptase-polymerase chain reaction (RT-PCR) |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSImmunocytochemistryAQP4 immunoprecipitation from...Preparation of brain membrane...Western blot analysisQuantitative reverse...RESULTSDISCUSSIONREFERENCES |
|---|
, 20)
. Relative amounts of AQP4 transcript were estimated by
direct comparison between multiple samples after standardization with
coamplification of 18S rRNA (QuantumRNA kit, Ambion, Austin, Tex.).
Densitometric analysis was performed as reported for the Western blot
experiments. The statistical analysis was conducted by the Student’s
t test for unpaired data. |
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSImmunocytochemistryAQP4 immunoprecipitation from...Preparation of brain membrane...Western blot analysisQuantitative reverse...RESULTSDISCUSSIONREFERENCES |
|---|
. To determine
precisely which fibers are affected, we first investigated the
expression of AQP4 in the skeletal muscle of control mice using
affinity-purified AQP4 antibody together with specific antibodies
against slow (type I) and fast (type IIA, IIB) MHC isoforms.
Figure 1
shows double immunostaining experiments in soleus (slow muscle) and
tibialis anterior (fast muscle). In soleus, type IIA fibers were
stained (Fig. 1A
) whereas fibers that contained type I MHC
were negative for AQP4 immunostaining (Fig. 1B
). Using
tibialis anterior muscle, we found that both type IIB (Fig. 1D
) and type IIA fibers (Fig. 1C
) expressed the
AQP4 water channel. These results show that the expression of AQP4 in
mouse skeletal muscle is restricted to fast contracting fibers and
seems related to the glycolitic metabolism of the fiber.
|
Analysis of AQP4 in the tibialis anterior of the mdx mouse
AQP4 sarcolemma staining in the tibialis anterior of mdx mice is
progressively reduced. In young mdx mice (1 month old), type IIB fibers
were more affected than type IIA fibers. Although numerous type IIA
fibers appeared to have quite a normal sarcolemmal AQP4 staining
(Fig. 2A
), the majority of type IIB fibers underwent a progressive
reduction in staining (Fig. 2B
). Our results show that a
subset of fast fibers, the type IIB fibers, are the first to manifest a
reduction in AQP4 staining. In adult mdx mice (1 year old), the
reduction in AQP4 staining was more general. Both type IIA (Fig. 2C
) and type IIB (Fig. 2D
) fibers manifest a
drastic reduction in AQP4 staining. In many areas of tibialis anterior,
only a weak fluorescence signal was detected on some fibers.
Immunofluorescence analysis indicates that the number of AQP4-positive
fibers in 1-month-old mdx mice represent 70% (P<0.01,
n=5) of the AQP4-expressing fibers in control mice whereas
only 8% (P<0.001, n=5) of 1-year-old mdx mice
show a weak immunofluorescence labeling. No labeling was observed in
the cytoplasm of muscle fibers.
|
We next used immunoprecipitation experiments to examine whether the
reduction in sarcolemma staining was caused by an improper assembly of
AQP4 protein in the membrane and/or a reduction in protein content.
Immunoprecipitated proteins were analyzed by immunoblot. Compared to
age-matched control mice, AQP4 protein is reduced by 20%
(P<0.01, n=3) and 90% (P<0.001,
n=4) in 1-month- and 1-year-old mdx mice, respectively
(Fig. 3B
). Coomassie blue-stained gels showed that the overall
protein band pattern is relatively comparable between normal mouse and
mdx skeletal muscle homogenates (Fig. 3A
). Thus, necrosis in
mdx muscle fibers does not seem to trigger general proteolitic
degradation of bulk skeletal muscle protein. Instead, the loss of
dystrophin seems to cause a specific reduction in AQP4 protein content.
|
To determine whether the reduction in the amount of AQP4 protein in the
skeletal muscle of mdx mice was caused by changes at the level of AQP4
mRNA, we performed semi-quantitative RT-PCR experiments (Fig. 3E
). The amount of AQP4-PCR product was analyzed together
with the amplification of 18S RNA, an invariant internal standard.
Results showed that the level of AQP4-PCR product in mdx mice was
similar to that of control mice (P>0.5, n=4)
indicating that the AQP4 mRNA level does not change significantly
during the progression of muscle pathology. Looked at together, these
results show that, in the skeletal muscle of mdx mice, AQP4 expression
is progressively reduced with the age of the animal and that this
reduction does not occur at transcription level.
Evaluation of AQP4 expression in mdx mouse brain
Western blot and RT-PCR
Since the mdx mouse brain metabolism is abnormal and associated
with an increase in extracellular volume (17)
, we next
evaluated whether changes in AQP4 expression also occur in the brain.
Coomassie blue-stained gels showed that the overall protein band
pattern exhibits no major differences between normal and pathological
specimens (Fig. 3C
), indicating that general proteolitic
degradation does not occur in dystrophic brain. AQP4 protein expression
was evaluated by densitometric analysis of immunoblot experiments using
a membrane fraction of brain homogenate. AQP4 content was progressively
reduced with age in mdx mice. Compared to control mice, AQP4 protein in
the brain is reduced by 30% (P<0.01, n=4) in
1-month-old mice. Strong reductions were observed in 1-year-old animals
(70%) (P<0.001, n=7) compared to the
age-matched control (Fig. 3D
). AQP4 mRNA levels in the brain
were analyzed by RT-PCR experiments. The results showed no significant
differences (P>0.5, n=4) in the amounts of AQP4
PCR product in both control and dystrophic mice, indicating that the
reduction in AQP4 in dystrophic brain mice was not at transcription
level (Fig. 3E
).
Immunocytochemistry
Immunohistochemistry experiments were performed in order to
evaluate the site of AQP4 alterations in the brain. At light
microscopy, the observation carried out on the immunostained sections
showed that the cytochemical localization of AQP4 in mdx mice brain was
similar to that in the controls, whereas in the former the reaction
intensity was diminished. The immunocytochemical analysis carried out
in the brain of control mice, including telencephalon, diencephalon,
and mesencephalon, showed that the anti-AQP4 antibody exclusively
stained bodies and processes of astrocytic cells, forming a dense glial
network throughout the neuropile and a continuous layer around the
vessels. These vessels showed a marked, continuous staining along the
abluminal side; labeled bodies and processes of glial cells were
recognizable, even close to the endothelium-pericyte layer (Fig. 4A
, B
, C
, D
, E
). In 1-year-old mdx mice, the bodies and processes of
astrocytes scattered in the neuropile appeared weakly AQP4
immunolabeled, and a remarkable labeling reduction was also observed in
the perivascular glial end-feet (Fig. 4G
, H
, I
, J
, K
). The
perivascular staining was point-by-point and discontinuous, labeled
vascular tracts alternating with unlabeled ones. The walls of some
vessels appeared quite unlabeled. In normal mice, a strong AQP4
immunoreaction was recognizable in the glial-limiting membrane formed
by the astrocyte end-feet in the submeningeal region (Fig. 4A
), whereas the submeningeal glial limiting membranes of
mdx brain showed a slight AQP4 expression or appeared completely
unstained (Fig. 4G
).
|
The brain osmosensory regions of control mice, including hypothalamic
magnocellular nuclei (Fig. 4D
, E
) and circumventricular
organs, showed strong AQP4 glial immunolabeling, especially at
perivascular level, whereas the neuronal cells, including the
magnocellular neurons in the paraventricular and supraopticus
hypothalamic nuclei, appeared completely unlabeled. In the diencephalic
brain regions of the mdx mice, including the osmosensory areas, the
AQP4 labeling appeared remarkably reduced. The glial processes
distributed throughout the neuropile and perivascularly arranged were
scantily stained, so that the brain parenchyma and vessel walls
displayed a weak labeling (Fig. 4I
, J
, K
).
A noticeable AQP4 reduction was observed in the periventricular
epithelium whose basolateral cellular membranes as well as the
subependymal layer glial cells appeared unstained or scarcely marked
(Fig. 4L
), whereas in the control brain the ependymal
epithelium appeared heavily labeled by anti-AQP4 antibody (Fig. 4F
). Similar results were obtained with four different
samples.
All the localization sites of AQP4 expression in mouse brain were
identical to those reported for rat brain (14
, 15
, 19
20
21,)
.
Immunogold electron microscopy confirmed the expression of astrocyte
bodies and processes. The astrocytic processes closely arranged around
the blood vessels were intensely labeled. Numerous 6 nm gold particles,
isolated or clustered, were seen on the membranes of the astrocytic
perivascular processes facing the capillaries (Fig. 5A
). On the ultrathin sections of mdx brain, the immunogold
reaction revealed a small number of AQP4 gold particles attached to the
bodies and processes of astrocytes. A strong reduction in labeling was
also observed in the glial end-feet facing the blood vessels. These
were surrounded by strongly swollen astrocytic processes whose plasma
membranes showed very rare, isolated gold particles (Fig. 5B
). Morphometric analysis of immunogold labeling indicated
a 85% reduction in gold particles (32±2/µm vs. 4.7±0.61/µm,
P<0.001, n=4). No labeling of intracellular
compartments was detected; the reduced AQP4 protein levels are thus not
a consequence of altered protein distribution within the cell.
|
Analysis of AQP4 expression in other mdx mouse tissues
To assess whether the reduction in AQP4 involves other tissues in
which this aquaporin is expressed (14
, 21)
, we
investigated AQP4 expression level in the kidney and stomach of mdx
mice. Immunofluorescence analysis indicated no major modifications of
AQP4 expression in the oxyntic glands of the lower stomach (Fig. 6A
) with a regular staining of the basolateral membrane of
parietal cells (Fig. 6B
), but not of the chief cells. Normal
staining was also observed in mdx mouse kidney. The medullary
collecting ducts all appeared to be intensely labeled by AQP4
antibodies (Fig. 6C
) and retained the specific staining of
the basolateral membrane of principal cells (Fig. 6D
).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSImmunocytochemistryAQP4 immunoprecipitation from...Preparation of brain membrane...Western blot analysisQuantitative reverse...RESULTSDISCUSSIONREFERENCES |
|---|
.
Thus, the purpose of this study was to analyze the expression of AQP4
in mdx mice, with particular regard to skeletal muscle and brain of
dystrophic mdx mice, and to determine whether AQP4 expression in brain
could also be affected in such animal models. Our immunofluorescence results showed that AQP4 reduction in skeletal muscle is progressive and begins by involving a subset of fast-twitch fibers identified as type IIB. The immunofluorescence and immunoblotting results together with quantitative PCR data indicate that AQP4 reduction seems to be related to a premature protein degradation. In fact, the AQP4 mRNA content was found to be unchanged, indicating that the protein reduction is at posttranslational level. Moreover, since the reduction in sarcolemma staining was not accompanied by an increase in intracellular labeling and the content of AQP4 protein was reduced in the whole muscle homogenate, we can conclude that the reduction in sarcolemma staining is not associated with an intracellular compartmentalization of mistargeted protein.
Many studies have analyzed the water and ion concentration in muscle
(22
, 23)
. These studies have demonstrated that the
contractile cycle appears to be associated with water entry into and
exit from the cell. In fact, during muscle exercise, there is a rapid
flux of water from the vascular compartment into contracting muscles,
due to an increase in intracellular osmolytes such as lactate
(24)
. This water displacement may affect the membrane
potential and thus muscle membrane excitability. Thus, AQP4 determining
a rapid osmotic transfer of water from blood to muscle during
contraction would therefore be important in establishing muscle fatigue
and so protect the cells against overload.
As demonstrated by the present study, mdx mice fast fibers
progressively reduce the sarcolemma expression of AQP4 protein. The
reduction seems to involve type IIB first and subsequently type IIA.
This is confirmed by analyzing the tibialis anterior fibers of
1-year-old mdx mice, where a weak residue of AQP4 labeling was only
associated with type IIA fibers. Normally dystrophin is expressed in
all myofiber types (25)
; nevertheless, type IIB fibers are
preferentially affected in DMD patients (26)
. Although the
mechanisms underlying this pathology remain to be determined, it is
possible that type IIB fibers are more sensitive to dystrophin
deficiency than others, and this may have a consequence in the
expression of AQP4. A previous study reported that mdx muscle fibers
show a decreased osmotic stability compared to normal muscle fibers
(27)
, but no indications on water content or changes were
reported. Other studies report variations in water content in the
muscles of mdx mice (28)
and dystrophic chicks
(29)
and significant increases in intracellular water
associated with a pronounced type II fiber deficiency in dystrophic
canine muscle (30)
. We can suspect that the reduced
presence of AQP4 in mdx fibers would determine important modifications
in the rapid and reversible cell volume equilibration in muscle and
thus affect many biological processes.
With respect to the pathophysiological mechanisms that may lead to
brain abnormalities in DMD, a deficiency in dystrophin may cause a
reduction in other associated proteins, such as the dystroglycan
complex (31)
. This would render specific regions of the
central nervous system more susceptible to cellular disturbances and
may result in cognitive impairment in some Duchenne patients. Similar
cognitive dysfunction also seems to occur in mdx mice
(32)
. Furthermore, in mdx mice, a significant increase in
extracellular brain volume leading to possible abnormal development has
been reported (17
, 33)
.
In the brain, AQP4 protein is strongly expressed in perivascular
astrocyte processes and ependymoglial cells, where it may regulate
water permeability at the blood–brain barrier (BBB) and cerebrospinal
fluid (CSF) reabsorption at the ventricular side (14
, 15
, 21)
. Therefore, we investigated the expression of AQP4 in
dystrophic brains. Our results indicate a marked reduction in AQP4
protein in mdx brain astrocytic processes and ependymoglial cells, and
demonstrate that this reduced staining is not due to an altered
distribution of AQP4 in the plasma membrane nor to a reduction in AQP4
mRNA content.
It is well known that in several CNS diseases as well as many
pathological conditions, BBB injury occurs tightly associated with an
increased vascular permeability (for a review, see ref
34
). Moreover, astrocytic cells have been demonstrated to
play a control role in BBB exchanges, maintaining CNS homeostasis by a
direct or indirect regulation of the ionic flux occurring during
neuronal activity with consequent osmotic water redistribution
(35
36
37)
. Therefore, the detection of a drastic AQP4
reduction in the perivascular astrocytic end-feet suggests that in the
mdx brain the regulation of the vascular water transport is altered
with a probable modification of the cerebral microenvironment and
neuronal cell activity. The reduction in AQP4 we found in the brains of
mdx mice was often associated with swollen astrocyte processes,
suggesting that its function is strictly related to the reabsorption of
water from the extracellular fluid to the blood and CSF. This agrees
with a recent study on rat brain AQP4 ontogeny, showing that it
accumulates in the brain as the volume fraction of the extracellular
space rapidly declines (16)
. Consequently, the reduction
of AQP4 protein in mdx mouse brain would determine a slower drainage of
water out of the brain, astrocyte swelling, and cellular brain edema.
This conclusion is further supported by a recent study reporting that
up-regulation of AQP4 mRNA occurs after brain injury that determined
BBB disruption and edema (38)
. Another possibility is that
the reduction of AQP4 protein in brain may be a consequence of an
altered vascular permeability of the BBB barrier.
Dystrophin deficiency results in the depletion of several associated
proteins (39)
. Moreover, the number of newly discovered
dystrophin-associated proteins (40)
is making the
dystrophin complex more complicated to study (41)
. From
our data, we can hypothesize that AQP4 in skeletal muscle interacts
with other sarcolemma proteins, such as those of the dystrophin
glycoprotein complex (DAPs). An accurate analysis of the AQP4 amino
acid sequence reveals a carboxyl-terminal tetrapeptide motif
X-Ser/Thr-X-Val-COO-, which is known to interact with PDZ domains. The
PDZ domain is composed of 90 amino acids and was originally identified
in postsynaptic density-95, disc large, and ZO-1 (42)
. The
PDZ domain is present in diverse families of structural proteins and
appears to be involved in the targeting and clustering of membrane
proteins. In skeletal muscle there are a few PDZ domain proteins; these
include neuronal nitric oxide synthase and the family of syntrophins
(42)
, both of which are components of the dystrophin
complex (43)
. In particular, the role of syntrophin in
skeletal muscle seems to be to recruit signaling proteins, such as
voltage-gated sodium channels, to the membrane (44)
, thus
acting as a modulator adaptor protein. Of all the components of the
DAPs complex, syntrophin seems most likely to be the one that interacts
with AQP4 in skeletal muscle. This is further supported by a recent
study showing that in alpha 1-syntrophin knockout mice, aquaporin-4 is
absent at the sarcolemma and at perivascular astrocyte end-feet
(45)
. Identifying other proteins associated with AQP4 may
be particularly relevant in order to understand the molecular origin of
the DMD myopathy as well as the physiological role of AQP4 in skeletal
muscle.
The reduction in AQP4, which is not seen at transcription level, could
be due to a reduced stability of the DAP complex when there is no
dystrophin to anchor the complex. Further studies are needed to
determine the biochemical relationship between AQP4 and the ensemble of
proteins bound to dystrophin in order to determine whether AQP4
contributes directly to the pathogenesis of muscular dystrophy, as
occurs with some individual components of the DAP complex.
Immunoreactive dystrophin has recently been detected in perivascular
astrocytes and its developmental expression coincides with the
development of the BBB (46)
. Since AQP4 expression seems
unaltered in stomach and kidney, this suggests a different membrane
organization of this water channel in the neuromuscular system compared
to the other epithelial membranes in which this water channel is
expressed. A recent study (47)
demonstrating that other
AQP4 polypeptides as well as the two major isoforms (30 and 32 kDa)
exist in the brain and many other tissues agrees with this conclusion.
This suggests that different AQP4 isoforms may play a role in the
formation and distribution of AQP4-containing orthogonal arrays or in
their trafficking.
A transgenic null mouse lacking AQP4 has been generated recently
(48)
. According to the data, no abnormalities in
neuromuscular function were observed in homozygous AQP4 knockout mice,
indicating that AQP4 could not play a key role in the regulation of
water homeostasis in the neuromuscular system. More recently, the same
group reported that AQP4 null mice manifested less brain edema after
water intoxication and stroke, pointing to a role played by AQP4 in
pathological situations that determine brain edema (49)
.
However, the possibility that another unknown water channel may be
induced when AQP4 is absent as in transgenic null mouse cannot be ruled
out. This is a consistent hypothesis, since it has been reported that
the last cloned water channel (AQP9) seems to be expressed in brain
astrocytes (50)
. A preliminary immunolocalization study
performed by us confirms that AQP9 is expressed in brain GFAP positive
cells (unpublished observation), indicating that this aspect needs to
be studied in AQP4 null mice.
In conclusion, mdx mice manifest a reduction in AQP4 expression in muscle and brain. This reduction is age related; in the brain it is associated with swollen perivascular astrocyte processes. We suppose that some neurological dysfunctions of mdx mice and DMD patients could be associated with changes in brain osmotic equilibrium.
|
ACKNOWLEDGMENTS |
|---|
Received for publication April 24, 2000.
Revision received July 5, 2000.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSImmunocytochemistryAQP4 immunoprecipitation from...Preparation of brain membrane...Western blot analysisQuantitative reverse...RESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  E. Migliati, N. Meurice, P. DuBois, J. S. Fang, S. Somasekharan, E. Beckett, G. Flynn, and A. J. Yool Inhibition of Aquaporin-1 and Aquaporin-4 Water Permeability by a Derivative of the Loop Diuretic Bumetanide Acting at an Internal Pore-Occluding Binding Site Mol. Pharmacol., July 1, 2009; 76(1): 105 - 112. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. E. Benarroch Aquaporin-4, homeostasis, and neurologic disease Neurology, December 11, 2007; 69(24): 2266 - 2268. [Full Text] [PDF]
|
|
|
|
 |
|
 B. Gavillet, J.-S. Rougier, A. A. Domenighetti, R. Behar, C. Boixel, P. Ruchat, H.-A. Lehr, T. Pedrazzini, and H. Abriel Cardiac Sodium Channel Nav1.5 Is Regulated by a Multiprotein Complex Composed of Syntrophins and Dystrophin Circ. Res., August 18, 2006; 99(4): 407 - 414. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. Laforenza, E. Cova, G. Gastaldi, S. Tritto, M. Grazioli, N. F. LaRusso, P. L. Splinter, P. D'Adamo, M. Tosco, and U. Ventura Aquaporin-8 Is Involved in Water Transport in Isolated Superficial Colonocytes from Rat Proximal Colon J. Nutr., October 1, 2005; 135(10): 2329 - 2336. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. A. Lennon, T. J. Kryzer, S. J. Pittock, A.S. Verkman, and S. R. Hinson IgG marker of optic-spinal multiple sclerosis binds to the aquaporin-4 water channel J. Exp. Med., August 15, 2005; 202(4): 473 - 477. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. W. Yeung, N. P. Whitehead, T. M. Suchyna, P. A. Gottlieb, F. Sachs, and D. G. Allen Effects of stretch-activated channel blockers on [Ca2+]i and muscle damage in the mdx mouse J. Physiol., January 15, 2005; 562(2): 367 - 380. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Dalloz, R. Sarig, P. Fort, D. Yaffe, A. Bordais, T. Pannicke, J. Grosche, D. Mornet, A. Reichenbach, J. Sahel, et al. Targeted inactivation of dystrophin gene product Dp71: phenotypic impact in mouse retina Hum. Mol. Genet., July 1, 2003; 12(13): 1543 - 1554. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Amiry-Moghaddam, T. Otsuka, P. D. Hurn, R. J. Traystman, F.-M. Haug, S. C. Froehner, M. E. Adams, J. D. Neely, P. Agre, O. P. Ottersen, et al. An alpha -syntrophin-dependent pool of AQP4 in astroglial end-feet confers bidirectional water flow between blood and brain PNAS, February 18, 2003; 100(4): 2106 - 2111. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. Vajda, M. Pedersen, E.-M. Fuchtbauer, K. Wertz, H. Stodkilde-Jorgensen, E. Sulyok, T. Doczi, J. D. Neely, P. Agre, J. Frokiar, et al. Delayed onset of brain edema and mislocalization of aquaporin-4 in dystrophin-null transgenic mice PNAS, October 1, 2002; 99(20): 13131 - 13136. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Zelenina, S. Zelenin, A. A. Bondar, H. Brismar, and A. Aperia Water permeability of aquaporin-4 is decreased by protein kinase C and dopamine Am J Physiol Renal Physiol, August 1, 2002; 283(2): F309 - F318. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. H. CROSBIE, S. A. DOVICO, J. D. FLANAGAN, J. S. CHAMBERLAIN, C. L. OWNBY, and K. P. CAMPBELL Characterization of aquaporin-4 in muscle and muscular dystrophy FASEB J, July 1, 2002; 16(9): 943 - 949. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Agre, L. S King, M. Yasui, W. B Guggino, O. P. Ottersen, Y. Fujiyoshi, A. Engel, and S. Nielsen Aquaporin water channels - from atomic structure to clinical medicine J. Physiol., July 1, 2002; 542(1): 3 - 16. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. L. Anderson, S. I. Head, C. Rae, and J. W. Morley Brain function in Duchenne muscular dystrophy Brain, January 1, 2002; 125(1): 4 - 13. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
