| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
-Glucosidase
From the Divisions of Human Genetics*
and
Pathology,
Children's Hospital Medical
Center, Cincinnati, Ohio, and the Department of Internal
Medicine,
New York University Medical Center,
New York, New York
| Abstract |
|---|
|
|
|---|
-glucosidase (GAA) cleaves the
1-4 and
1-6 glycosidic
linkages of glycogen and related
-glucosyl substrates within
lysosomes. Its deficiency results in glycogen storage disease type II
(GSDII) variants including Pompe disease. To gain insight into the
tissue patterns of involvement by glycogen storage in GSDII,
GAA mRNA expression in mouse tissues was evaluated by Northern blot and
in situ hybridization analyses. Extensive temporal and
spatial variation of GAA mRNA was observed. During preterm
maturation, GAA mRNA levels of whole mice progressively
increased as assessed by Northern analysis. By in situ
hybridization with GAA antisense mRNA, low signals were
detected in most tissues throughout gestation. However,
increased expression in specific cell types of different tissues was
observed beginning at 16 days post coitum in developing brain
neurons, primitive inner ear cells, and seminiferous
tubular epithelium. In adult mice, whole-organ GAA mRNA levels
were highest in brain, moderate in heart,
liver, and skeletal muscle, and lowest in the series
kidney > lung > testis > spleen. By in
situ hybridization, the highest-intensity signals were in
neurons of the central and peripheral nervous systems whereas
neuroglial cells had only low-level signal. Signals of moderate
intensity were in cardiomyocytes whereas low signals were in
hepatocytes and skeletal muscle myocytes and very low in cells of the
lungs, thymus, pancreas, spleen, and
adrenal glands. However, testicular Sertoli cells and kidney
tubular epithelial cells had significant signals even though
surrounding cells had very low signals. The discrete temporal and
spatial variations of GAA mRNA during development indicate different
physiological roles for this enzyme in various cell types and
developmental stages.
| Introduction |
|---|
|
|
|---|
-glucosidase or acid maltase
(GAA; EC 3.2.1.3) is an essential lysosomal enzyme that cleaves
1-4 and
1-6 glycosidic bonds of glycogen, maltose, and
intermediate oligosaccharides yielding glucose.1,2
Recessively inherited GAA deficiencies lead to glycogen storage
disease type II (GSDII) variants that are characterized by lysosomal
glycogen accumulation in many tissues (reviewed in Ref. 3
). The GSDII
clinical variants differ in age of onset, tissue involvement, and
severity. In the fatal infantile variant, Pompe disease, progressive
massive glycogen accumulation is predominant in heart, skeletal muscle,
and liver and leads to death from cardiorespiratory failure by 1 or 2
years of age1
(reviewed in Ref. 3
). Glycogen accumulation
occurs in neurons and glial cells of the central nervous system (CNS)
and Schwann cells of the peripheral nervous system (PNS) of the
patients.4-8
These pathological findings have not been
emphasized recently but suggest a neural component to the Pompe disease
pathogenesis. Later-onset forms of the disease are clinically
heterogeneous and present as muscular dystrophies primarily affecting
skeletal muscles.1,3,9,10
In these cases, the slower
progressive glycogen accumulation leads to a limb-girdle-like myopathy
with minor cardiac involvement. Frequently, death results from
respiratory failure. Although the pathophysiological bases of these
diseases are not well understood, most signs and symptoms have been
attributed to primary muscle deterioration and dysfunction related to
glycogen accumulation. The human GAA gene maps to chromosome 17q25,11 spans ~20 kb, and has 20 exons.12 Characterization of its promoter region revealed characteristics of a housekeeping gene.13,14 The human cDNA, ~3.4 kb in length, encodes a 952-amino-acid polypeptide.13,15,16 A high degree of conservation between mouse and human cDNAs is evidenced by a 75% nucleotide identity and 79% identity of their predicted amino acid sequences. Numerous mutations of the GAA gene have been described, and many result in absent or abnormal mRNAs.17-19 In general, correlation exists between residual levels of enzyme activity and phenotype.3 Infantile-onset variants of GSDII have low to undetectable enzyme activity levels in skeletal muscle and cultured skin fibroblasts, whereas in later-onset variants these are higher.
GAA enzyme activity levels vary among tissues and throughout development.5,20-24 In human fetuses, GAA activity peaks toward the end of gestation in many tissues (liver and kidney > skeletal muscle and lung > spleen and heart).22 The brain has the lowest enzyme activity and this remains relatively constant throughout prenatal development. In adult humans, the highest GAA activities are in the kidney and prostate.20 Relative to kidney GAA activity, that in liver, spleen, and adrenal glands is ~10% to 30%, and that in skin, skeletal muscle, and heart is ~2% to 5%. From pathological studies in Pompe disease patients, the greatest glycogen storage is present in liver, spleen, and adrenals that also have low enzyme activity.22 In mice, the highest GAA activity levels are in intestine, brain, liver, kidney, and testis with only ~10% to 20% of these levels present in lungs and cardiac and skeletal muscle and intermediate levels in the thymus.21,24 To date, GAA knockout mouse models have variably mild phenotypes, but glycogen storage is predominant in skeletal and cardiac muscle.24,25
To gain insight into the physiological role of GAA, tissue-specific pathology in GSDII deficiency, and pathogenesis of the disease, the GAA mRNA expression was characterized in murine tissues and cells. These studies show discrete temporal and cellular specific GAA mRNA expression in regions of the CNS and some non-neural tissues.
| Materials and Methods |
|---|
|
|
|---|
A 1.36-kb 32P-labeled mouse GAA cDNA fragment (nucleotides 634 to 1997 from the ATG) was hybridized to BALB/c adult mice poly A+ RNA (2 µg) from a variety of tissues (Clontech Laboratories, Palo Alto, CA). Hybridization with a human ß-actin cDNA probe was used for RNA quality control. Quantitation of probe binding was by scanning densitometry of autoradiograms of 24 hours of exposure (LKB, Pharmacia, Uppsala, Sweden).
In Situ Hybridization Analysis
Mouse GAA 35S-labeled sense and antisense riboprobes were synthesized by in vitro transcription26 from linearized templates containing ~1 kb (nucleotides 634 to 1635 from the ATG). In situ hybridization of B6C3F1/J (C57BL/6J x C3H/HeJ) mouse tissues (Harland Animal, Indianapolis, IN) was performed essentially as described.26 Briefly, cryosections of 4% paraformaldehyde-fixed and embedded tissues were post-fixed and prepared for in situ hybridization as described.26 The sections were first hybridized with [35S]UTP-labeled 1-kb sense or antisense GAA riboprobes under high-stringency conditions in a mixture containing 50% formamide. Hybridization was followed by ribonuclease A/T1 digestion and washes under progressively higher-stringency conditions, including 0.1x SSC at 55°C. This was followed by dehydration in graded ethanol solutions, dipping in Kodak NTB2 emulsion, and exposure at 4°C for 10 to 15 days. Slides were developed and stained with hematoxylin and eosin (H&E). Positive signal, obtained with the antisense probe, appears as white or light pink grains under dark-field microscopy. Duplicate sections hybridized with the sense probe were used as negative controls.
| Results |
|---|
|
|
|---|
Differential expression of GAA mRNA was observed by Northern blot
analysis and in situ hybridization. Multiple Northern blots
with mRNA from whole tissues showed a single 3.8-kb GAA band with the
highest levels in the brain (100%), intermediate levels in skeletal
muscle, liver, and heart (50% to 35%), and low to very low levels in
lung and kidney (15%) (Figure 1A)
. In
spleen and testis (3% to 1.6%), 10 times longer exposures were
required to develop signals of similar intensity to those of lung and
kidney (not shown). Rehybridization of this membrane to a cDNA
ß-actin probe verified integrity of the mRNA samples. The
characteristic ß-actin 2.0-kb band was present in all tissues at
relatively high levels (Figure 1B)
. Faster migrating bands of ~1.8
kb, prominent in heart and skeletal muscle, are due to hybridization of
the probe to the
or
forms of actin (Figure 1B)
.
|
|
|
In the kidney, a gradient of signal was detected in the tubular
epithelial cells with low signals in the renal cortex and barely
detectable signals in the medulla (Table 1)
. In the intestine, GAA mRNA
signals were undetectable in duodenal, jejunal, or ileal cells even
after long-term 20- to 30-day exposures. Using similar exposure times,
the colon showed very low-level signals in stromal cells of the lamina
propria (Table 1)
. In addition, low-level signals were detected in
neurons of the intestinal ganglia (Table 1)
. Differential expression
was apparent in the liver. Low-intensity GAA mRNA signals were in
hepatocytes, but no signals were appreciated in cells of the portal
ducts or sinusoids (Table 1)
. The stomach showed low signals throughout
(Table 1)
. GAA mRNA signals in cells of other visceral organs,
including the lungs, thymus, pancreas, adrenal glands, and spleen, were
slightly above background levels (Table 1)
.
Analysis of cardiac and skeletal muscles showed uniformly distributed
low signals in myocytes (Figure 2, IL)
. Prolonged exposures (20 to 30
days) were needed to discern distinct signals. Cardiac myocytes of the
atria and ventricles showed low-intensity signals (Figure 2J)
. No
significant signals above background were detected in cardiac
interstitial or valvular cells (Table 1)
. Skeletal muscle evaluation
included sections of the quadriceps, anterior tibialis, triceps,
biceps, diaphragm, tongue, and paraspinal (Figure 2, K and L)
and
extraocular muscles. Uniformly distributed low-intensity signals were
detected in all muscle groups. No qualitative or quantitative
differences were observed regionally within a particular muscle or
among different muscle groups (Table 1)
.
Developmental Studies
A single specific signal of 3.8 kb was detected throughout
gestation (days 11, 15, and 17) by Northern hybridization of poly
A+ RNA of whole-mouse embryos (not shown). The levels of
mRNA at days 15 and 17 were three to four times greater than that at
day 11. By in situ hybridization, mouse embryo sagittal
sections at days 9.5, 12, 14, and 16 of gestation showed ubiquitous
low-level GAA mRNA expression with slightly increasing overall levels
at day 16 (Table 2)
. By gestational day
16, differential expression in specific cells was evident. GAA mRNA
signals of moderate and high intensity were present in clusters of
differentiating neurons of the hypothalamic region (Figure 2M)
and
epithelial cells lining the inner ear (Figure 2N)
. Higher-intensity
signals also were found in neurons of the peripheral sympathetic
ganglia (Figure 2O)
and in cells lining the
developing seminiferous tubules of the testis (Figure 2P)
. The
epithelial cells lining the developing intestinal tract had low-level
signals. Skeletal muscles showed low-level expression in most skeletal
muscle groups. Higher-intensity signals were present in developing
skeletal muscles from head and neck regions (Table 2)
. In the
developing heart, low- and very low-intensity GAA mRNA signals were
seen in myocytes of atria and ventricles (Table 2)
.
|
| Discussion |
|---|
|
|
|---|
An objective of this work was to evaluate the potential for correlations between GAA mRNA expression in particular cell types and the phenotypic involvement resulting from GAA deficiency in mice and humans. This hypothesis would be based on the commonly held concept that sites of highest expression directly correlate with sites of major pathology. Such correlation could not be established due to extreme variation in mRNA signal in tissues affected by the disease. In general, the low-level mRNA expression in normal murine cardiac and skeletal myocytes is consistent with the low GAA activities in normal mice and also in humans.20-22,24,25 However, both cell types have major phenotypic involvement in the murine and human GSDII diseases.3,24,25 Of possible significance, glycogen is most abundant in normal muscle and liver. The major site of normal glycogen synthesis and breakdown is the cytoplasm, not obviously involving lysosomes. Autophagy has been proposed as the mechanism whereby cytoplasmic glycogen is transferred to lysosomes for breakdown, although the precise regulation of this autophagic event is not clear. Accumulation of glycogen within lysosomes of muscle and liver, organs severely involved in the disease process, may simply reflect higher autophagic activity in tissues with normally high glycogen content.
Unexpectedly, high levels of GAA mRNA signals were observed in neurons of the CNS, and increasing GAA mRNA signal was shown with in utero maturation in mouse CNS and PNS neurons. In both cases, normal enzyme activities20-22,25 parallel mRNA levels. The mouse GAA knockout models have glycogen accumulation in Schwann cells and some neurons in addition to liver and muscle. Interestingly, review of the neuropathology of Pompe disease reveals major lysosomal glycogen accumulation in Schwann cells of the PNS, in glial cells, and, in decreasing amounts, in neurons of spinal cord, brainstem, and cerebrum.4,6-8 These results suggest roles for GAA in specific cell types in fetal and adult CNS tissues. The lack of clear documented CNS effects in the mouse or human GAA deficiencies suggests either a lack of toxic effects of stored glycogen in the CNS neurons or the need for more intensive efforts to document potential sequelae in the nervous system. With the recent interest in development of enzyme replacement and/or gene therapy for GSDII, intensive surveillance for CNS involvement may be essential during assessment for overall efficacy. Additionally, such examinations are critical for the full evaluations of the mice homozygous for the targeted disruption of the GAA locus.24
The differences in histopathological changes in muscle and in functional compromise between the different GAA knockout models may reflect the known differences in their genetic background, and would be consistent with the possibility that other specific metabolic differences play a significant role in the pathophysiological mechanisms and as modifier phenotype. These differences among tissues and during the maturation process remain to be elucidated and are critical to understanding the manifestations of GAA deficiency in humans and knockout mice.
| Acknowledgements |
|---|
| Footnotes |
|---|
Supported by grants from Children's Hospital Research Foundation (E. Ponce), March of Dimes National Foundation (R. Hirschhorn), Muscular Dystrophy Association (R. Hirschhorn), and NIH grant DK36729 (G. A. Grabowski).
Accepted for publication December 22, 1998.
| References |
|---|
|
|
|---|
-Glucosidase in generalized glycogen-storage disease (Pompe's disease). Biochem J 1963, 86:11-16[Medline]
-1,4-glucan glucohydrolase in human tissues. Biochim Biophys Acta 1981, 110:124-133
-glucosidase (acid maltase) deficiency. Scriver CR Beaudet AL Sly WL Valle D eds. The Metabolic and Molecular Bases of Inherited Disease. 1995, :pp 2443-2464 McGraw Hill, New York
-glucosidase (GAA) and thymidine kinase (TK1) by fluorescence in situ hybridization. Hum Genet 1996, 97:404-406[Medline]
-glucosidase gene. Biochem J 1990, 272:493-497[Medline]
-glucosidase, detection of an intron in the 5' untranslated leader sequence, definition of 18-bp polymorphisms, and differences with previous cDNA and amino acid sequences. DNA Cell Biol 1990, 9:85-94[Medline]
-glucosidase in transiently transfected mammalian cells. Biochem J 1990, 272:485-492[Medline]
-glucosidase and detection of genetic heterogeneity for mRNA in three
-glucosidase-deficient patients. Proc Natl Acad Sci USA 1986, 83:9641-9644
-glucosidase: homology with the intestinal sucrase-isomaltase complex. EMBO J 1988, 7:1697-1704[Medline]
-glucosidase deficiency as detected by abnormalities of DNA and mRNA. Am J Hum Genet 1990, 47:73-78[Medline]
-glucosidase activity and glycogen content in patients with generalized glycogenosis. Isr J Med Sci 1967, 3:411-421[Medline]
-glucosidase gene from barley. Plant Mol Biol 1996, 30:229-241[Medline]
-glucosidase gene in mice causes an illness with critical features of both infantile and adult human glycogen storage disease type II. J Biol Chem 1998, 273:19086-19092
-galactosidase A. Gene 1995, 166:277-280[Medline]
This article has been cited by other articles:
![]() |
L. R. DeRuisseau, D. D. Fuller, K. Qiu, K. C. DeRuisseau, W. H. Donnelly Jr, C. Mah, P. J. Reier, and B. J. Byrne Neural deficits contribute to respiratory insufficiency in Pompe disease PNAS, June 9, 2009; 106(23): 9419 - 9424. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. Yan, N. Raben, and P. Plotz The Human Acid alpha -Glucosidase Gene Is a Novel Target of the Notch-1/Hes-1 Signaling Pathway J. Biol. Chem., August 9, 2002; 277(33): 29760 - 29764. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. Yan, J. Heus, N. Lu, R. C. Nichols, N. Raben, and P. H. Plotz Transcriptional Regulation of the Human Acid alpha -Glucosidase Gene. IDENTIFICATION OF A REPRESSOR ELEMENT AND ITS TRANSCRIPTION FACTORS Hes-1 AND YY1 J. Biol. Chem., January 12, 2001; 276(3): 1789 - 1793. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |