(Received for publication, March 31, 1997, and in revised form, May 20, 1997)
From the Lysosomal acid phosphatase (LAP) is a
tartrate-sensitive enzyme with ubiquitous expression. Neither the
physiological substrates nor the functional significance is known. Mice
with a deficiency of LAP generated by targeted disruption of the LAP
gene are fertile and develop normally. Microscopic examination of
various peripheral organs revealed progredient lysosomal storage in
podocytes and tubular epithelial cells of the kidney, with regionally
different ultrastructural appearance of the stored material. Within the central nervous system, lysosomal storage was detected to a regionally different extent in microglia, ependymal cells, and astroglia concomitant with the development of a progressive astrogliosis and
microglial activation. Whereas behavioral and neuromotor analyses were
unable to distinguish between control and deficient mice, ~7% of the
deficient animals developed generalized seizures. From the age of 6 months onward, conspicuous alterations of bone structure became
apparent, resulting in a kyphoscoliotic malformation of the lower
thoracic vertebral column. We conclude from these findings that LAP has
a unique function in only a subset of cells, where its deficiency
causes the storage of a heterogeneously appearing material in
lysosomes. The causal relationship of the enzyme defect to the clinical
manifestations remains to be determined.
Lysosomal acid phosphatase (LAP1; EC
3.1.3.2) is an orthophosphoric monoesterase of the endosomal/lysosomal
compartment (1, 2). Like the highly homologous secretory acid
phosphatase of the prostate (3), LAP is sensitive to inhibition by
L-tartrate (4). In vitro, LAP cleaves various
phosphomonoesters (e.g. adenosine monophosphate,
thymolphthalein phosphate, and glucose 6-phosphate) at an acidic pH
(3.5-5.0) (4, 5). Despite the pivotal role LAP has played in the
discovery and histochemical visualization of lysosomes (6), neither its
in vivo substrates nor its functional role is known.
The biosynthetic route of LAP is quite unusual since it is synthesized
and transported as an integral type I membrane glycoprotein with a
large luminal domain containing the active site of the enzyme, a single
transmembrane domain, and a short cytoplasmic domain (7). The LAP
precursor is initially transported to the plasma membrane, from where
it is rapidly internalized by receptor-mediated endocytosis via
clathrin-coated pits (8, 9). Before its eventual delivery to the
lysosomes, the membrane-bound LAP precursor repeatedly recycles from
the early endosomes to the cell membrane (8). The cytoplasmic domain of
the LAP precursor harbors a tyrosine-containing internalization signal
(10, 11) as well as a signal for basolateral sorting (12). The
membrane-bound LAP precursor, which forms homodimers in
vivo, is converted to soluble mature LAP after delivery to the
lysosomal compartment by limited proteolysis (13).
In situ hybridization showed an almost uniform expression of
LAP in all organs of the mouse, except for the testis and brain, which
stand out through their considerably higher level of expression. In the
testis, the higher expression is restricted to spermatocytes, whereas
in the brain, expression in cerebellar Purkinje cells and telencephalic
pyramidal neurons as well as in epithelial cells of the choroid plexus
predominates (14).
Besides tartrate-sensitive LAP, the tartrate-resistant type 5 acid
phosphatase (TRAP), another acidic orthophosphoric monoesterase, has
been identified in the lysosomal compartment (15). TRAP enzyme activity
has been detected in the mononuclear phagocyte system, i.e.
alveolar macrophages and osteoclasts (16). The functional relationship
between both phosphatases is still unknown.
To elucidate the physiological role of LAP, we have inactivated the
corresponding gene in the mouse by targeted disruption. In this study,
we describe the phenotype of LAP-deficient mice, which is characterized
by lysosomal storage phenomena in the kidney and central nervous
system, progressive skeletal disorder, and an increased disposition
toward generalized seizures.
A C57BL/6J mouse cosmid library (Stratagene, La Jolla, CA)
was screened with a full-length mouse LAP cDNA clone (17). The isolated mouse LAP cosmid clone mLAP1 contained all 11 exons of the
gene as well as several kilobase pairs of the 5
The recombination construct (pLAP-9HIIIneo; see Fig.
1A, panel II) was linearized with XhoI
and introduced into the ES cell line E-14-1 by electroporation. ES
cells were cultured as described previously (19, 20). G418-resistant
colonies were screened by Southern analysis of genomic DNA digested
with SstI or XbaI and probed with probe A (see
Fig. 1A, panels I and III). Mutated ES
cells were microinjected into blastocysts of C57BL/6J mice as described
(20). The resulting male chimeras were mated with C57BL/6J females.
Mice were genotyped for the introduced LAP gene mutation by Southern
analysis of SstI-digested genomic DNA using probe A (see
Fig. 1A, panels I and III). The mice
were kept in a conventional animal facility at the Zentrum für
Biochemie und Molekulare Zellbiologie, Universität
Göttingen (Göttingen, Germany).
Total RNA from the livers and
kidneys of 3-month-old mice was prepared as described by Chirgwin
et al. (21). Ten micrograms of total RNA were separated on a
formaldehyde-agarose gel and processed as described (22). Filters were
hybridized with the mouse LAP cDNA (17) and a 280-base pair
cDNA fragment from glyceraldehyde-3-phosphate dehydrogenase (23).
Hybridization and washing of filters were performed as described
(10).
Spleen and liver tissues were homogenized in 10 mM Tris-HCl and 150 mM NaCl, pH 7.4 (w/v, 1:3),
at 4 °C. Homogenates were adjusted to 0.1% Triton X-100, sonicated
for 30 s on ice, and centrifuged at 12,000 × g
for 5 min at 4 °C. The protein concentration of the supernatant was
determined (24). Aliquots of the supernatant were incubated with 100 µl of 10 mM p-nitrophenyl phosphate in 0.1 M sodium citrate, pH 4.5, and either 50 µl of 40 mM NaCl or 50 µl of 40 µM tartrate. After
incubation for 20 min at 37 °C, the reaction was stopped by the
addition of 0.5 ml of 0.4 M glycine/NaOH, pH 10.4. Absorbance at 405 nm was determined. The tartrate-inhibitable LAP
activity was determined as described (1).
For standard
light microscopic histology and immunohistochemistry, brains were fixed
by cardiac perfusion with 4% paraformaldehyde in 0.1 M
phosphate buffer and shock-frozen or fixed by perfusion with Bouin's
solution diluted 1:3 with 0.1 M phosphate buffer and
embedded in paraffin. Series of frozen sections were processed for
lectin histochemistry using Griffonia simplicifolia
agglutinin (GSA) coupled to peroxidase (Sigma, Deisenhofen, Germany) at
a concentration of 5 mg/ml, followed by enzyme histochemistry of peroxidase according to Adams (25). Alternating sections were immunohistochemically reacted with monoclonal antibodies against lamp-1
(lysosome-associated membrane
protein-1; 3 mg/ml; Developmental Studies
Hybridoma Bank, University of Iowa) or cathepsin D-specific antiserum
(26) in a 1:1000 dilution in phosphate buffer following the
peroxidase/anti-peroxidase method of Sternberger (27). Similarly, antibodies (mono- and polyclonal) against tyrosine hydroxylase (Chemicon International, Inc.), glutamate decarboxylase (a gift from W. Oertl), phosphorylated and dephosphorylated neurofilaments (Sternberger-Meyer Immunocytochemistry Inc.), GAP-43 (Sigma), GFAP
(glial fibrillary acidic
protein) (Sigma), S100 (DAKO, Hamburg, Germany), tenascin
(a gift from M. Schachner), c-Fos (Oncogene Science Inc.), and
Ulex europaeus agglutinin I lectin (Sigma) were applied
using standard histochemical techniques.
To visualize the distribution of some antigens with better tissue
preservation, paraffin sections were dewaxed and reacted with
antibodies against GFAP, lamp-1, cathepsin D, and MHC II (Pharmingen,
Hamburg) and the microglia-specific lectins GSA-I-B4 and
Ricinus communis agglutinin I (RCA-I), the latter one
predominantly reacting with activated microglia. Binding of both
lectins and primary antibodies was detected by tyramide amplification
(reagents purchased from NEN Life Science Products) developed by
peroxidase histochemistry (28, 29).
For electron microscopy, the fixative was modified by adding 0.5%
glutardialdehyde, while the histochemical procedures remained unchanged. Finally, tissue sections were treated with 2% osmium tetroxide, dehydrated, and embedded in Epon 812. For standard transmission electron microscopy without the application of
histochemical procedures, animals were perfused with 6%
glutardialdehyde, and tissues were embedded in Araldite.
Mice were briefly narcotized to facilitate
radiography. For comparative purposes, high resolution Eastman Kodak
mammography x-ray film was employed, and identical radiation energy (35 kV) and photographic exposures were used to prepare radiographs on separate portions of the same film.
A mouse genomic DNA clone covering the entire
LAP gene was isolated from a C57BL/6J mouse cosmid library (for
details, see "Experimental Procedures"). The target construct
pLAP-9HIIIneo (Fig. 1A, panel II)
was used for disruption of the LAP gene in ES cells. The open reading
frame of the gene is interrupted in exon 7, and the truncated open
reading frame encodes 243 of 423 amino acids of wild-type LAP. Since
conversion of histidine 256 of human LAP, which is conserved among
lysosomal, prostatic, yeast, and Escherichia coli acid
phosphatases and was proposed to be part of the active sites of these
enzymes (30), to serine yielded a catalytically inactive
LAP,2 the introduced mutation is most
likely a null mutation.
pLAP-9HIIIneo was introduced into E-14-1 ES cells (31), and
G418-resistant colonies were analyzed by Southern blotting. In 3 out of
532 independent clones tested, an additional SstI DNA
restriction fragment was detected with probe A (Fig. 1A), indicating a homologous recombination event in one of the LAP alleles
(Fig. 1B, panel I). This was confirmed by probing
XbaI-digested DNA with probe A (Fig. 1B,
panel II). The targeted ES cell clones were microinjected
into C57BL/6J blastocysts. The resulting chimeras were used to generate
heterozygous and subsequently homozygous mutant offspring in an outbred
129SVJ-C57BL/6J genetic background.
Heterozygous offspring identified by hybridization of
SstI-digested DNA with probe A (Fig. 1C) did not
show differences in phenotype or fertility as compared with wild-type
littermates (data not shown). Genotyping of 92 offspring from
heterozygous crosses revealed a frequency of 24% for homozygous mutant
mice, closely resembling the expected mendelian frequency (25%).
Hence, disruption of the LAP gene does not result in embryonic
lethality.
To test for expression of LAP at RNA and enzyme activity
levels, Northern blot analyses and acid phosphatase enzyme assays were
performed. A 3.2-kilobase LAP-specific RNA hybridized in liver RNA from
a wild-type animal, whereas no LAP transcripts were detectable in liver
RNA from a homozygous mutant animal (Fig. 2A). For determination of LAP enzyme
activity, spleen and kidney homogenates were incubated with
p-nitrophenyl phosphate in the presence and absence of
L-tartrate, a competitive inhibitor of LAP. LAP-specific,
L-tartrate-sensitive enzyme activity was undetectable in
homogenates from homozygous mutant animals (Fig. 2B).
Northern blot analyses as well as LAP enzyme activity assays showed
that the LAP gene had been inactivated and that homozygous mutant mice were devoid of lysosomal acid phosphatase activity.
To identify phosphorylated
storage material in LAP-deficient cells, embryonic fibroblasts were
metabolically labeled with [32P]orthophosphate. After
subcellular fractionation by Percoll density centrifugation, the
fractions enriched in markers for lysosomes or endosomes were subjected
to analysis for low molecular weight, water-soluble material by gel-
and ion-exchange chromatography; for polypeptides by two-dimensional
electrophoresis; and for lipids by extraction with chloroform/methanol
and thin-layer chromatography. None of these experimental procedures
revealed the accumulation of 32P-labeled material in
LAP-deficient fibroblasts (data not shown).
Homozygous mutant, heterozygous, and wild-type mice
resulting from heterozygous crosses did not exhibit differences in
growth and weight development (data not shown). LAP-deficient mice were fertile and did not show an elevated mortality up to an age of 18 months. Gross anatomy and both light and electron microscopic observations of tissues such as liver, spleen, muscle, retina, and
thyroid gland of 3-9-month-old mice showed no significant differences
compared with control mice. After the age of 6 months, alterations of
bone structure became apparent, resulting in a kyphoscoliotic
malformation of the lower thoracic and lumbar vertebral column (see
Fig. 7).
Among the peripheral organs
analyzed, lysosomal storage was observed only in the kidney.
Ultrastructural examination revealed enlarged lysosomes in some
podocytes (Fig. 3A) and those portions of
Henle's loop that are located within the inner stripe of the outer
medulla, i.e. the uppermost portion of the thin descending limbs (Fig. 3, B and D) and the thick ascending
limbs. In wild-type mice, the podocytes (data not shown) and the cells
of Henle's loop (Fig. 3C) show only very few small
lysosomes. In LAP-deficient mice, the lysosomes were increased in size
and number and contained multilamellar or polymorphic material.
When analyzed with a
variety of standard techniques such as Nissl and hematoxylin/eosin
staining and incubation with antibodies to neuronal growth cones,
synaptophysin, phosphorylated and nonphosphorylated neurofilaments,
dopaminergic and GABAergic neurons and olfactory axons, S100 protein
(glial and neuronal marker), and tenascin (predominately a glial
product), no differences between LAP-deficient and control brains were
observed (data not shown) in brain sections of mice 3-11 months of
age. To provide a first survey of lysosomal alterations, these
organelles were detected by immunohistochemical labeling of the
corresponding marker antigens lamp-1 and cathepsin D. The
immunoreactivity for lamp-1 was evenly distributed at a comparatively
low level in control mice (Fig. 4, A,
C, and E), whereas antibodies to cathepsin D
strongly labeled the somata of all kinds of neurons (shown for control
cerebral cortex in Fig. 4G).
In LAP-deficient mice, strongly lamp-1-immunoreactive glial cells
appeared in deep layers of the cerebral cortex (Fig. 4B). At
higher magnification (shown for the striatum in Fig. 4, C
and D), it was apparent that the weak staining of glial
cells in control mice was replaced by a strong staining of the cell
bodies of glial cells in LAP-deficient mice. In the cerebella of
LAP-deficient mice, a prominent staining for lamp-1 was seen in
Bergmann glial cells, subpopulations of astrocytes in the granular
layer, and microglial cells in the granular and molecular layers (Fig.
4F), whereas in control mice, immunoreactivity for lamp-1
was weak and mainly confined to Bergmann glia (Fig. 4E). In
addition, in LAP-deficient mice, antibodies to lamp-1 strongly labeled
ependymal cells of the ventricular lining with a gradient of decreasing intensity from the spinal chord central canal to the hemispheral ventricles (data not shown).
In LAP-deficient mice, immunoreactivity for cathepsin D as a second
lysosomal marker protein was enhanced in most brain regions due to
staining of enlarged microglial and astroglial cell bodies and globular
structures within their processes (data not shown). Interestingly,
cathepsin D immunoreactivity tended to be reduced in neurons (Fig.
4H).
Routine immunohistochemistry for the glial marker GFAP revealed a
localized astrogliosis restricted to those brain regions that showed
increased lamp-1 staining. Thus, strongly GFAP-positive astroglia cells
were found in the deep layers (V/VI) of the isocortex, basal ganglia,
and cerebellar Purkinje cell layer (data not shown).
Electron microscopic analysis of LAP-deficient mice showed that the
majority of astrocytes, cerebellar Bergmann glial cells, and ependymal
cells contained numerous storage vacuoles, which could be identified as
lysosomes by staining with lamp-1 (shown for astroglia in Fig.
5B) and cathepsin D (data not shown)
antibodies. Conventional electron microscopy revealed that the storage
vacuoles either appeared electron-lucent or were filled with material
of variable electron density (shown for Bergmann glia in Fig.
5D and ependymal cells in Fig. 5F). In control
mice, lysosome-like structures were barely detectable in astrocytes,
Bergmann glia, and ependymal cells (Fig. 5, A, C,
and E). Lysosomal storage in glial cells of LAP-deficient
brains appeared to be a progredient process and also exhibited a
gradient of decreasing severity from distal to proximal parts of the
brain.
Similar to astroglia, microglial cells exhibited conspicuous
alterations within LAP-deficient brains. Thus, in control mice, staining of microglial cells with B4 isolectin from
G. simplicifolia seeds (GSA) gave a rather uniform staining
pattern throughout the brain (Fig. 6A). This
Up-regulation of GSA-binding sites is regarded as one of many
parameters characterizing microglial cell activation. Binding sites for
the lectin RCA-I, which are present only at low levels in resting
microglia, were strongly expressed by microglia of LAP-deficient mice
in the same brain regions that were distinguished by the most intense
up-regulation of GSA-binding sites. Thus, cells labeled by RCA-I were
scattered e.g. along the isocortical lamina V and the
cerebellar Purkinje cell layer (data not shown).
In younger LAP-deficient animals, both lectins stained cells of the
already activated, but still ramified phenotype. A change in microglial
morphology toward an epithelioid (phagocytic) morphology was observed
at later stages (>8 months), which was then encompassed by the
expression of MHC II antigens by a subset of these cells. However,
other markers of microglial activation (NADH diaphorase and
nitrogen-monoxide synthetase) were not elevated in deficient animals
(data not shown).
Among 270 LAP-deficient mice, 19 animals (7%) repeatedly presented epileptic seizures at >8 weeks of
age. These seizures were generalized and tonic-clonic in nature. They
were triggered by the weekly cage changing and followed by a short
period of excitement caused by placement of the animal in a new
environment. The convulsions lasted between 5 and 30 s, during
which the animal lost consciousness and balance. The seizures were
followed by 5-15 min of postepileptic depression, in which the animal
did not respond to external stimuli. Increased susceptibility to
auditory or electric shock provocation was not
observed.3 In about the same number of
wild-type mice with the same genetic background, seizures were not
observed. Behavioral tests including the open-field test (33), rotarod
test, wire test, and hot-plate test (34) performed at an age of 3 months failed to distinguish between LAP-deficient and wild-type
control mice (data not shown).
LAP-deficient
animals of later adult stages developed conspicuous and progressive
alterations of bone structure already visible in x-ray imaging whereby
lesions were most prominent in vertebrae and long bones of the upper
and lower limbs (shown for 6- and 15-month-old animals in Fig.
7). Specifically, areas of compact bone as well as the
endplates of vertebral bodies appeared less electron-dense, whereas
regions of cancellous bone (again most notably within vertebral bodies)
featured an abnormally coarse and locally hyperdense meshwork of mostly
longitudinally oriented trabeculations. As a consequence, the clear-cut
radiographic differences between cancellous and compact bone were
almost lost.
Concomitant to this apparent reduction in calcium salts within compact
bone, a progressive change in bone morphology was noted, which was most
prominent in thoracic and lumbar vertebrae. Thus, at the age of 15 months (Fig. 7, C-F), the individual vertebral bodies were
thickened along their dorsoventral axis and also exhibited a reduced
craniocaudal height (Fig. 7, E and F, short
arrows and arrowheads). Vertebral arches had stunted
pediculi that extended over almost the entire length of the vertebral
bodies (Fig. 7, E and F, long arrows),
thereby leading to a dramatic reduction in the size of the
intervertebral foramina, possibly causing a generalized compression of
spinal nerves (Fig. 7F). Within the thoracic vertebral
column, animals of this age exhibited a scoliotic malformation with an
augmented lower cervical lordosis and lower thoracic kyphosis.
Interestingly, the alterations described developed comparatively late.
Thus, at the age of 6 months (Fig. 7, A and B),
only a slight augmentation of the lordotic and kyphotic bends of the
lower cervical and thoracic vertebral column was noted.
Even though L-tartrate-sensitive LAP is the
enzyme that led De Duve (35) to the discovery of lysosomes, its
in vivo functions remained elusive until today. The gene
coding for LAP has been shown to be a housekeeping gene, and LAP
activity has been found in all tissues and nucleated cells that have
been examined (1, 14). If LAP is critical for the catabolism of one or
several of its phosphorylated substrates, we reasoned that a deficiency of LAP would cause the accumulation of these substrates in lysosomes. In analogy to the >30 different lysosomal storage disorders caused by
the inactivity of one or several lysosomal proteins (36), we expected
that this accumulation would lead to the morphological, biochemical,
and clinical manifestation of a lysosomal storage disorder.
In the early seventies, patients in two unrelated families were
described that were characterized by a complete deficiency of
phosphatase activity in lysosomes. The familial occurrence of the
disease and the intermediate phosphatase activities in the parents
suggested an autosomal recessive inheritance of the disease and an
essential role of an acidic phosphatase in lysosomal catabolism (37,
38). The relation of the acid phosphatase deficiency to the clinical
manifestations (progressive lethargy, opisthotonus, bleeding, and death
within the first year of life) and the molecular basis remained
speculative, and no further patients have been described.
Among the tissues from LAP-deficient mice investigated, lysosomal
storage phenomena were observed only in podocytes and distinct segments
of the nephron in the kidney as well as microglia and different
subtypes of macroglial cells within the brain. In all cases, the
morphology of the stored material was heterogeneous, possibly
indicating the presence of different substrates in different cells,
which in turn would concur with the broad spectrum of substrates cleaved by LAP in vitro. All other tissues were free of
morphological signs of lysosomal storage. Moreover, the search for
storage material in LAP-deficient fibroblasts with biochemical methods
remained elusive. This indicates that LAP is not critical for the
lysosomal catabolism in these tissues either because the substrates of
LAP have a tissue-specific occurrence or because the function of LAP can be passed by alternative lysosomal phosphatases or transporters that allow the exit of phosphorylated compounds from the lysosomal compartment.
Storage vacuoles were found
in astroglia, cerebellar Bergmann glia, ependymal cells, and microglial
cells. They were identified at the ultrastructural level as lysosomes
by their immunoreactivity for the lysosomal membrane marker lamp-1 and
the lysosomal matrix marker cathepsin D. The material contained in
these storage lysosomes was of heterogeneous morphological appearance
and remains to be identified. It is not clear whether the stored
material originates from the glial cells or is taken up from the
surrounding tissue. It should be noted that the storage in astrocytes
and microglial cells is not uniform, but shows clear regional
differences.
In addition to lysosomal storage, the microglial cells display
characteristics of a progressive activation, i.e. the
enhanced expression of binding sites for GSA and RCA-I and
up-regulation of the MHC II molecules during later stages of
development. At present, it is not known if the lysosomal storage
per se or if signals from the surrounding milieu trigger the
observed microglial activation.
The lysosomal storage in restricted cells
might functionally be related to neuronal and skeletal alterations in
LAP-deficient mice. The molecular correlation between the observed
functional changes and the LAP deficiency is still unknown and deserves
further investigations.
Contrasting to the moderate extent of neuropathologic damage observed
in LAP-deficient mice, these animals repeatedly presented generalized
tonic-clonic seizures at >8 weeks of age. The low penetrance of these
seizures (~7% of LAP-deficient mice) may be linked to the genetic
background of the animals investigated. It has been described earlier
that C57BL/6J mice, which provide half of the 129SVJ-C57BL/6J outbred
background used in this study, are very resistant to epileptic seizures
at all ages (39, 40). The causal relationship between the predominantly
glial pathology and the development of epileptic seizures remains
unexplained. The astroglia has repeatedly been involved in the
spreading of epileptic potentials by its ion transport and buffering
capabilities (41). Thus, one might speculate that the lysosomal storage
impairs the buffering capacity of glial cells and thereby facilitates generation of seizures. Furthermore, activated microglia may contribute to ictogenesis through secretion of neurotoxic agents that increase glutamate receptor-mediated excitotoxicity (42).
For most of the clinical
manifestations in lysosomal storage disorder, the link between the
molecular defect and the phenotypic alteration is not understood. This
applies also for the alterations of bone structure observed in
LAP-deficient mice, of which the decreased mineralization observed from
the age of 6 months is the most notable feature. According to data from
human bone pathology, a loss of calcium salts from bones has to be
quite substantial (~30%) to be detected by standard radiographic
procedures as employed here (43). It is therefore likely that the
decreased mineralization of compact bone and vertebral endplates would
be detectable much earlier with more sensitive methods. The coarse
structure of cancellous bone especially noted in vertebrae can be
interpreted as a compensatory reaction to the apparent loss of minerals
from the compacta and the resulting destabilization. The foreshortening
of individual bones indicates that this compensation remains
inadequate.
Contrasting to frequently observed perturbations of enchondral
ossification in lysosomal storage diseases (44), alterations of
enchondral ossification have not been detected so far in LAP-deficient animals, and skeletal abnormalities were found only at later stages. A
defective enchondral ossification with delayed mineralization of the
cartilage was observed in 6-8-week-old mice with a deficiency of TRAP
(45). TRAP is also localized in the lysosomal compartment (15). The
expression of TRAP, however, is limited to the mononuclear phagocyte
system (16). In older mice with a TRAP deficiency, an osteopetrosis
with increased mineralization of bone tissue was observed.
The different phenotypes of LAP- and TRAP-deficient mice clearly
indicate that the two lysosomal phosphatases are critical for the
catabolism of distinct substrates. Mice deficient in both LAP and TRAP,
however, display a severe phenotype that is more than a mere addition
of clinical signs seen in LAP- and TRAP-deficient mice from which the
LAP/TRAP-deficient mice were generated by mating,4 e.g. cells that appear
normal in mice with a deficiency of either LAP or TRAP, such as Kupffer
cells of liver, show excessive lysosomal storage. This strongly
suggests that for a number of substrates, LAP and TRAP can substitute
for each other. Storage of such substrates is therefore only observed
if both lysosomal phosphatases are deficient.
We thank N. Hartelt, H. Boetcher, A. Wolff,
and B. Rühlke for excellent technical assistance and W. Schmahl
and C. Rahner for performing analyses of various tissues of
LAP-deficient animals. We are grateful to W. Danysz (Merck, Frankfurt,
Germany) for performing behavioral studies and to W. Löscher for
examining epileptic seizures in LAP-deficient mice. We thank R. Kunze
for two-dimensional lipid analysis of LAP-deficient fibroblasts.
Zentrum Biochemie und Molekulare
Zellbiologie,
Zentrum Anatomie,
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
Isolation of a Genomic Clone and Construction of a Target
Vector
- and 3
-flanking regions. DNA sequence analyses revealed complete sequence identity of
exon sequences to the nucleotide sequence of the mouse LAP cDNA
used for screening (data not shown). For construction of a target
vector, a 9-kbp HindIII DNA restriction fragment of mLAP1 covering 3.7 kbp of 5
-untranslated region and exons 1-8 (see Fig.
1A, panel I) was subcloned into the plasmid
vector pBluescript SK II+ (Stratagene). The neo
expression cassette from pMC1neopA (Stratagene) (18) was inserted as a
BglII DNA restriction fragment into a BamHI
restriction site located in exon 7 of the LAP gene. The insertion of
the neo cassette introduces a premature translational stop
codon into the open reading frame of the LAP gene.
Fig. 1.
Disruption of the LAP gene. A,
strategy for disruption of the LAP gene by homologous recombination in
mouse ES cells. Panel I, genomic structure and partial
restriction map of the LAP gene. Exons are numbered and indicated by
open boxes. Exon 7, used for interruption of the open
reading frame, is indicated by the closed box. Introns and
5- and 3
-flanking regions are indicated by the solid line.
The black bar designated probe A denotes the DNA
probe used for Southern blot analyses. Panel II, the target
vector pLAP-9HIIIneo with 9.0-kilobase homology to the LAP gene locus.
The neo cassette (open box) was inserted into a
BamHI restriction site located in exon 7; the
arrowhead marks the direction of transcription of the
neo gene. The broken line indicates the plasmid
vector pBluescript SK II+. Panel III, predicted
LAP gene locus after homologous recombination. B,
BamHI; H, HindIII; S,
SstI; X, XbaI. B, Southern
blot analyses of ES cell clones. Panel I, probe A was
hybridized to SstI-digested DNA from wild-type E-14-1 ES
cells and two ES cell clones (EL-5 and EL-178) with a targeted allele
as indicated by an additional 4.6-kbp DNA fragment. Panel
II, probe A was hybridized to XbaI-digested DNA from
the same ES cells (as in panel I), confirming homologous recombination in EL-5 and EL-178 as indicated by an additional 4.2-kbp
fragment. C, Southern blot analysis of tail DNA. DNA was digested with the restriction enzyme SstI and hybridized
with probe A (wild-type allele: 3.4 kbp; mutant allele: 4.6 kbp). +/+, homozygous wild-type; +/
, heterozygous;
/
, homozygous
mutant.
[View Larger Version of this Image (37K GIF file)]
Generation of Mice with Targeted Disruption of the Lysosomal Acid
Phosphatase Gene
Fig. 2.
The LAP gene is inactivated in homozygous
mutant mice. A, Northern blot analysis of LAP expression.
Total RNA (10 µg) was hybridized using a LAP mouse cDNA probe
(17) and a murine glyceraldehyde-3-phosphate dehydrogenase
(G3PD) probe (internal control) (23). LAP mRNA (3.2 kilobases) was absent in the livers and kidneys of homozygous mutant
mice (results shown for the liver only). B, LAP enzyme
activity. Spleen and kidney homogenates from homozygous mutant mice
(/
) and wild-type controls (+/+) were incubated with
p-nitrophenyl phosphate in the presence and absence of 50 mM L-tartrate. The tartrate-inhibitable enzyme
activity was determined. No LAP enzyme activity was found in
LAP-deficient tissues (n = 2; range is given).
[View Larger Version of this Image (17K GIF file)]
Fig. 7.
Radiologic examination of LAP+/+
and LAP/
mice at 6 (A and B)
and 15 (C-F) months of age. A and B,
after an apparently normal development of the skeletal system,
pathological alterations were first noted at the age of 6 months, when
LAP-deficient mice (B) exhibited an augmented lower thoracic
kyphosis and an upper thoracic lordosis of the vertebral column
(arrowheads in B) as compared with controls
(A). At this stage, the overall bone structure appears
otherwise normal. C-F, until 15 months of age,
LAP-deficient mice (D and F) have developed a
considerable S-shaped deformation of the entire thoracic vertebral
column and severe alterations of bone structure. Specifically,
individual vertebral bodies feature an increased sagittal diameter
(short arrows in E and F), whereby individual vertebral bodies exhibit a reduced axial height (indicated by arrowheads for LIII). In LAP-deficient mice, the pediculi
of vertebral arches extend over almost the entire length of the corpus vertebrae (long arrows), thus leading to a dramatic
reduction in size of the intervertebral foramina (small
arrows pointing to intervertebral foramen LIV) and possibly to a
compression of the corresponding spinal nerves. Control mice are shown
in C and E.
[View Larger Version of this Image (106K GIF file)]
Fig. 3.
Lysosomal storage in LAP-deficient kidney.
A, podocyte with a storage vacuole (arrow)
containing flocculent material. B, epithelial cell of an
intermediate tubule. The lysosomes contain multilamellar and granular
materials (arrows). C and D, light micrographs of intermediate tubules of a control (C) and a
LAP-deficient (D) mouse. The storage lysosomes appear as
intensely staining cytoplasmic inclusions (arrows).
Bars = 0.5 µm (A), 0.25 µm
(B), and 10 µm (C and D).
[View Larger Version of this Image (187K GIF file)]
Fig. 4.
Enhanced immunoreactivity of LAP-deficient
glial cells for lamp-1 and cathepsin D. Shown is the
immunohistochemical distribution of the lysosomal marker lamp-1 in
parasagittal sections of control (A, C, and
E) and LAP-deficient (B, D, and
F) mouse brain. A and B, cerebral
cortex. In superficial cortical layers of normal brains (A),
the immunoreactivity for lamp-1 is weak, whereas in the LAP-deficient
cerebral cortex (B), the glial cells of the lower half of
the cortex show a strong labeling for lamp-1. C and
D, striatum. At higher magnification, only some glial cells of the normal striatum (C) are faintly stained, whereas in
the LAP-deficient striatum (D), the number of strongly
stained glial cells, both microglia and astroglia, is higher than in
controls. E and F, cerebellar cortex. In the
cerebellar cortex of control brains (E), immunoreactivity is
concentrated in the Purkinje cell layer (pl). Here the
neurons (marked by arrows) are less stained than cell bodies
of Bergmann glia (bg). In LAP-deficient mice (F),
Bergmann glial cells (arrows), subpopulations of astrocytes (a) in the granular layer (gl), and microglial
cells (m) in the granular and molecular layers are strongly
labeled. G and H, cerebral cortex. The
immunoreactivity for cathepsin D in laminae V and VI of the dorsal
isocortex is shown for controls (G) and LAP-deficient mice
(H). Note the intense staining of neuronal perikarya in
controls, which is almost lost in LAP-deficient animals.
Bars = 100 µm (A and B), 20 µm
(C and D), 50 µm (E and
F), and 200 µm (G and H).
[View Larger Version of this Image (106K GIF file)]
Fig. 5.
Lysosomal storage in LAP-deficient macroglial
cells. Shown are electron micrographs of macroglial cells from
control (A, C, and E) and
LAP-deficient (B, D, and F) mice.
A and B, astroglial cells. Immunoelectron
microscopy for lamp-1 reveals labeling of a lysosome-like structure
(arrow) in the control astrocyte (A), but a
strong labeling of storage vacuoles (arrows) in the
LAP-deficient astrocyte (B). C and D,
Bergmann glial cells. Conventional electron microscopy of Bergmann
glial cells demonstrates storage vacuoles (arrowheads) in
LAP-deficient animals (D), which were absent in controls
(C). The storage vacuoles are partly filled with
heterogeneously appearing material. E and F,
ependymal cells. Conventional electron microscopy of ependymal cells
reveals lysosomal storage (arrow) in LAP-deficient ependymal
cells (F), which was not found in controls (E).
Bars = 1 µm (A and B), 0.5 µm
(C and D), and 1.5 µm (E and F).
[View Larger Version of this Image (204K GIF file)]
-D-galactosyl-specific lectin has been shown to
selectively label endothelia and microglial cells in rodents (32). In
control mice, the lectin outlined the extensive processes of microglial
cells (Fig. 6C). Electron microscopy revealed that binding
of GSA was confined to the cell surface and a few lysosome-like
structures (Fig. 6E, arrows). A remarkably
enhanced staining was found when brain sections of LAP-deficient mice
were stained with GSA. In LAP-deficient mice, the staining intensity of
microglial cells was increased in almost the entire brain. The GSA
staining pattern closely mimicked that obtained with lamp-1. As for
astroglia, the deep isocortical laminae, basal ganglia, cerebellum, and
brain stem were most intensely labeled, whereas other areas such as the
superficial layers of the neocortex (laminae I-II), the first layer of
the paleocortex, and the dorsal segment of the hippocampus (Fig.
6B, stars) remained unchanged with respect to
control mice. The increased staining intensity of microglial cells was
due to an enhanced labeling of their cell bodies (Fig. 6D),
which were filled by GSA-positive vesicles (Fig. 6F,
arrows). Conventional electron microscopy showed that the
vacuoles were filled with material of variable electron density, which
apparently was lost during the sample preparation for lectin staining
(data not shown). As revealed by immunohistochemistry for lamp-1 and cathepsin D, the vacuolar storage granules of microglial cells from
LAP-deficient mice (shown for cathepsin D in Fig. 6, G and H, arrows) were of lysosomal origin.
Fig. 6.
Microglial cells in LAP-deficient mice
contain cytoplasmic vacuoles that stain with GSA and cathepsin D. A and B, parasagittal sections of 3-month-old
mouse brains. At low magnification, a uniform and weak staining is
observed with GSA throughout the control brain (A), with
somewhat stronger staining along marginal and paraventricular brain
surfaces. In a 3-month-old LAP-deficient mouse brain (B),
enhanced GSA staining is based on a spotty staining pattern in most
brain parts. Zones with weak staining are labeled (*) in superficial
cortex layers. The strongest staining is observed in the anterior parts
of the striatum (s) and cerebral cortex (c) as
well as in some parts of the hippocampus (h). C
and D, deep cortical layers. At higher magnification, GSA
labels normal microglial cells at variable density in deep cortical
layers (C). At higher magnification in LAP-deficient brains
(D) (picture shows a comparable part of the deep cortical
layer in C), the cell bodies of almost all microglia cells
are strongly labeled with GSA, whereas labeling appears equal or even
reduced in many microglial cell processes. E and
F, microglial cells. Electron microscopy shows that in
control animals (E), GSA labels selectively the surface of
microglial cells and a few vacuolar, lysosome-like structures (arrows). Electron microscopy of microglial cells of
LAP-deficient mice (F) shows a prominent GSA labeling of
polymorphic vacuole membranes, whereas labeling of the cell surface is
sparse or absent. Labeled vesicular structures are indicated by
arrows. G and H, microglial cells.
Immunoelectron microscopy for cathepsin D shows that in controls
(G), weak labeling of the microglial cell surface (arrows) is observed, whereas in LAP-deficient cells
(H), labeling of the limiting membrane of cytoplasmic
vacuoles is prominent and that of the cell-surface is scarce.
Bars= 1 mm (A and B), 10 µm
(C and D), 1 µm (E and
F), and 0.5 µm (G and H).
[View Larger Version of this Image (112K GIF file)]
Deficiency of LAP Causes Tissue-specific Lysosomal
Storage
*
This work was supported in part by Deutsche
Forschungsgemeinschaft Grant Pe 310/6-1 and by the Fonds der Chemischen
Industrie.The costs of publication of this
article were defrayed in part by the
payment of page charges. The article
must therefore be hereby marked
"advertisement" in
accordance with 18 U.S.C. Section
1734 solely to indicate this fact.
§
To whom correspondence should be addressed. Tel.: 49-551-395932;
Fax: 49-551-395979; E-mail: saftig{at}uni-bc2.gwdg.de.
**
Present address: Medizinische Molekularbiologie, Innere Medizin I,
Universitätsklinik Freiburg, 79106 Freiburg, Germany.
Present address: Hoffmann-La Roche, 4058 Basel,
Switzerland.
§§
Supported by a Federation of European Biochemical Societies
short-term fellowship. Present address: Nencki Inst. of Experimental Biology, Warsawa 02-093, Poland.
1
The abbreviations used are: LAP, lysosomal acid
phosphatase; TRAP, tartrate-resistant type 5 acid phosphatase; kbp,
kilobase pair(s); ES, embryonic stem; GSA, G. simplicifolia
agglutinin; RCA, R. communis agglutinin.
2
R. Pohlmann, unpublished data.
3
W. Löscher and W. Danysz, personal
communication.
4
A. Suter and P. Saftig, unpublished data.
©1997 by The American Society for Biochemistry and Molecular Biology, Inc.