Download - Wiley Online Library

Journal of Neurochemistry
Lippincott—Raven Publishers, Philadelphia
© 1997 International Society for Neurochemistry
Increased Expression of Cathepsins E and D in Neurons of
the Aged Rat Brain and Their Colocalization with Lipofuscin
and Carboxy-Terminal Fragments of Alzheimer
Amyloid Precursor Protein
Hiroshi Nakanishi, Tutomu Amano, Dewi F. Sastradipura, *yoshito Yoshimine,
Takayuki Tsukuba, Kazunari Tanabe, tlchirou Hirotsu,
lTomochika Ohono, and Kenji Yamamoto
Departments of Pharmacology and * Periodontics and Endodontics, Kyushu University Faculty of Dentistry, Fukuoka, and
t Laboratory of Experimental Pharmacology, Suntory Institute for Biomedical Research, Osaka, Japan
Abstract: Age-related changes in the expression and
localization of two distinct intracellular aspartic proteinases, cathepsin F (CE) and cathepsin D (CD), were investigated in the rat cerebral cortex and the brainstem
by immunocytochemical and quantitative methods using
discriminative antibodies specific for each enzyme. Nonlysosomal CE was barely detectable in these two brain
tissues in the embryonic stages, whereas relatively high
expression of lysosomal CD was observed in embryonic
tissues. After birth, CE was increasingly expressed in
these tissues with aging to attain maximal levels at 30
months of age. Western blot analyses revealed that CE
existed predominantly as the mature enzyme at 2 and
17 months of age, whereas it was present as not only
the mature enzyme but also the proenzyme at 30 months
of age. On the other hand, CD was mainly present in the
mature form throughout development, although its level
in these tissues was also significantly increased with
aging. The CE-positive cortical and brainstem neurons
of the aged rat corresponded well with cells emitting
autofluorescence for lipopigments. By the double-staining technique, most of the CE-positive cortical and
brainstem neurons of the aged rat were also positive for
antibody to the carboxyl-terminal fragments of amyloid
precursor protein (APP534_695), intracellular accumulation of which is thought to be associated with age-related changes in the endosome/lysosome system. It is
important that electron microscopy revealed that CE in
brainstem neurons of the aged rat colocalized with CD
in the lipofuscin-containing lysosomes. These results indicate that aging results in the increased expression and
lysosomal localization of CE in cortical and brainstem
neurons and changes in the endosomal/lysosomal proteolytic system, which may be related to lipofuscinogenesis and altered intracellular APP metabolism. Key
Words: Cathepsin E—Cathepsin D—Aspartic proteinase—Lipofuscin—Amyloid precursor protein—Aging
—Neurodegeneration.
J. Neurochem. 68, 739—749 (1997).
In past years attention has been paid to the altered
intracellular proteolytic system in relation to different
neuropathological conditions, as irregular cellular proteolysis of physiologically important neuronal proteins
has severe consequences for the integrity of neuronal
structure and function. The disorder or breakdown of
the endosomal/lysosomal proteolytic system has also
been suggested to be of pathological significance in
neurodegenerative diseases such as Alzheimer’ s disease (Nixon and Cataldo, 1993). Therefore, it has been
believed that alterations in proteolytic systems in neurons play a crucial role in cellular events that leads to
neuropathological conditions.
Lipofuscin is a major type of residual body observed
in various types of neurons during the aging process
and is considered to be formed initially by autophagocytosis of cytoplasmic components and ultimately defined as a membrane-bound lysosomal organelle containing lipoidal moieties. Although the chemical composition of lipofuscin is very heterogeneous, the
formation of lipofuscin is thought to be related to a
decrease in lysosomal proteolytic activity (Sohal and
Wolfe, 1986). On the other hand, several lines of evidence indicate that levels of amyloidogenic C-terminal
fragments generated from amyloid precursor protein
(APP) increase in neurons of the aged brain (Benowitz
Received July 2, 1996; revised manuscript received October 7,
1996; accepted October 8, 1996.
Address correspondence and reprint requests to Dr. H. Nakanishi
at Department of Pharmacology, Kyushu University Faculty of Dentistry, Fukuoka 812, Japan.
Abbreviations used: AEC, aminoethylcarbazole; APP, amyloid
precursor protein; CD, cathepsin D; CE, cathepsin E; LDH, lactate
dehydrogenase; P fraction, particulate fraction; PBS, phosphate-buffered saline; S fraction, soluble fraction; SDS, sodium dodecyl sulfate.
739
740
H. NAKANISHI ET AL.
et a!., 1989; Takemura et al., 1993; Beeson et a!.,
1994). It is generally accepted that the formation and
deposition of the amyloidogenic APP fragments are
closely associated with specific proteolytic events, especially the altered endosomal/lysosomal system (Benowitz et al., 1989; Cole et a!., 1989; Golde et a!.,
1992; Haass et al., 1992), and that the amyloidogenic
APP fragments are routinely produced in an acidic
compartment by noncysteine proteases and that they
are eliminated by lysosomal cysteine proteinases such
as cathepsins B and L (Siman et al., 1993). The importance of lysosomal cysteine proteinases in elimination
of the potentially amyloidogenic APP has also been
suggested by accumulation of potentially amyloidogenic C-terminal APP fragments after treatment with
the general cysteine proteinase inhibitors E-64 and leupeptin (Hayashi et al., 1992; Bahr et al., 1994; 1-lajimohammadreza et al., 1994) and by its possible association with the decreased activities of cathepsins B and
L in aged rat brain tissues (Nakanishi et al., 1994;
Amano et al., 1995a). Taken together, it is of special
interest to note that the injection of leupeptin or chloroquine, both inhibitors of endosomal/lysosomal proteolysis, into young rat brains causes the formation of
lysosomal granular aggregates that closely resemble
the ceroid-lipofuscin (Ivy et al., 1984). Therefore, agerelated accumulation of lipofuscin and the C-terminal
APP fragment is thought to beassociated with an alteration of the endosomal/lysosomal system.
Recently, the activity of cathepsin D (CD; EC
3.4.23.5), a typical lysosomal aspartic proteinase, has
been shown to increase in various brain regions of
both the aged rat and human (Matus and Green, 1987;
Kenessey et al., 1989; Nakamura et al., 1989u; BanaySchwartz et al., 1992), especially in neurons from aged
rat brain (Nakanishi et al., 1994), suggesting the agedependent enhancement of the endosomal/lysosomal
system in these cells. It has been shown that the enzymatically active CD, as well as the lysosomal cysteine
proteinase cathepsins B and L, is present extracellularly at high levels in the senile plaques (Bernstein et
al., 1989; Cataldo and Nixon, 1990; Cataldo et al.,
199 I). Furthermore, secondary lysosomes and residual
bodies are markedly increased in affected neurons of
Alzheimer’s disease brain (Nixon and Cataldo, 1993;
Cataldo et al., 1994). On the other hand, cathepsin E
(CE; EC 3.4.23.34) is another intracellular aspartic
proteinase, whose biochemical and catalytic features
such as substrate specificity, susceptibility to various
inhibitors, and pH optimum, are similar to those of CD
(Yamamoto Ct al., 1978, 1979, 1985; Yonezawa et al.,
1987). However, CE is apparently distinct from CD
in tissue distribution and cellular localization. CE is
found to be localized mainly in the endosome-like
structure, the endoplasmic reticulum, and the plasma
membrane but not in a typical lysosomal compartment
in normal tissues (Ichimaru et al., 1990; Saku et al.,
1991; Solcia et al., 1993). Taken together, the limited
J. Neurochem., Vol. 68, No. 2, /997
tissue distribution of CE suggests its regulated gene
expression and its involvement in different proteolytic
events from CD. Increased expression and abnormal
localization of CD have been well described in the
brain during aging, but less attention has been given
to age-dependent changes in CE of the brain. Recently,
CE was shown to be highly expressed in degenerating
neurons and to be localized in the core of the senile
plaques of Alzheimer’s disease brain (Bernstein and
Wiederanders, 1994). In a previous study (Nakanishi
et al., 1994), we found that CE was greatly expressed
in both neurons and microglia from the aged rat brain,
although it was barely detectable in the normal young
rat brain. We also showed the alteration in immunostaining pattern of CE in neurons from a diffuse type
to a granular type during the aging process. However,
it is not clear whether the age-related increase in activity of CE and CD is associated with the maturity or
senescence processes. Also, there is little understanding of the molecular forms of these two enzymes and
their accumulation sites in aged neurons.
In view of the evidence linking the endosomal/lysosomal proteolytic system to the pathogenesis of lipofuscin and amyloid deposits, it is important to determine the association of the proteolytic events relating
to CE and CD with the age-related neuronal degeneration process. In this study we report their expression,
molecular forms, and cellular localization in aged neurons. Furthermore, to define the altered endosomal/
lysosomal system in aged neurons, the cellular localization of CE and CD was compared with that of Iipofuscin and C-terminal APP fragments.
MATERIALS AND METHODS
Materials
Fischer 344 rats were divided into four different age
groups: embryonic (embryonic day 18), young (2 months
of age), middle-aged (17 months of age), and aged (30
months of age) rats. Antibodies against rat spleen CE (Yamamoto et al., 1978) and CD (Yamamoto et al., 1979) were
raised in rabbits and purified by affinity chromatography as
described previously (Yamamoto et a!., 1985). Antibody
specific for the synthetic peptide Ser-Gln-Leu-Ser-Glu-PheTrp-Lys-Ser-His-Asn-Leu-Asp-Met, which corresponds to
2~-Met3~
of human
the
prosequence
comprising
pro-CE
cDNA (Azuma
et a!.,residues
1989), Ser
was raised in rabbits
and purified on the peptide-Sepharose 4B affinity column.
Monoclonal antibodies against the N-terminus of APP
(22C1 1 clone; its epitope lies between residues 60 and 100
of APP) and the C-terminal portions (anti-APP
64~9~,
its
epitope lies between residues 643 and 695) were purchased
from Boehringer Mannheini Biochemica.
Sodium dodecyl sulfate (SDS)-gel electrophoresis
and immunoblotting
Four rats each were used for immunoblotting analysis at
2, 17, and 30 months of age. Fourteen embryonic rats were
used for immunoblotting analyses. The detailed immunoblotting procedure has been described previously (Nakanishi et
CATHEPSINS E AND D IN AGED RAT BRAIN
al., 1994; Amano et al., 1995b). In brief, the cerebral cortex
and the brainstem were obtained from rats in each age group
that were anesthetized with sodium pentobarbital (40 mgI
kg, i.p.) and killed by intracardiac perfusion with isotonic
saline. The soluble and particulate fractions obtained from
the cerebral cortex and brainstem homogenates by differential centrifugation as described below were electrophoresed
in SDS-polyacrylamide gels. For immunoblotting, proteins
on SDS gels were transferred electrophoretically at 100 V
for 12—15 h from the gels to nitrocellulose membranes according to the method of Towbin et al. (1979) and then
incubated at 4°Covernight under gentle agitation with one
of the following primary antibodies: anti-CE IgG (0.50 ,ugl
ml), anti-CD IgG (1.55 ~g/ml), 22C11 (5 ~tg/ml), or antiAPP
643.695 IgG (10 jig/ml). After washes, the membranes
were incubated with 0.5% horseradish peroxidase-labeled
donkey anti-rabbit IgG (Amersham). Subsequently, mem-
brane-bound horseradish peroxidase-labeled antibodies were
detected by the enhanced chemiluminescence detection system (ECL kit; Amersham) on x-ray film (X-Omat; Kodak)
30—60 s after exposure. As a control, the primary antibody
was replaced by nonimmune rabbit IgG. The protein bands of
CE and CD on x-ray film were scanned and densitometrically
analyzed by a densitometer (Personal Scanning Imager
PD11O; Molecular Dynamics).
Subcellular fractionation
Three rats each were used for subcellular fractionation at 2
and 30 months of age. The cerebral cortex and the brainstem
obtained from rats in each age group were homogenized in
3 volumes of 0.25 M sucrose/0.2 M KC1 (pH 7.4) with a
Teflon glass homogenizer (20 strokes). Each homogenate
was centrifuged at 650 g for 10 mm at 4°C.The supernatant
—
was further centrifuged at 10,000 g for 20 mm at 4°C.The
resulting precipitate was resuspended in the same buffer and
designated as the particulate fraction (P fraction), which is
rich in lysosomes and mitochondria. The supernatant was
centrifuged at 105,000 g for 1 h at 4°C.The final supernatant
was designated as the soluble fraction (S fraction).
Determinations
Acid proteinase activity was determined at pH 3.5 using
1.5% acid-denatured hemoglobin as a substrate as described
(Sakai et al., 1989; Nakanishi et al., 1993). ~-Glucuronidase
was assayed with 4-methylumbelliferyl-/3-D-glucuronide as
a substrate by the method of Robins et al. (1968). Lactate
dehydrogenase (LDH) was assayed by the method of
Wacker et al. (1956). Protein content was determined by the
method of Lowry et al. (1951) with bovine serum albumin as
a standard.
Preparation of cell cultures
Primary cultures of cortical neurons were prepared from
embryonic day 18 rats. The cerebral cortices were dissected
and transferred into ice-cold oxygen-bubbled L- 15 medium
(GIBCO, Gaithersburg, MD, U.S.A.). Then, the tissues were
enzymatically 2~
dissociated
- and Mg2 by
-free
incubation
phosphate-buffered
twice for 15saline
mm
at 37°Cin
(PBS)
containing
Ca
papain (Worthington, Freehold, NJ,
U.S.A.; 90 units), DNase I (Worthington; 2,000 units), D,Lcysteine-HC1 (Sigma, St. Louis, MO, U.S.A.; 2.23 mg),
bovine serum albumin (Nacalai, Kyoto, Japan; 2 mg), and
glucose (50 mg). After digestion, the tissue fragments were
resuspended in a culture medium consisting of 90% of a
741
1:1 mixture of Dulbecco’s modified Eagle’s (GIBCO) and
Ham’s F12 (GIBCO) media, 5% heat-inactivated horse serum (GIBCO), and 5% heat-inactivated fetal bovine serum
(GIBCO) containing 15 mM HEPES buffer (pH 7.2), 30
nM selenium, 1.9 mg/ml sodium bicarbonate, 100 units/mI
penicillin G, and 100 mg/mI streptomycin sulfate. The tissue
fragments were then dissociated by gentle passage through
plastic tips with three different diameters. Dissociated cells
were filtered through a nylon cell strainer with a pore size
of 70 /.tm (Falcon, Franklin Lakes, NJ, U.S.A.) and centrifuged; then the cells were resuspended in the culture medium
and plated in dishes (106 cells per dish) precoated with
polyethylenimine (Sigma) and maintained in a humidified
CO
2 incubator (95% air/5% CO2) at 37°C. After 2 days,
nonneuronal cell division was halted by 24 h of exposure to
i0~ M cytosine arabinoside (Sigma). Subsequent medium
replacement was carried out every 2 days. After 7 days in
vitro, cultures were subjected to immunoblot analyses.
Immunolight microscope
Four rats each were used for immunohistochemical analyses at 2 and 30 months of age. The detailed immunohistochemical procedure has been described previously (Nakanishi et al., 1994; Amano et al., l995b). In brief, specimens
were obtained from rats in each age group, which were anesthetized with sodium pentobarbital (40 mg/kg, i.p.) and
killed by intracardiac perfusion with isotonic saline followed
by a chilled fixative consisting of 4% paraformaldehyde in
0.2 M PBS (pH 7.4). After perfusion, the brain was removed
and further fixed by immersion in the same fixative for 24
h at 4°Cand then immersed in 20% sucrose (pH 7.4) for
24 h at 4°C.Floating parasagittal sections (30 ~smthick) of
the brainstem and the cerebral cortex were prepared by a
cryostat and stained by anti-CD IgG (1.55 ~sg/ml) or antiCE IgG (0.50 ~ig/ml) with the avidin—biotin—peroxidase
complex method. After PBS washes, sections were reacted
with aminoethylcarbazole (AEC; Dako Corp., Carpinteria,
CA, U.S.A.) for 5—20 mm. For double immunostaining,
AEC was used as a chromogen yielding a xylene-soluble
orange precipitate. After immunostaining for CE or CD,
AEC precipitates for CE or CD labeling were removed by
incubating the sections in 50% alcohol for 5 mm and then
in xylene for 30 mm at room temperature. After washing
with PBS, the sections were stained by 22C1 1 (10 ~ig/ml)
or anti-APP643_695 IgG (5 ~sg/ml) with the avidin—biotin—
peroxidase method followed by reaction with 0.025% 3,3 ‘diaminobenzidine/0.4% (NH4)2Ni (SO4)2/0.09% H202/0. I
M Tris-buffered saline solution for 5—10 mm. All sections
were thoroughly rinsed with PBS, mounted, and coverslipped with glycerol gelatin (Sigma). As an immunohistochemical control, the sections were incubated with nonimmune rabbit IgG and treated as described above.
Immunoelectron microscopy
For immunoelectron microscopy, a mixture of 4% paraformaldehyde and 0.1% glutaraldehyde containing 0.05%
CaC12 and 0.1% tannic acid in 0.1 M cacodylate buffer (pH
7.4) was used for a fixative. After perfusion, the brain was
removed and further fixed by immersion in the same fixative
for 1 h. After quenching in 50 mM glycine in cacodylate
buffer, the tissue pieces were rinsed at 4°C overnight in
cacodylate buffer containing 7% sucrose. The specimens
were subsequently postfixed in ferrocyanide-reduced I %
0s04 at 4°Cfor 30 mm, dehydrated, and embedded in LR
J. Neurochem., Vol. 68, No. 2, /997
742
H. NAKANISHI ET AL
White. They were then polymerized by ultraviolet irradiation
at —25°Cfor 72 h. Ultrathin sections were mounted on Formvar-coated 150-mesh nickel grids. Nonspecific staining was
blocked by treatment in 1% bovine serum albumin in PBS
containing 0.1 M lysine hydrochloride for 30 mm. The grids
were reacted for 3 h by floating them on drops of anti-CE
IgG (10 ~sg/ml) or anti-CD IgG (37 ~sg/ml). After rinsing
with PBS, the grids were incubated for 1 h on 10-nm colloid
gold-conjugated anti-rabbit IgG diluted to 1:40 in 1% bovine
serum albumin in PBS. After rinsing with PBS, the immuno-
forms of CE, revealed that a definitive immunoreactive
labeled sections were stained with uranyl acetate for 5 mm
age were increased -~1.4 and 1.7 times as much as
those at 2 months of age, respectively. On the other
hand, immunoblot analyses using the antibody specific
for CD showed an intense band with a molecular mass
of 44 kDa, which corresponded to the mature form,
throughout the entire age (Fig. I c and d). The CD
levels were also increased with advancing age by up
to 1.5-fold in the cerebral cortex and 1.7-fold in the
cerebral cortex at 30 months of age as compared with
those at the embryonic stage.
The cortex extracts from 2-month-old rats and the
extracts of primary cultured cortical neurons from the
embryonic rat showed a 42-kDa protein band, whereas
two bands with apparent molecular sizes of 46 (major)
and 42 (minor) kDa were clearly observed in the
spleen extracts of 2-month-old rats (Fig. 2a). The 46kDa form in the spleen extract was sensitive to conversion to the 42-kDa form by a brief acid treatment at
pH 3.5 at 37°Cfor 10 mm, whereas the 42-kDa form
was resistant to this treatment (Fig. 2a). Furthermore,
the antibody specific for the synthetic peptide Ser-GlnLeu-Ser-Glu-Phe-Trp-Lys-Ser-His-Asn-Leu-Asp-Met
and observed using an H-7000 transmission microscope (Hitachi, Tokyo, Japan) operated at 75 kV.
RESULTS
Immunoblot analyses of CE and CD
The activity levels of aspartic proteinase (CE plus
CD) were analyzed in the brain extracts prepared from
the cerebral cortex and the brainstem of embryonic
and 2-, 17-, and 30-month-old rats. Increased age was
found to be associated with elevated levels of the activity in both the cerebral cortex and the brainstem. The
mean ±SEM specific activity in the cerebral cortex
of embryonic and 2-, 17-, and 30-month-old rats was
2.5 ±0.1 (n = 14), 5.1 ±0.7 (n = 4), 5.0 ±0.1 (n
= 4), and 9.6 ± 1.8 (n = 4) units/mg of protein,
respectively. On the other hand, the mean ± SEM
specific activity in the brainstem of embryonic and 2and 30-month-old rats was 2.5 ±0.1 (n = 14), 15.3
±4.6 (n = 4), and 29.0 ±4.2 (n = 4) units/mg of
protein, respectively. Figure 1 shows the results of
immunoblot analyses of CE and CD in the cerebral
cortex and brainstem from rats of four age groups.
Immunoblot analyses using the polyclonal antibody for
CE, which can recognize both the pro- and mature
band was barely detectable in the tissue extract of embryonic brain. However, the extracts of both brain tissues at 2, 17, and 30 months of age revealed the increased expression of the 42-kDa band, and an additional band with a molecular mass of 46 kDa was
observed especially at 30 months of age (Fig. la and
b). Densitometric scanning showed that the CE levels
in the cerebral cortex and brainstem at 30 months of
reacted with the 46-kDa band in the spleen extracts
but not with the 42-kDa bands in either tissue (Fig.
2b). Therefore, the 46- and 42-kDa forms were found
to be the mature and proform of CE, respectively.
FIG. 1. Immunoblot analyses of whole extract of cerebral cortex (CX) and brainstem (BS) from rats of
four different ages by anti-CE (a and b) and anti-CD
(c and d) antibodies. Each column represents the
mean relative immunoreactivity of four experiments.
The relative immunoreactivity of CE and CD was determined as compared with 2 months of age and
embryonic, respectively. E, embryonic day 18; 2M,
2 months old; 1 7M, 17 months old; 30M, 30 months
old.
J. Neurochem., Vol. 68, No. 2, 1997
CATHEPSINS E AND D IN AGED RAT BRAIN
743
the other hand, the LDH content recovered in the P
and S fractions was independent of age, suggesting
FIG. 2. Immunoblot analyses of whole extracts from various rat
tissues by (a) anti-CE and (b) anti-pro-CE antibodies. Extracts
without (—) and with (÷)a brief acid treatment were analyzed
in a. CX, cerebral cortex; BS, brainstem; SP, spleen; CL, cortical
cell cultures.
Subcellular fractionation
To examine the age-related alteration in intracellular
distribution of CE and CD, their levels in the subcellular fractions prepared from tissue extracts of the cerebral cortex and the brainstem of 2- and 30-month-old
rats were measured. At 2 months of age, the P and S
fractions were found to be rich in the lysosome and
the cytoplasmic matrix, respectively, as judged by the
distribution of /3-glucuronidase (a marker for the lysosome) and LDH (a marker for the cytosol). A small
amount of /3-glucuronidase was detected in S fractions
of the cerebral cortex (6%) and the brainstem (7%).
However, its level in the S fractions was markedly
increased in the brain from 30-month-old rats (16%
for the cerebral cortex and 21 % for the brainstem). On
that the age-dependent increase of autophagy has no
significant effect on distribution of the cytosolic components. As shown in Fig. 3a and b, CE was exclusively confined to P fractions of both the cerebral cortex and the brainstem from 2-month-old rats. However,
in tissues from 30-month-old rats, CE was found in
the S fractions to a similar or even higher extent than
shown in the P fractions. The CE levels in the P fraction between 2- and 30-month-old rats were not significantly changed in the cerebral cortex but increased
1.7-fold in the aged brainstem. On the other hand, CD
was also mainly found in the P fraction and partially
in the S fraction of 2-month-old rat brain (Fig. 3c and
d). The CD levels in both brain tissues were significantly increased in 30-month-old rat brain as compared
with 2-month-old rats. The marked increase of CD
level in the S fractions of the aged rat tissues may
represent the increased fragility of the lysosomal membrane of aged cells.
Localization of CE and CD at light and electron
microscopic levels
At 2 months of age, CE was barely detectable in
both the frontal portion of the cerebral cortex (Fig.
4a) and the pontine reticular formation of the brainstem (Fig. 5a), whereas CD was widely but unevenly
distributed as coarse intraplasmic granules in these tissues (Figs. 4c and Sc). At 30 months of age, marked
increase in CE immunoreactivity was observed in both
cortical (Fig. 4b) and brainstem (Fig. Sb) neurons. It
was noted that CE was greatly accumulated as large
granular immunoreactive products in large neurons of
Immunoblot analyses of the extracts of P and
S fractions of brain tissues from young and aged rats
by (a and b) anti-CE and (C and d) anti-CD antibodies. Each column represents the mean relative immunoreactivity of three experiments. The relative immunoreactivity of both CE and CD was determined as
compared with the S fraction of 2-month-old rats. CX,
cerebral cortex: BS, brainstem; 2M, 2 months old;
30M, 30 months old.
FIG. 3.
J. Neurochem., Vol. 68, No. 2, /997
744
H. NAKANISHI ET AL.
FIG. 4. Age-related increase in CE and CD immunoreactivities (upper panels) and their colocalization with autofluorescent lipopigments
(lower panels) in cortical neurons. a and b: Immunostaining of CE in (a) 2- and (b) 30-month-old rats. c and d: Immunostaining of
CD in (c) 2- and (d) 30-month-old rats. e—h: Autofluorescent lipopigments observed under the ultraviolet light in the same section
shown in a—d before immunostaining, respectively. Colocalization of CE immunoreactivities and autofluorescent lipopigments in cortical
neurons of the aged rat was indicated by arrowheads in b and f. Colocalization of CD immunoreactivities and autofluorescent lipopigments in cortical neurons of the aged rat was indicated by arrowheads in d and h. Bar = 50 ~.tm.
these brain regions. Microglia in the white matter were
also intensely immunostained with anti-CE antibody
(data not shown). Immunoreactivity of CD was increased in both neurons and glial cells in these brain
regions of 30-month-old rats (Figs. 4d and Sd).
The lower panels in Figs. 4 and 5 show age-dependent accumulation of lipofuscin granules in the same
sections of the upper panels. At 30 months of age,
intense autofluorescent granules were found to accumulate in the focal cytoplasmic site of the majority of
cortical (Fig. 4f and h) and brainstem (Fig. 5f and h)
neurons. These autofluorescent pigments were considered to be lipofuscins because they exhibited yellowbrown fluorescence under ultraviolet light. The localization of the autofluorescent pigments corresponded
well with that of CE (Figs. 4b and f and Sb and f).
CD was also partially colocalized with autofluorescent
pigments (Figs. 4d and h and Sd and h). In contrast,
the cortical and brainstem neurons from 2-month-old
rats showed no such autofluorescence (Figs. 4e and g
and Se and g). The immunoreaction products with both
antibodies had to be weakly demonstrated, because the
intense immunostainings interfered with clear demon-
stration of the autofluorescence of lipofuscin. The dis-
J. Neurochem., Vol. 68, No. 2, 1997
crepancy in immunoreactivity between CE and CD was
mainly due to the practical difficulty in demonstrating
low concentrations of the cellular CE. However, it was
confirmed that CE was present in most of the lipofuscm-positive neurons, if not all (data not shown).
Ultrastructurally, lipofuscin is known to accumulate
in lysosomes as electron dense granules and lipid-like
granules. As shown in Fig. 6b, immunogold particles
indicating antigenic sites for CD were mainly associated within such lipofuscin-containing organelles in
the pontine reticular formation neurons of the brainstem of 30-month-old rats. A small number of lysosomes without lipofuscin-like materials was also labeled by immunogold particles for CD. Because CD,
as well as cathepsins B and L, is known as a representative lysosomal hydrolase, the lipofuscin-like materials
were found to accumulate in secondary lysosomes. It
is interesting that immunogold particles indicating antigenic sites for CE were predominantly found in lipofuscin-contamning lysosomes accumulated near the nuclear membrane (Fig. 6a). In serial thin sections, the
same granules containing lipofuscin-like materials
were labeled by immunogold particles for both CD
and CE (data not shown). These results indicate the
CATHEPSINS E AND D IN AGED RAT BRAIN
745
FIG. 5. Age-related increase in CE and CD immunoreactivities (upper panels) and their colocalization with autofluorescent lipopigments
(lower panels) in brainstem neurons. a and b: Immunostaining of CE in (a) 2- and (b) 30-month-old rats. c and d: Immunostaining
of CD in (c) 2- and (d) 30-month-old rats. e—h: Autofluorescent lipopigments observed under the ultraviolet light in the same section
shown in a—d before immunostaining, respectively. As shown by arrowheads in b and f, CE immunoreactivities in brainstem cells of
the aged rat corresponded well with autofluorescent lipopigments. As shown by arrowheads in d and h, CD immunoreactivities in
brainstem neurons of the aged rats partially colocalized with autofluorescent lipopigments. Bar = 50 tim.
colocalization of CE and CD in lipofuscin-loaded lysosomes of neurons from 30-month-old rats.
ing with the antibody to the C-terminal fragment of
APP (APP643_695). At 2 months of age, the immunore-
Colocalization of CE and CD with APP
Figures 7 and 8 show a marked difference between
2- and 30-month-old rat brain tissues in immunostain-
activity was scarcely detectable in either the frontal
portion of the cerebral cortex or the pontine reticular
formation of the brainstem (Figs. 7b and 8b). At 30
months of age, however, the immunoreaction with this
antibody was evident in the majority of large frontal
cortical pyramidal cells in the layer V (Fig. 7d) and
relatively large pontine reticular neurons (Fig. 8d).
The number of the positive frontal cortical pyramidal
cells in the layer V and pontine reticular neurons
counted through a drawing tube attached to the
micro2, respecscope
176 ±
61 and 66 ±24
cells/mm
tively. was
Double
immunostaining
revealed
that the appearance of cells positive for the C-terminal fragment
was highly consistent with the appearance of CE-posi-
FIG. 6. Immunocytochemical staining of (a) CE and (b) CD in
brainstem neurons of a 30-month-old rat. Gold particles indicating antigenetic sites of CE are seen predominantly in the lipofuscm. Bar = 0.65 tim.
tive cells in the cortical (Fig. 7c and d) and brainstem
(Fig. 8c and d) neurons. On the other hand, antiAPP
6O_ 100 antibody intensely stained the cortical and
brainstem neurons of 2- and 30-month-old rats, where
there was no obvious difference in staining patterns
between both groups (data not shown).
Immunoblot analyses of APP
To confirm the age-related neuronal accumulation
of the C-terminal fragment of APP, the tissue extract
J. Neurochem., Vol. 68, No. 2, 1997
746
H. NAKANISHI ET AL.
FIG. 7. Double-staining immunohistochemistry shows the colocalization of CE and APP in
cortical neurons of a 30-month-old rat. a and
b: Double-staining with anti-CE antibody and
anti-APP antibody in cerebral cortex of a 2month-old rat. c and d: Double-staining with
anti-CE antibody and anti-APP antibody in cerebral cortex of a 30-month-old rat. As shown
by arrowheads in c and d, most CE-positive
cortical cells were contained with APP-immunoreactivity. Bar = 100 ~sm.
of brainstems from embryonic and 2- and 30-monthold rats was subjected to SDS-polyacrylamide linear
gradient gel electrophoresis from S to 12% (wt/vol)
to separate full-length APP and APP fragments. In
the brainstem extracts from 2- and 30-month-old
rats, two immunoreactive bands of 19 and 22 kDa
corresponding to the C-terminal fragment of APP
were clearly detected with anti-APP643695 antibody,
and their levels greatly increased at 30 months of
age (Fig. 9). However, these protein bands were
FIG. 8. Double-staining immunohistochemistry shows
the colocalization of CE and APP in brainstem neurons of a 30-month-old rat. a and b: Double-staining
with anti-CE antibody and anti-APP antibody in
brainstem of a 2-month-old rat. c and d: Doublestaining with anti-CE antibody and anti-APP antibody in brainstem of a 30-month-old rat. As shown
by arrowheads in c and d, most CE-positive brainstem cells were contained with APP-immunoreactivity. Bar = 100 tim.
J. Neurochem., Vol. 68, No. 2, 1997
barely detectable at the embryonic stage. The discrepancy between immunohistochemical (Figs. 7
and 8) and immunoblotting data (Fig. 9) on the
young brainstem with regard to immunoreaction
with the antibody APP643 695 appeared to be due to
the difference in immunoreactivity between the natural and denatured proteins. Namely, it is likely that
APP in the young neurons exists mostly as the native
and normally processed forms that are unable to react with the antibody, probably owing to the steric
CATHEPSINS E AND D IN AGED RAT BRAIN
747
They have also shown that the thymocyte pro-CE was
converted into the mature form by in vivo dexamethasone administration, suggesting that the proenzyme is
transported from the primary storage sites, such as the
endoplasmic reticulum and the cis- or trans-Golgi, to
the functional sites, where it is processed to the catalytically active enzyme. Although the functional sites of
CE are currently unknown, the proteolytic maturation
of pro-CE appears to proceed within an acidic environment, because in the pulse—chase experiment in cultured rat microglia the proteolytic processing has been
shown to be blocked by an increase in the intracellular
pH with bafilomycin A1 and NH4C1 (D. F. Sastradipura
et al., unpublished data).
In the present study, we provide the first evidence
for the colocalization of CE with CD in the lipofuscincontaining lysosomes in the brainstem neurons of aged
FIG. 9. lmmunoblot analyses of APP in whole extracts of brain-
stem from rats of three different ages: embryonic day 18 (E), 2
months old (2M), and 30 months old (30M).
hindrance, whereas the denatured molecules
by
SDS-polyacrylamide gel electrophoresis become reactive to the antibody.
DISCUSSION
The present study clearly showed that CE and CD
were increasingly expressed in both the cerebral cortex
and the brainstem during the normal aging process and
that these changes were associated with senescence of
the brain. CD was exclusively found in the mature form
in both young and aged rat brain tissues. Although CE
was mainly present in the mature form in the young
rat brain tissues, a significant amount of pro-CE as
well as the predominant mature form was detected in
the aged brain tissues. These results suggest that CE
is increasingly expressed in these rat brain tissues with
aging and greatly processed to the mature CE. On the
other hand, CE in most rat tissues such as spleen and
stomach has been shown to be predominantly present
as the enzymatically inactive proform (Yonezawa et
al., 1993; Okamoto et al., 199S). This protein is also
known to exist in normal human erythrocyte membranes in the proform, which can be easily activated
by cell aging or oxidant challenge (Yamamoto et al.,
1989). Furthermore, the recombinant human CE expressed in Chinese hamster ovary cells (Tsukuba et
al., 1993) and mouse L cell fibroblasts and monkey
Cosl kidney cells (Finley and Kornfeld, 1994) was
primarily present as the proenzyme in the endoplasmic
reticulum. Finley and Kornfeld (1994) have suggested
that CE in the endoplasmic reticulum served as a reservoir that could be mobilized in response to certain
stimuli. More recently, Nishishita et al. (1996) reported that CE in rat thymus was exclusively confined
to thymocytes, in which it was present in the proform.
rats. Because CE predominantly existed as the mature
form in the brain tissues and cortical cell cultures, it
seems likely that the newly synthesized pro-CE in
these tissues is transported to the intracellular processing sites connected with its activation. Previous
immunoelectron microscopic studies have consistently
demonstrated that intracellular localization of CE in
nonneuronal cells was different from that of CD (Ichimaru et al., 1990; Saku et al., 1991; Solcia et al.,
1993), where the enzyme appeared to be present
mostly in the proform (Finley and Kornfeld, 1994).
In gastric and lymphoid cells, CD was exclusively confined to lysosomes, whereas CE was distributed diffusely in the cisternae of the rough endoplasmic reticulum and of the nuclear envelope and often in the cytoplasmic matrix. Therefore, CE in nonneuronal cells,
except for macrophages and microglia, is assumed to
be exclusively concentrated in the storage sites, but
the nature of the storage sites of CE remains to be
determined. In contrast, CE in neurons is likely to be
transported to the final processing acidic sites.
It is widely accepted that the initial step in the formation of lipofuscin is autophagocytosis of cytoplasmic
components. These autophagosomes fuse with primary
or secondary lysosomes to form autolysosomes. In the
next step, lipofuscin undergoes maturation to convert
into the characteristic fluorescent material. Therefore,
colocalization of CE with CD or lipofuscin suggests
the age-related alteration of the endosomal/lysosomal
system in the neurons, together with the association
with enhanced autophagy in the neuronal aging process. In addition to the immunocytochemical observation, the fusion of CE-containing organelles with lysosomes and the enhanced endosomal/lysosomal system
in the aged brain tissues are supported by the fact that
the amount of CE in the S fraction of the aged rat
brain tissues was very large, although the enzyme was
barely detectable in the S fraction of the young brain
tissues. This may also be related to the age-dependent
increase in the fragility of the endosomal/lysosomal
system. It has also been proposed that the increase in
J. Neurochem., Vol. 68, No. 2, 1997
748
H. NAKANISHI ET AL.
the cytosolic CD level in the aged brain is probably
due to disintegration of the lysosomal membrane (Matus and Green, 1987; Nakamura et al., 1989b). The
leakage of CE as well as CD into the cytoplasm of
neurons may be associated with age-related pathological changes because CE and CD have been shown
to be capable of degrading cytoskeletal proteins and
bioactive peptides (Matus and Green, 1987; Kageyama, 1993).
The present results also demonstrate that CE was
colocalized with the C-terminal APP fragments that
reacted with anti-APP643_695 antibody in the aged neurons. Immunoblot analysis revealed the age-related increase in content of two smaller proteins of APP, with
apparent molecular masses of 19 and 22 kDa, which
appeared to represent proteolytic C-terminal APP fragments. The size of these proteins strongly suggested
that they might contain an intact /3/A4 domain and
therefore be amyloidogenic. As APP has been shown
to accumulate in lipofuscin-containing lysosomes in
neurons from normal and Alzheimer’ s disease subjects
(Benowitz et al., 1989) and as CE was predominantly
found in lipofuscin-containing lysosomes of aged rat
neurons (Fig. 6), CE may be involved in the altered
APP catabolism to produce amyloidogenic C-terminal
APP fragments. Recently it has been suggested that
CD as well as CE efficiently cleaves the wild-type
APP, as well as a Swedish mutant, to yield amyloidogenic fragments (Ladror et al., 1994). Further study
will be needed to determine whether CE can process
full-length APP into small amyloidogenic fragments.
REFERENCES
Amano T., Nakanishi H., Kondo T., Tanaka T., Oka M., and Yamamoto K. (1995a) Age-related changes in cellular localization
and enzymatic activities of cathepsins B, L and D in the rat
trigeminal ganglion neuron. Mech. Ageing Dev. 83, 133—141.
Amano T., Nakanishi H., Oka M., and Yamamoto K. (1995b) Increased expression of cathepsins E and D in reactive microglial
cells associated with spongiform degeneration in the brain stem
of senescence-accelerated mouse. Exp. Neurol. 136, 171—182.
Azuma T., Pals G., Mohandas T. K., Couvreur J. M., and Taggart
R. T. (1989) Human gastric cathepsin E. Predicted sequence,
localization to chromosome 1, and sequence homology with
other aspartic proteinases. J. Biol. Chem. 264, 16748—16753.
Bahr B. A., Abai B., Gall C. M., Vanderklish P. W., Hoffman K. B.,
and Lynch G. (1994) Induction of f3-amyloid-containing polypeptides in hippocampus: evidence for a concomitant loss of
synaptic proteins and interactions with an excitotoxin. Exp. Neurol. 129, 81—94.
Banay-Schwartz M., DeGuzman T., Kenessey A., Palkovits M., and
Lajtha A. (1992) The distribution of cathepsin D activity in
adult and aging human brain regions. J. Neurochem. 58, 2207—
2211.
Beeson J. G., Shelton E. R., Chan H. W., and Gage F. H. (1994)
Age and damage induced changes in amyloid protein precursor
immunohistochemistry in the rat brain. J. Comp. Neurol. 342,
69—77.
Benowitz L. I., Rodriguez W., Paskevich P., Mufson E. J., Schenk
D., and Neve R. L. (1989) The amyloid precursor protein is
concentrated in neuronal lysosomes in normal and Alzheimer
disease subjects. Exp. Neurol. 106, 237—250.
J. Neurochem., Vol. 68, No. 2, 1997
Bernstein H.-G. and Wiederanders B. (1994) Animmunohistochemical study of cathepsin E in Alzheimer-type dementia brains.
Brain Res. 667, 287—290.
Bernstein H.-G., Bruszis S., Schmidt S., Wiederanders B., and Dorn
A. (1989) Immunodetection of cathepsin D in neuritic plaques
found in brains of patients with dementia of Alzheimer type.
J. Hirnforsh. 30, 613—618.
Cataldo A. M. and Nixon R. A. (1990) Enzymatically active lysosomal proteinases are associated with amyloid deposits in Alzheimer brain. Proc. Nati. Acad. Sci. USA 87, 3861 —3865.
Cataldo A. M., Paskevich P. A., Kominami E., and Nixon R. A.
(1991) Lysosomal hydrolases of different classes are abnormally distributed in brains of patients with Alzheimer disease.
Proc. Nati. Acad. Sci. USA 88, 10998—11002.
Cataldo A. M., Hamilton D. J., and Nixon R. A. (1994) Lysosomal
abnormalities in degenerating neurons link neuronal compromise to senile plaque development in Alzheimer disease. Brain
Res. 640, 68—80.
Cole G.M., Huynh T. V., and Saito T. (1989) Evidence for lysosomal processing of amyloid /3-protein precursor in cultured
cells. Neurochem. Res. 14, 933—939.
Finley E. M. and Komfeld 5. (1994) Subcellular localization and
targeting of cathepsin E. J. Biol. Chem. 269, 31259—31266.
Golde T. E., Estus S. S., Younkin L. H., Selkoe D. J., and Younkin
S. G. (1992) Processing of the amyloid protein precursor to
potentially amyloidogenic derivatives. Science 255, 728—730.
Haass C., Koo E. H., Mellon A., Hung A. Y., and Selkoe D. J.
(1992) Targeting of cell-surface /3-amyloid precursor protein
to lysosomes: alternative processing into amyloid-bearing fragments. Nature 357, 500—503.
Hajimohammadreza I., Anderson V. E. R., Cavanagh J. B., Seville
M. P., Nolan C. C., Anderton B. H., and Leigh P. N. (1994) /3Amyloid precursor protein fragments and lysosomal dense bodies are found in rat brain neurons after ventricular infusion of
leupeptin. Brain Res. 640, 25—32.
Hayashi Y., Kashiwagi K., and Yoshioka K. (1992) Protease inhibitors generate cytotoxic fragments from Alzheimer amyloid protein precursor in cDNA-transfected glioma cells. Biochem. Biophys. Res. Commun. 187, 1249—1255.
Ichimaru E., Sakai H., Saku T., Kunimatsu K., Kato Y., Kato I.,
and Yamamoto K. (1990) Characterization of hemoglobin-hydrolyzing acidic proteinases in human and rat neutrophils. .1.
Biochem. (Tokyo) 108, 1009—1015.
Ivy G. 0., Schottler F., Wenzel J., Baudry M., and Lynch G. (1984)
Inhibitors of lysosomal enzymes: accumulation of lipofuscinlike dense bodies in the brain. Science 226, 985—987.
Kageyama T. (1993) Rabbit procathepsin E and cathepsin E. Nudeotide sequence of eDNA, hydrolytic specificity for biologically
active peptides and gene expression during development. Eur.
J. Biochem. 216, 717—728.
Kenessey A. M., Banay-Schwartz M., DeGuzman T., and Lajtha A.
(1989) Increase in cathepsin D activity in rat brain in aging.
J. Neurosci. Res. 23, 454—456.
Ladror U. S., Snyder S. W., Wang G. T., Holzman T. F., and Krafft
G. A. (1994) Cleavage at the amino and carboxyl termini of
Alzheimer’s amyloid-/3 by cathepsin D. J. Biol. Chem. 269,
18422—18428.
Lowry 0. H., Rosebrough N. J., Fan A. L., and Randall R. J. (1951)
Protein measurement with the Folin phenol regent. J. Blot.
Chem. 193, 265—275.
Matus A. and Green G. D. J. (1987) Age-related increase in a cathepsin D like protease that degrades brain microtubule-associated proteins. Biochemistry 26, 8083—8086.
Nakamura Y., Takeda M., Suzuki H., Tada K., Hariguchi S., and
Nishimura T. (1989a) Age-related change in activities oflysosomal enzymes in rat brains. Mech. Ageing Dev. 50, 215—225.
Nakamura Y., Takeda M., Suzuki H., Tada K., Hariguchi S., and
Nishimura T. (1989b) Lysosomal instability in aged rat brain.
Neurosci. Lett. 97, 215—220.
Nakanishi H., Tsukuba T., Kondou T., Tanaka T., and Yamamoto
CATHEPSINS E AND D IN AGED RAT BRAIN
K. (1993) Transient forebrain ischemia induces increased expression and specific localization of cathepsins E and D in rat
hippocampus and neostriatum. Exp. Neurol. 121, 215—223.
Nakanishi H., Tominaga K., Amano T., Hirotsu I., Inoue Y., and
Yamamoto K. (1994) Age-related changes in activities and
Iocalizations ofcathepsins D, E, B, and L In the rat brain tissues.
Exp. Neurol. 126, 119—128.
Nishishita K., Sakai H., Sakai E., Kato Y., and Yamamoto K. (1996)
Age-related and dexamethasone-induced changes in cathepsins
E and Din rat thymic and spleniccells. Arch. Biochem. Biophys.
333, 349—358.
Nixon R. A. and Cataldo A. M. (1993) The lysosomal system on
neuronal cell death: a review. Ann. NYAcad. Sd. 679, 87—109.
Okamoto K., Yu H., Misumi Y., Ikehara Y., and Yamamoto K.
(1995) Isolation and sequencing of two clones encoding rat
spleen cathepsin E and analysis of the activation of purified
procathepsin E. Arch. Biochem. Biophys. 322, 103—111.
Robins E., Hiesh H. E., and Emmons S.S. (1968) Glycosidase in
nervous system I: assay, some properties and distribution of
/3-galactosidase, /3-glucuronidase and /3-glucosidase. J. Biol.
Chem. 243, 4246—4252.
Sakai H., Saku T., Kato K., and Yamamoto K. (1989) Quantitation
and immunohistochemical localization of cathepsins E and D
in rat tissues and blood cells. Biochim. Biophys. Acta 991, 367—
375.
Saku T., Sakai H., Shibata Y., Kato Y., and Yamamoto K. (1991)
An immunocytochemical study on distinct intracellular localization of cathepsin E and cathepsin D in human gastric and various
rat cells. J. Biochem. (Tokyo) 110, 956—964.
Siman R., Mistretta S., Durkin J. T., Savage M. J., Loh T., Trusko S.,
and Scott R. W. (1993) Processing of the /3-amyloid precursor;
multiple proteases generate and degrade potentially amyloidogenie fragments. J. Biol. Chem. 268, 16602—16609.
Sohal R. S. and Wolfe L. 5. (1986) Lipofuscin: characteristic and
significance. Prog. Brain Res. 70, 171—183.
Solcia E., Paulli M., Silini E., Fiocca R., Finzi G., Kindl S., Boveri
E., Bosi F., Cornaggia M., Capella C., and Samloff I. M. (1993)
Cathepsin E in antigen-processing Langerhans and interdigitat-
749
ing reticulum cells. Its possible role in antigen processing. Eur.
J. Histochem. 37, 19—26.
Takemura M., Nakamura S., Akiguchi I., UenoM., Oka N., Ishikawa
S., Shimada A., Kimura J., and Takeda T. (1993) /3/A4 proteinlike immunoreactive granular structures in the brain of senescence-accelerated mouse. Am. .1. Pathol. 142, 1887—1897.
Towbin H., Staehelin T., and Gordon J. (1979) Electrophoretic transfer ofproteins from polyacrylamide gels to nitrocellulose sheets:
procedure and some applications. Proc. Nati. Acad. Sci. USA
76, 4350—4354.
Tsukuba T., Hon H., Azuma T., Takahashi T., Taggart R. T., Akamine A., Ezaki M., Nakanishi H., Sakai H., and Yamamoto K.
(1993) Isolation and characterization of recombinant human
cathepsin E expressed in Chinese hamster ovary cells. J. Biol.
Chem. 268, 7276—7282.
Wacker W. E. C., Ulmer D. D., and Vallee B. L. (1956) Metalloenzymes and myocardial infarction. II. Malic and lactic dehydrogenase activities and zinc concentrations in serum. N. Engi. J.
Med. 55, 449—456.
Yamamoto K., Katsuda N., and Kato K. (1978) Affinity purification
and properties of cathepsin E-like acid proteinase from rat
spleen. Eur. I Biochem. 92, 499—508.
Yamamoto K., Katsuda N., Himeno M., and Kato K. (1979) Cathepsin D of rat spleen. Affinity purification and properties of two
types of cathepsin D. Eur. J. Biochem. 95, 459—467.
Yamamoto K., Takeda M., Yamamoto M., Tatsumi M., and Kato
Y. (1985) Human erythrocyte membrane acid proteinase
(EMAP). Sidedness and relation to cathepsin D. J. Biochem.
(Tokyo) 97, 821—830.
Yamamoto K., Yamada M., and Kato Y. (1989) Age-related and
phenylhydrazine-induced activation of the membrane-associated cathepsin E in human erythrocytes. J. Biochem. (Tokyo)
105, 114—119.
Yonezawa S., Tanaka T., and Miyauchi T. (1987) Cathepsin E from
rat neutrophils: its properties and possible relations to cathepsin
D-like and cathepsin E-like acid proteinases. Arch. Biochem.
Biophys. 256, 499—508.
Yonezawa S., Maejima Y., Hagioka N., Aratani T., Shoji R., Kageyama T., Tsukada S., Miki K., and Ichinose M. (1993) Changes
with development in the expression of cathepsin E in the fetal
rat stomach. Dev. Growth Differ. 35, 349—356.
.J. Neurochem., Vol. 68, No. 2, 1997