| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* Department of Clinical Chemistry, Kuopio University Hospital, FIN-70211 Kuopio, Finland;
Department of Pathology and
§ Division of Haematology/Oncology, Children’s Hospital Los Angeles Research Institute, USC School of Medicine, Los Angeles, California 90027, USA;
Department of Pathology and Forensic Medicine, Kuopio University, FIN-70210, Kuopio, Finland; and
|| Department of Clinical Chemistry and Hematology, Turku University Hospital, FIN-20521 Turku, Finland
1Correspondence: Department of Clinical Chemistry, Kuopio University Hospital, P.O. Box 1777, FIN-70211 Kuopio, Finland. E-mail: ilkka.mononen{at}messi.uku.fi
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: aspartylglycosylaminase • recombinant proteins • lysosomal storage diseases • animal disease models • lysosomes
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 2)
. Glycosylasparaginase (AGA;
aspartylglucosaminidase, EC 3.5.1.26) participates in stepwise
degradation of glycoproteins within lysosomes by hydrolyzing the
N-glycosidic carbohydrate-to-protein linkage. In AGU, the inherited
deficiency of glycosylasparaginase leads to accumulation of
aspartylglucosamine (GlcNAc-Asn) and other glycoasparagines in body
fluids and tissues starting in early fetal life. Although not proved,
it is generally accepted that the storage of aspartylglucosamine in
lysosomes results in this multisystem disease involving brain function.
AGU is characterized by a variety of clinical symptoms such as motor
and mental retardation, coarsening habitus, and shortened life span
(3)
. A mouse model of AGU that lacks any detectable AGA
activity has been generated by targeted disruption of the mouse
Aga gene in embryonic stem cells (4
5
6)
. The
AGU mice share most of the clinical, biochemical, and histopathologic
characteristics of the human disease.
Enzyme replacement therapy, i.e., replacement of the malfunctioning
enzyme by biologically active protein, has been tested as a therapeutic
approach in certain types of lysosomal storage diseases other than
glycoproteinoses. Enzyme replacement therapy with glucocerebrosidase is
routinely used to correct the metabolism of the accumulating
glycosphingolipid in the non-neuronopathic variant of Gaucher disease
in humans (7
, 8)
, but the therapy is not effective in
neuronopathic type II variant of Gaucher disease (9
, 10)
.
In mucopolysaccharidosis type VII (MPS VII) mice, reduction or
elimination of lysosomal storage in many tissues (including neurons)
has been achieved by intravenous (i.v.) injections of recombinant
ß-glucuronidase alone or in combination with bone marrow
transplantation (11
12
13)
. Long-term enzyme replacement
therapy was effective in non-neuronal tissues, but no improvement in
brain either in a canine model of MPS I (14)
or in a
feline model of MPS VI (15)
was reported.
So far, little is known about the efficacy of enzyme replacement
therapy in disorders of glycoprotein metabolism. Unlike the
accumulating substrates in glycosphingolipidoses and
mucopolysaccharidoses, the stored material (aspartylglucosamine) in AGU
is a small hydrophilic substance composed of a single amino acid and
sugar. Enzyme replacement with human recombinant glycosylasparaginase
efficiently corrects the aspartylglucosamine metabolism in cultured AGU
lymphoblasts and fibroblasts in vitro (16)
. The
therapeutic enzyme is targeted to the lysosomes through a
mannose-6-phosphate-mediated pathway, like several other lysosomal
enzymes (7
, 17
, 18)
.
Here we describe that adult AGU mice receiving i.v. injections of recombinant glycosylasparaginase rapidly restore aspartylglucosamine to normal levels in non-neuronal tissues. We also demonstrate that increased glycosylasparaginase activity and a decrease in the amount of the uncatabolized substrate are detected in brain tissue. The combined evidence suggests that enzyme replacement therapy with glycosylasparaginase has the potential to effectively treat AGU.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The mice, aged 9 to 19 months,
were wild-type Aga+/+, heterozygous
Aga+/-, or homozygous
Aga-/- (offspring of
Aga-/--
Aga-/- or
Aga +/-Aga+/-
matings) confirmed by Southern blot analysis of tail DNA samples.
Enzyme preparation
Human glycosylasparaginase was stably overexpressed in NIH-3T3
mouse fibroblasts (16)
. Recombinant human
glycosylasparaginase (isoform GA-1), which exposes mannose-6-phosphate
residues on its N-glycosidic carbohydrate chains and is actively
transported into AGU fibroblasts and lymphoblasts through
mannose-6-phosphate-mediated endocytosis, was purified from the
overexpressing cells as described (16)
. Specific activity
of the enzyme preparation was 336 mU/mg of protein.
Glycosylasparaginase activity was measured by a fluorometric assay
(19)
. One unit of enzyme leads to conversion of 1 µmol
of substrate/min under standard conditions. The enzyme preparation was
dialyzed into 150 mM NaCl, 10 mM Tris-HCl, pH 7.5, with 1 mM
ß-glycerophosphate, which was used as the injection buffer.
Treatment of animals and collection of samples
Animals were anesthetized for injection and sample collection
and handled according to institutional guidelines. The enzyme
preparation was injected into the tail vein of the mice in a volume of
100 µl. Two animals received a single 1 mg/kg injection, and blood
samples were collected 1 min, 5 min, 10 min, 15 min, 60 min, 12 h,
and 36 h after the injection. Four animals received 1 mg/kg and
another four mice received 10 mg/kg injections every second day. Urine
samples were collected on filter paper daily. Body weight and general
appearance of the mice in all groups were followed every day. All the
animals were anesthetized, killed, and perfused 36 h after the
last injection and tissues were collected for metabolite or
histological analysis. For tissue half-life of glycosylasparaginase,
six heterozygous mice received a single 1 mg/kg injection. Mice were
anesthetized, killed, and perfused 1, 6, 12, 24, 72, or 120 h
after injection and tissues were collected as described below.
Metabolite and protein analysis
The tissues were collected after intracardiac perfusion with
0.9% NaCl and homogenized in 50 mM Na-K-phosphate buffer, pH 7.5,
containing 0.1% Triton X-100. The aspartylglucosamine concentration
and glycosylasparaginase activity were analyzed by high-performance
liquid chromatography (HPLC) (20)
. For tissue half-life of
AGA, the enzyme activity in liver homogenates was measured by a
fluorometric method (19)
. The GlcNAc-Asn concentration in
urine samples collected on filter paper was measured as described
previously (21)
. Plasma was separated from heparinized
blood samples by centrifugation and AGA activity was measured by a
fluorometric method (19)
. Protein concentrations were
assayed by a Bio-Rad protein assay kit (Bio-Rad Laboratories, Richmond,
Calif.).
Histology
Animals were perfused with 2.5% glutaraldehyde in 0.1 M
phosphate buffer, pH 7.4, for electron microscopy. The tissues were
postfixed with 1% osmium tetroxide, dehydrated in ethanol, and
embedded in LX-112 resin. Ultra-thin sections were cut and
double-stained with uranyl acetate and lead citrate. Transmission
electron micrographs were taken on a Jeol JEM-1200EX electron
microscope (Jeol, Chicago, Ill.). For immunohistochemical studies, the
animals were perfused with saline; tissues were removed, formalin
postfixed for 48 h, and paraffin embedded. Sections (4 µm) were
immunostained with a polyclonal chicken antibody (1:300) directed
against human AGA peptide (16)
or rabbit immunoserum
(1:1000) directed against human AGA protein using the avidin-biotin
complex method (Vectastain Elite ABC Kit, Vector, Burlingame, Calif.).
Secondary antibody (polyclonal anti-chicken IgG, Zymed, South San
Francisco, Calif.) was used at a dilution of 1:200 for the peptide
antibody. Sections were counterstained with hematoxylin. Omission of
the antibody served as a negative control.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Tissue half-life of AGA was estimated to be 1.9 days in
liver (Fig. 1B
). Thirty six hours after the injection (1
mg/kg), AGA activity was almost in the normal range in the liver; in
the spleen it was 10–30% of that found in the wild-type tissue
(Table 1
).
|
|
AGA activity comparable to that in the wild-type tissues was measured
in the liver and spleen of the mice after eight 1 mg/kg injections
given every second day (Table 1)
. The enzyme activity in other tissues
remained low or undetectable (Table 1)
. A marked increase in enzyme
activity was seen in all the examined non-neuronal tissues with a
10-fold higher enzyme dose (8x10 mg/kg) (Table 1)
. Very high levels of
AGA activity were reached in the liver and spleen, somewhat higher than
normal activity was measured in the jejunum and heart, and enzyme
activity close to that of wild type was detected in the kidney and lung
tissues of the treated animals. The enzyme activity in brain tissue
increased to 10% of that in the wild-type tissue.
Effects of glycosylasparaginase on aspartylglucosamine in
tissues
The exogenous enzyme functioned by clearing accumulated
glycoasparagines from tissues. A single injection of the enzyme at a
dose of 1 mg/kg drastically reduced the amount of GlcNAc-Asn by
80–90% in the spleen and liver (Fig. 2
). At the same time, the amount of storage material was reduced in
jejunum by ca. 50%, by 30% in lung, and by 15% in heart (Fig. 2)
.
After eight injections of glycosylasparaginase (1 mg/kg), the metabolic
defect was completely corrected and the stored GlcNAc-Asn was
eliminated from the liver (Fig. 2)
. Accumulated aspartylglucosamine was
also reduced efficiently in all other tissues except the brain. In the
treated animals, the amount of GlcNAc-Asn had declined by 80–90% in
the spleen, jejunum, and kidney and by 60–65% in the heart and lung
(Fig. 2)
. After eight injections of the enzyme at a dose of 10 mg/kg,
the liver, heart, kidney, jejunum, and spleen tissues were completely
cleared of GlcNAc-Asn; only trace amounts of the compound were detected
in the lung tissue, and in brain tissue lysosomal storage decreased by
20% (Fig. 2)
.
|
Effects of glycosylasparaginase on excretion of aspartylglucosamine
in urine
The enzyme injections resulted in a rapid reduction of the
GlcNAc-Asn excretion in urine (Fig. 3
). This effect was dose dependent: the higher the enzyme dose, the lower
the excretion. A single dose of either 1 mg/kg or 10 mg/kg of human
recombinant AGA lowered the excretion of the storage compound by more
than 35% at 24 h after injection. The first three or four enzyme
injections every second day progressively lowered the excretion of
GlcNAc-Asn; after that it plateaued, even though further enzyme
injections were administered. With the smaller enzyme dose (8x1
mg/kg), the excretion of GlcNAc-Asn plateaued at a mean level of 108
µmol per mmol of creatinine (range 68–151 µmol GlcNAc-Asn per mmol
of creatinine; n=15), which corresponds to ~40% of that
in the untreated AGU mice (mean 256 µmol and range 204–307 µmol
GlcNAc-Asn per mmol of creatinine; n=8) (Fig. 3)
. The
therapeutic protocol with the higher enzyme dose (8x10 mg/kg) reduced
the mean excretion of GlcNAc-Asn into urine to 24 µmol GlcNAc-Asn per
mmol of creatinine (range 11–51 µmol GlcNAc-Asn per mmol of
creatinine; n=12), which corresponds to ~9% of that in
untreated animals (Fig. 3)
.
|
Tissue pathology
Tissues of AGU mice were examined after injections of AGA (8x10
mg/kg every second day) by light and electron microscopy. The massive
lysosomal distention of hepatocytes, sinusoidal lining cells, and
Kupffer cells in the liver tissue of untreated animals was markedly
reduced during the treatment (Fig. 4A, B
). The finely granular, amorphous
storage material in intracellular vacuoles typical of AGU had
disappeared, and only single electron-dense membranous structures
against a clear background were left in the vacuoles of the liver
tissue of the treated animals (Fig. 4A, B
). The epithelial
and meseangial cells of renal glomeruli appeared free from storage
vacuoles in the treated mice, in contrast to their variable
vacuolization in the untreated tissues (Fig. 4C
,
D). The same applied to proximal and distal tubules and
collecting tubes of kidney (data not shown). In the spleens of the
treated animals, sinus lining cells and macrophages were reduced in
size and contained very few enlarged lysosomes (Fig. 4E, F
).
In the spleens of untreated mice, lymphoid and other cells were only
slightly affected and showed a normal appearance in the spleens of
treated animals. Immunostaining with AGA antibodies was extensive and
widely distributed in the liver of treated animals. The most intensive
staining was seen in the sinusoidal lining cells, whereas staining of
hepatocytes was less intensive (data not shown). In the spleen of the
treated animals, lymphoid cells and macrophages in the red pulp were
strongly positively stained (Fig. 5A, B
). AGA immunostaining in the brain of the treated animals
did not demonstrate consistent accumulation of the enzyme protein to
any particular cell type or structure, including meninges and blood
vessels (data not shown).
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and the availability of a mouse model for
AGU (4
5
6)
have made enzyme replacement therapy an
attainable goal for treatment of AGU. Glycosylasparaginase
administrated i.v. to the enzyme-deficient mice was rapidly cleared
from the systemic circulation in two phases
(t1/2=4 min and t1/2=39
min). The first rapid phase probably represents distribution of the
enzyme within the vascular space or extracellular fluids; the second,
slower phase probably represents uptake of the enzyme by tissues mostly
through mannose-6-phosphate-mediated endocytosis (16)
.
Similar bimodal decrease in plasma activity has been reported with
Hurler syndrome (14)
. The enzyme functioned in tissues by
degrading accumulated glycoasparagines, as demonstrated by the
disappearance of aspartylglucosamine from them. The estimated half-life
of AGA in liver tissue was ~2 days. Corresponding tissue half-lives
have been reported for recombinant human
N-acetylgalactosamine-4-sulfatase in cats (15)
and for
recombinant ß-D-glucuronidase in MPS VII mice (22)
.
The response of enzyme activity in tissues to an increase in enzyme
dose appeared not to be linear with the dosage schedules used.
Injection of 8 x 1 mg/kg of AGA increased the liver and spleen
tissue activity of the enzyme by 2–5-fold compared to single dose (1
mg/kg) of AGA, but in several tissues the enzyme activity remained
undetectable. This and the presence of undegraded aspartylglucosamine
in the tissues indicate that the enzyme dosage of 8 x 1 mg/kg was
too low for efficient enzyme replacement therapy in AGU mice. After a
10-fold increase in the amount of enzyme delivered (8x10 mg/kg vs.
8x1 mg/kg), the final tissue activity levels were highly increased in
liver and spleen and essentially normalized in all non-neuronal tissues
analyzed. This and the complete clearance of aspartylglucosamine from
other non-neuronal tissues except lung, in which a trace amount of the
compound was left, indicate effective correction of the glycoprotein
degradation defect in non-neuronal tissues of AGU mice with the high
enzyme dose. Eight i.v. injections of the corrective enzyme (10 mg/kg)
into adult mice during a 2 wk period increased enzyme activity in brain
tissue to 10% of that present in the wild-type tissue. With a 10%
increase in enzyme activity the amount of aspartylglucosamine was
decreased by 20%. Microscopic examination demonstrated that tissue
pathology was effectively corrected in the spleen and kidney of the AGU
mice given the high enzyme dosage. Microscopic evaluation of liver, the
most severely affected organ in AGU mice, demonstrated less but not an
absent vacuolization in hepatocytes of animals treated with 8 x
10 mg/kg of glycosylasparaginase. Biochemical analyses indicated rapid
disappearance of aspartylglucosamine from the liver tissue during the
therapy, suggesting that histological changes in the organ are
corrected more slowly than the biochemical defect. This is probably
related to the severity of the liver damage, which eventually leads to
massive coagulative hepatic necrosis in many terminally ill AGU mice
(6)
. Reduction of lysosomal storage in the brain tissue
could not be localized to any particular region, including meninges on
histopathologic analysis of the treated mice. Immunostaining with AGA
antibodies did not localize accumulation of the enzyme to any
particular cell type or structure, including meninges or blood vessels,
suggesting that the enzyme was rather diffusely distributed in the
brain. This widespread distribution in enzyme activity within brain
tissue is further supported by the significant decrease in the storage
of aspartylglucosamine in the tissue. This finding is interesting
compared to other enzyme replacement studies performed in adult animals
with types of lysosomal diseases other than glycoproteinoses (14
, 15
, 23)
. Additional long-term enzyme replacement experiments are
required to study whether the reduction in lysosomal storage of
aspartylglucosamine in AGU mice brain will improve the mental function
of the animals. In MPS VII mice, improvements in the behavioral and
auditory performance in addition to improved histopathologic and
biochemical findings have been achieved by enzyme replacement therapy
(24)
.
In AGU, the storage material aspartylglucosamine is excreted in urine in large amounts, and urine analysis of the GlcNAc-Asn can be used in the diagnosis of the disease. Detection of GlcNAc-Asn in urine also seems to be useful for monitoring the therapeutic response in the disease, since it indicates the level of glycoprotein degradation in the whole body. The urine analyses demonstrated that the therapeutic effect on GlcNAc-Asn excretion was achieved with the first three to four enzyme injections every second day, but the next four to five injections maintained the response. The excretion of GlcNAc-Asn into urine was not corrected to normal, i.e., <1 µmol GlcNAc-Asn per mmol of creatinine, which suggests that aspartylglucosamine continues to leak into the systemic circulation from tissues that are not readily accessible to the enzyme, and the metabolite escapes undegraded into urine due to the short half-life of the enzyme in the systemic circulation.
Here we demonstrated using quantitative biochemical assays that enzyme
replacement therapy restores the normal AGA activity and reverses the
pathology in many important somatic AGU tissues. Also, the activity of
the corrective enzyme in brain tissue increased to a higher extent than
expected. The results are encouraging when keeping in mind that this
study was performed with adult mice over a 2 wk period of time. It
cannot be ruled out that these rapid therapeutic effects could somehow
be related to the type of the basic metabolic defect—disorder of
glycoprotein degradation—in AGU in contrast to more widely studied
lysosomal diseases involving disorders in mucopolysaccharide or
glycosphingolipid catabolism. It remains to be investigated whether the
effects of external enzyme in the central nervous system of AGU mice
would be more marked with other dosage schedules or when the therapy is
initiated with newborn mice with an immature blood–brain barrier and
an early stage of the disease, as shown with MPS VII mice
(22)
.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-L-iduronidase, by cultured human fibroblasts. Cell 12,619-627[Medline]
-glucosidase leads to uptake of enzyme in heart and skeletal muscle of mice. J. Clin. Invest. 87,513-518
This article has been cited by other articles:
|
|
 |
|
  J. Blanz, S. Stroobants, R. Lullmann-Rauch, W. Morelle, M. Ludemann, R. D'Hooge, H. Reuterwall, J. C. Michalski, J. Fogh, C. Andersson, et al. Reversal of peripheral and central neural storage and ataxia after recombinant enzyme replacement therapy in {alpha}-mannosidosis mice Hum. Mol. Genet., November 15, 2008; 17(22): 3437 - 3445. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Tomatsu, A. M. Montano, A. Ohashi, M. A. Gutierrez, H. Oikawa, T. Oguma, V. C. Dung, T. Nishioka, T. Orii, and W. S. Sly Enzyme replacement therapy in a murine model of Morquio A syndrome Hum. Mol. Genet., March 15, 2008; 17(6): 815 - 824. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. H. Grubb, C. Vogler, B. Levy, N. Galvin, Y. Tan, and W. S. Sly Chemically modified {beta}-glucuronidase crosses blood-brain barrier and clears neuronal storage in murine mucopolysaccharidosis VII PNAS, February 19, 2008; 105(7): 2616 - 2621. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Urayama, J. H. Grubb, W. A. Banks, and W. S. Sly Epinephrine enhances lysosomal enzyme delivery across the blood brain barrier by up-regulation of the mannose 6-phosphate receptor PNAS, July 31, 2007; 104(31): 12873 - 12878. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. S. Sly, C. Vogler, J. H. Grubb, B. Levy, N. Galvin, Y. Tan, T. Nishioka, and S. Tomatsu Enzyme therapy in mannose receptor-null mucopolysaccharidosis VII mice defines roles for the mannose 6-phosphate and mannose receptors PNAS, October 10, 2006; 103(41): 15172 - 15177. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Vogler, B. Levy, J. H. Grubb, N. Galvin, Y. Tan, E. Kakkis, N. Pavloff, and W. S. Sly From The Cover: Overcoming the blood-brain barrier with high-dose enzyme replacement therapy in murine mucopolysaccharidosis VII PNAS, October 11, 2005; 102(41): 14777 - 14782. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. Matzner, E. Herbst, K. K. Hedayati, R. Lullmann-Rauch, C. Wessig, S. Schroder, C. Eistrup, C. Moller, J. Fogh, and V. Gieselmann Enzyme replacement improves nervous system pathology and function in a mouse model for metachromatic leukodystrophy Hum. Mol. Genet., May 1, 2005; 14(9): 1139 - 1152. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Kelo, U. Dunder, and I. Mononen Massive accumulation of Man2GlcNAc2-Asn in nonneuronal tissues of glycosylasparaginase-deficient mice and its removal by enzyme replacement therapy Glycobiology, January 1, 2005; 15(1): 79 - 85. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. S. Sands, C. A. Vogler, K. K. Ohlemiller, M. S. Roberts, J. H. Grubb, B. Levy, and W. S. Sly Biodistribution, Kinetics, and Efficacy of Highly Phosphorylated and Non-phosphorylated beta -Glucuronidase in the Murine Model of Mucopolysaccharidosis VII J. Biol. Chem., November 9, 2001; 276(46): 43160 - 43165. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) and those treated with 1 x 1 mg/kg
(
), 8 x 1 mg/kg (
), or 8 x 10 mg/kg (
) of
glycosylasparaginase. The GlcNAc-Asn amount was measured by HPLC. An
asterisk indicates GlcNAc-Asn levels below the detection limit of the
assay (0.3 nmol/mg cell protein). Each column represents the mean value
of homogenates of two different animals, and the bars represent ±
range of the mean values.
) of glycosylasparaginase/kg of animal
weight was injected into the tail vein of AGU mice every second day for
15 days. Urine samples were collected on filter paper and the amount of
excreted GlcNAc-Asn was measured by HPLC. Each data point represents
the mean value of samples of three different animals in both groups.
The bars represent ± 1 SD of the values. The mean
GlcNAc-Asn concentration in urine of untreated AGU mice is 256
µmol/mmol of creatinine (range 204–307 µmol/mmol of creatinine;
n=8) (gray area).
