![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Previous Article | Next Article 
Molecular and Cellular Biology, August 2006, p. 5621-5635, Vol. 26, No. 15
0270-7306/06/$08.00+0 doi:10.1128/MCB.02357-05
Copyright © 2006, American Society for Microbiology. All Rights Reserved.
Center for Human Genetics, Katholieke Universiteit Leuven, Leuven, Belgium,1 Department of Pathology, Katholieke Universiteit Leuven, Leuven, Belgium,2 Laboratory of Biological Psychology, Department of Psychology, Katholieke Universiteit Leuven, Leuven, Belgium,3 Laboratory of Neurochemistry and Behavior, Born-Bunge Institute, University of Antwerp, Antwerp, Belgium,4 Biochemistry II, University of Goettingen, Goettingen, Germany,5 Center for Human Genetics and Flanders Interuniversity Institute for Biotechnology, VIB4, Katholieke Universiteit Leuven, Leuven, Belgium,6 Universitätskinderklinik, Uni-Heidelberg, Heidelberg, Germany7
Received 10 December 2005/ Returned for modification 26 February 2006/ Accepted 25 April 2006
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Phosphomannomutase 2 (PMM2) is a key enzyme in the initial steps of N glycosylation. PMM2 catalyzes the conversion of mannose-6-P to mannose-1-P (21), which, through the binding of GTP, generates GDP-mannose. This GDP-mannose is the source of the mannose core of the N-glycan precursors (1).
Mutations in the PMM2 gene result in reduced phosphomannomutase activity (18, 19, 23) and depletion of the GDP-mannose pool. As a result, numerous serum (e.g., transferrin, haptoglobin, and 
1-antitrypsin) (4, 26, 28) and cerebrospinal fluid glycoproteins (e.g., transferrin and ß-trace protein) (9, 11, 24) in CDG-Ia patients are underglycosylated, whereas many lysosomal and other cellular enzymes loose their activity, probably also as a result of the hypoglycosylation (17). This hypoglycosylation causes a severe clinical syndrome characterized by developmental abnormalities of mainly the nervous system (featuring cerebellar hypoplasia and demyelinization) and peripheral neuropathy, resulting in a generalized psychomotor retardation. Severe cases present additional failure of single or multiple visceral organs. Most often the liver, heart, gut, and/or kidney are affected (16). The severe consequences of a reduced PMM2 activity are puzzling, since in humans another PMM gene, PMM1 was identified (20, 27). PMM1 displays 66% homology to PMM2 on an amino acid level and 65% on a nucleotide level. A comparison of the genomic structure of the PMMs indicated that the genes have probably arisen by gene duplication 75 to 110 million years ago (13, 25). The PMMs are highly conserved during evolution: the murine genes are located on syntenic regions and display 90% identity with the human genes. The presence of these two highly conserved PMM genes in the genome implies that both have crucial functions, whereby the physiologic role of PMM1 has remained elusive. Specifically, PMM1 is not implicated in any known disease (20-22), and solely mutations of PMM2 that cause a reduced enzyme activity (loss-of-function seems to be intolerable) are associated with disease, i.e., CDG-Ia. Since both the subcellular localization and the catalytic abilities are quite similar, it is counterintuitive that the PMM1 enzyme does not compensate for the reduced PMM2 activity and argues for a function unrelated to mannose-1-P synthesis.
In line with this reasoning, some biochemical differences between both PMMs have been identified. Besides its phosphomannomutase activity in vitro, PMM1 has an additional phosphoglucomutase activity and in vitro also converts glucose-1-P into glucose-6-P, whereas PMM2 converts glucose-1-P 20 times more slowly than mannose-1-P. PMM1 has a higher Ka value for both mannose-1,6-bisphosphate and glucose-1,6-bisphosphate compared to the Ka values for PMM2. Fructose-1,6-bisphosphate can also stimulate PMM1, in contrast to PMM2. When incubated with fructose-1,6-bisphosphate, PMM2 activity depends on the formation of mannose-1,6-bisphosphate from fructose-1,6-bisphosphate and mannose-1-phosphate by PMM1 (21).
The phosphomannomutases not only differ in kinetics. Previous Northern blot analysis on adult human tissues has indicated some differences in expression pattern. PMM1 mRNA was abundant in brain, liver, pancreas, kidney, skeletal muscle, and heart samples, and lower levels were detected in placenta and lungs. Highest expression of PMM2 was found in pancreas, liver, kidney, and placenta. Skeletal muscle, heart, and lung showed lower PMM2 levels, whereas in brain almost no signal was observed (19, 20). These results suggest PMM1 is the brain specific phosphomannomutase, which is in contrast with the patient's clinical picture of severe neurological involvement due to decreased PMM2 activity.
To unravel the biological role of PMM1 and PMM2, the murine orthologous genes were previously cloned (13), and their expression in adult mice was analyzed on a Northern blot. In comparison to the published human expression profile described above, this revealed a quite similar expression pattern for the murine Pmm1 and Pmm2 and especially showed high Pmm1 levels in adult mouse brain and visceral organs (13). Enzyme activity measurements of adult rat brain extracts after Pmm2 immunoprecipitation indicated that Pmm1 accounts for 66% of the total phosphomannomutase activity in brain (21).
To elucidate the physiological role of Pmm1 and to gain a better understanding of its expression profile within the tissues, we documented the expression pattern of the Pmm1 protein in embryonic and adult tissues of wild-type mice on the tissue level by immunohistochemistry and ISH. We especially expected additional insights into possible differences in the microdistribution of both enzymes that could have escaped Northern blot approaches. We then generated Pmm1-deficient mice and looked for phenotypic changes in adult mice and embryos. Glycan structures were studied by lectin glycohistochemical analysis of Pmm1-deficient tissues.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
For construction of the targeting vector, a pBSK vector (Stratagene) without XhoI restriction site was constructed by digestion with XhoI and HindIII and subsequent Klenow polymerase blunting of the overhangs and ligation. A 6-kb EcoRI subclone from the BACclone 85J13 (Research Genetics), containing the entire Pmm1 gene was inserted in this vector. This subclone consisted of a fragment containing the region between the last 50 bases of intron 1 and the first 750 bases of intron 6. Exon 3 of Pmm1 was mutated to create unique XhoI and NheI sites by insertion of a C and mutating a CT into GC, respectively. A XhoI-XbaI digestion thus resulted in a loss of 29 bp from exon 3 and allowed the insertion of a XhoI-XbaI PGK-Neo cassette in inverse orientation (Fig. 1A).
|
Behavioral assessment in Pmm1 knockout mice. Analysis for a behavioral phenotype was done by subjecting wild-type and knockout mice (aged 1 year) to a modified and extended version of the behavioral test battery described previously (6). To assess a variety of different functions, the battery included tests for locomotion and exploratory activity (cage activity test, open-field activity, dark-light transition, and social exploration test), gait and neuromotor coordination (walking pattern and rotarod), muscle strength (wire suspension test), and learning or memory (passive avoidance and Morris water maze).
Briefly, cage activity was tested by counting the number of beam crossings in a standard home cage equipped with three infrared beams during dusk-phase (2-h) and overnight (16-h) observation periods. Open-field recording was performed to evaluate the general exploratory behavior when mice were placed for 10 min in an open-field arena equipped with a camera. In addition, exploratory behavior was tested in a dark-light transition box. The number of transitions between the two chambers was measured as the number of infrared beam interruptions. A first beam was placed close to the dark box, and a second beam was located some distance from the dark compartment. To assess social and/or sexual exploration in male mice, different ambulatory measures were recorded in an open-field arena containing a round cage with two wild-type female mice.
Gait ability and rotarod performance were recorded to examine neuromotor coordination and motor learning. During four consecutive trials on an accelerating rotarod (4 to 40 rpm, 5 min), time on the rod was recorded for up to 300 s.
To test for muscle strength, mice were to grasp a wire with their forepaws, and their ability to remain suspended for 120 s was measured.
In the training trial of the passive avoidance test, mice were placed in the light compartment of a step-through box, and the latency to enter the dark compartment was recorded. After entry in the dark compartment, a mild footshock was delivered through the metal grid floor (0.3 mA, 1 s). In the retention test (24 h after training), the mice were again placed in the illuminated box, and latency to re-enter the dark compartment was timed. Morris water maze training consisted of eight acquisition trial blocks (each comprising four daily swim trials), followed by a retention test during which the hidden escape platform was removed from the pool (probe trial), and the search path of the mice was video tracked for 100 s.
Analysis of the Pmm1 expression pattern by Western blot and immunohistochemistry. Euthanasia of the animals, isolation, and processing of the tissues or whole embryos for either Western analysis or immunohistochemistry were done as described in reference 7. Tissues investigated included brain, intestine, pancreas, liver, lung, testis, ovary, and adrenal gland. Western blots were done with a specific affinity-purified polyclonal rabbit anti-mouse Pmm1 antibody as described previously (7). Western analysis of embryonic and adult non neural tissues (15 µg) required antibody concentration of 1 in 500 and overnight exposure. Blots of adult endocrine glands such as testis, ovary, and adrenal gland were exposed for only 5 to 10 min. In all tissues a specific band corresponding to Pmm1 was detected, except for lung tissues, where a smaller protein showed nonspecific cross-reactivity with our antibody.
Immunohistochemistry was done with a polyclonal anti-Pmm1 antiserum whose specificity had been previously demonstrated in brain (7). However, this antiserum revealed a nonspecific reaction in embryonic heart and acrosomes of spermatozoa (not shown).
Experiments were repeated on tissues from four different wild-type and knockout mice at adult or embryonic stages and showed comparable results.
To identify the Pmm1 positive cells in pancreatic islets, double stainings were performed with either a commercial anti-mouse glucagon (Sigma) or a anti-swine insulin antibody (Dako).
(Glyco)histochemical analysis of Pmm1 knockout mice. Brains isolated from different developmental stages (postnatal day 0 [P0], P7, P20, and adult), adult testis and whole embryos (embryonic day 17 [E17]) from C57BL/6J wild-type mice, and Pmm1 knockouts were analyzed by lectin histochemistry to check for possible aberrant glycan structures. The tissues were prepared as described in reference 7. Deparaffinized sections were incubated for 1 h at room temperature with the lectins summarized in Table 1, together with their nominal specificities (Vector Laboratories). Brain sections were screened with antibodies to GFAP (antiglial fibrillary acid protein SMI-22 antibody, marker of astroglial cells; Steinberger Monoclonal, Inc.) and F4/80 (microglial/macrophage marker). Bis-benzimide was used as a nuclear counterstain. Sections were covered in Vectashield (Vector Laboratories).
|
|
RESULTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We traced Pmm expression by Western blotting expression analysis in the mouse both prenatally and in adulthood.
The data showed that Pmm1 expression levels are high in most tissues at both stages (Fig. 2A and B). The high expression level of Pmm1 in mouse embryos was characteristic for at least several embryonic tissues. Brain was the organ with the highest expression level. The signals for liver and lung were somewhat higher than for intestine and pancreas. These high expression levels were, however, downregulated postnatally. In adult intestine the Pmm1 concentration was below detection limits, whereas in the liver, pancreas, and lung the signal was low (Fig. 2B).
|
Immunohistochemistry. To further analyze the cell types expressing Pmm1, we investigated embryonic and adult tissues by immunohistochemistry. In all Pmm1-positive regions, the staining was distributed throughout the cell body, in line with their documented localization (12).
Pmm1 expression in the embryonic mouse tissues. Staining of the E17 embryo with the Pmm1 antibody revealed strong signals in most organ systems: nervous system, gastrointestinal tract, endocrine glands, urinary system, and respiratory system. Epithelia such as the lung, gut, liver, and kidneys and endocrine glands such as the pancreatic islets, hypophysis, and adrenal gland were especially highly immunoreactive for Pmm1, as well as nervous system components, developing bone, muscle, and thymus.
(i) Nervous system. Nervous tissue components displaying high Pmm1 signals included peripheral ganglia, cerebellum, and cerebral cortex, including the hippocampus (Fig. 3A). Neurons in peripheral ganglia such as the spinal ganglia and the trigeminal ganglion were intensely labeled, as well as the peripheral, but not the central neurite branches of the sensory ganglia (Fig. 3B and C).
|
In dendrites and axons the signals were less intense and the molecular zone always displayed a weak signal. Central white matter tracts were always negative (data not shown).
(ii) Gastrointestinal tract. In the E17 embryo, epithelial cells of the mucous membrane lining the gastrointestinal tract showed Pmm1-positive signals in the esophagus, stomach, and intestines. The submucosa and muscular compartment never displayed immunoreactivity for Pmm1. The body of the stomach showed intense signals in the glandular mucous membrane (Fig. 4A). The epithelial lining of the mucosa of the small intestine showed an equal distribution of the Pmm1 signal in villi and crypts (Fig. 4B).
|
(iii) Endocrine glands. The islets of Langerhans of the endocrine pancreas, visible as large clusters scattered throughout the secretory cells, showed intense Pmm1 immunoreactivity (Fig. 4D). In the developing adrenal gland isolated cell groups were highly Pmm1 immunoreactive (Fig. 4E).
(iv) Urinary system. Staining of the developing kidney with the Pmm1 antibody revealed high signals in the cortical region, whereas the medullary region was somewhat less stained. In the cortex, glomeruli were especially intensely labeled. Signals were less pronounced in the cortical proximal and distal convoluted tubules and in the components of the collecting system in the medullary region (Fig. 4E).
(v) Respiratory system. The lung alveoli, which at this stage are still collapsed, showed intense signals in the epithelial cells that do not yet have their mature squamous appearance as in expanded alveoli but are still cuboidal in shape. Terminal bronchi are less intensely stained (Fig. 4F).
Pmm1 expression in adult tissues. Tissues analyzed in detail included brain, intestine, pancreas, testis, ovary, and adrenal gland.
(i) Nervous system. The high Pmm1 immunoreactivity seen in the embryonic nervous system is maintained in the adult brain. Intensely labeled regions include the cortex, hippocampus, cerebellum, and olfactory bulb. As in the embryo, intense staining was almost completely restricted to neuronal cell bodies. Immunoreactivity with dendrites resulted in weak neuropil staining. Glial cells in the gray matter were mostly negative as were white matter tracts.
As described previously, Pmm1 signals in the adult brain were mostly restricted to neuronal cell bodies: the neurons of all cortical layers (Fig. 5A), Purkinje cells and, to a lesser extent, the cerebellar granular cells (Fig. 5B and C), granule cells in the adult dentate gyrus (Fig. 5D and E), pyramidal cells of CA1, CA2, and CA3, as well as some scattered polymorphic neurons.
|
Intense staining in the olfactory bulb was observed in the mitral cell layer and the glomerular layer, whereas the signals were less intense in the granular layer and subependymal zone (Fig. 5F to H). The external plexiform layer was almost negative for Pmm1. In the glomerular layer staining was restricted to the periglomerular cells, whereas the neuronal processes in the glomeruli were completely devoid of any signal (Fig. 5G). Also in the other layers, staining was restricted to the neuronal cells: mitral cells, granular cells, and a few scattered cells in the external plexiform layer (Fig. 5H).
(ii) Gastrointestinal tract. In the adult gut, Pmm1 expression was drastically downregulated. Whereas in the embryo the signals were equally distributed throughout the mucosa, in the adult gut the enterocytes of the villi showed remarkably low expression. Pmm1 signals were concentrated in the proliferative crypts (Fig. 6A), and occasionally signals were seen in goblet cells.
|
(iii) Endocrine glands. The adult adrenal gland revealed a region-specific expression with high Pmm1 signals in the cortex in contrast to the medullary region. In the cortex, the clusters of mineralocorticoid-secretoring cells of the zona glomerulosa showed particularly intense signals. The glucocorticoid-secreting cells in the other cortical layers, the zona fasciculata and reticularis, displayed considerably weaker signals. Catecholamine-secreting cells of the medullary region were completely devoid of Pmm1 signals (Fig. 6B).
Immunohistochemical analysis of the gonads with the affinity-purified Pmm1 antibody revealed intense signals in distinct cell populations. A specific staining was observed in the spermatogonia of the seminiferous tubuli, whereas the more differentiated stages of the germ cells did not show Pmm1 signals. Sertoli cells and Leydig cells were negative (Fig. 6C). These results are, however, in contrast with the in situ analysis that showed a specific Pmm1 signal in the postmeiotic cells (Fig. 6D).
In the adult female gonads the ovarian germ cells displayed Pmm1 immunoreactivity: secondary follicles showed strong cytoplasmic staining, and weaker staining was observed in the zona granulosa and theca interna (Fig. 6E). The graafian follicles showed a particularly strong staining in the endocrine cells in the theca interna (Fig. 6F).
In the adult pancreas the Pmm1 signals are restricted to the islets of Langerhans (Fig. 7A and C). Double stainings for glucagon and insulin revealed Pmm1 is expressed in both cell populations (Fig. 7B and D).
|
Southern blotting of eighty transfected ES cells revealed only one recombinant clone (Fig. 1B). After transfer of ES cells from this clone to pseudopregnant females, 19 pups were born. Eight of these pups were chimeric, and germ line transmission was observed with three male chimeras. Mutant mice homozygous for the Pmm1 gene disruption were obtained from heterozygote matings. Disruption of the Pmm1 gene was detected by allele specific PCR. The wild-type specific band of 330 bp was only apparent in the wild type and heterozygotes, whereas the 1.5-kb fragment was specific for the knockout allele (Fig. 1C). The disruption of the Pmm1 gene was confirmed by Southern blotting of genomic tail DNA, where the wild-type specific band of 11.5 kb is replaced by the targeted allele of 10 kb in Pmm1–/– mice (data not shown). The absence of Pmm1 protein in the Pmm1-deficient mice was shown by Western blotting of brain homogenates with the specific polyclonal anti-Pmm1 antibody (Fig. 1D).
Phenotypical analysis of Pmm1 knockout mice. Pmm1 knockout mice obtained on a mixed C57BL/6J background were backcrossed for five generations. The mice were viable and fertile and had normal litter sizes. They developed normally, without any major phenotype up to adulthood.
(i) Histological analysis. Analysis of the major Pmm1-expressing structures by light microscopy of hematoxylin-eosin sections did not reveal any pathologically alterations in the knockout mice (Fig. 8 and data not shown), although Pmm1 protein was clearly depleted, as demonstrated by immunohistochemistry of Pmm1 on sections of Pmm1-null mice (data not shown). Detailed histological screening of adjacent sections with markers of astroglial cells (antiglial fibrillary acid protein SMI-22 antibody) and microglia/macrophages marker (F4/80) did not reveal significant differences between wild-type and knockout mice (data not shown).
|
|
Finally, learning and memory were evaluated with the passive avoidance test and Morris water maze. In passive avoidance learning, no differences in performance were observed between the groups. Spatial navigation learning was unaffected, as assessed in the Morris water maze test. No distinguishable differences were observed during eight acquisition trial blocks, and all animals learned to reach the escape platform. During the probe trial, groups performed equally well and spent most time searching in the previously trained (target) quadrant (42% of the time for the knockout mice; 37% of the time for the wild-type mice).
In summary, Pmm1 knockout mice showed no abnormalities in general motor functions and performed similarly to their wild-type littermates in behavioral tests for learning abilities.
(iii) Glycohistochemical analysis. Although Pmm1 is highly expressed in brain throughout development, no differences in lectin binding patterns were observed between wild-type and knockout mouse brains in the stages investigated. None of the 21 lectins displayed an aberrant pattern in whole Pmm1 knockout embryos (details not shown).
(iv) Expression of Pmm2 in Pmm1-null mice. From previous studies it was clear that Pmm2 expression overlaps with Pmm1 in brain neurons (7). Because in vitro studies have shown Pmm2 is a functional homologue of Pmm1 (21), we tested whether Pmm2 was upregulated in Pmm1-deficient mice. The abundance of Pmm2 in homogenates from knockouts and wild-type littermates was assayed by Western blotting. The Pmm2 specific antibody showed similar levels in both Pmm1-knockout and wild-type mice (Fig. 9.).
|
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Our results first indicate that Pmm1 is almost ubiquitously expressed in the E17 embryo. In the adult mouse Pmm1 expression changes to a more restricted pattern, with brain and endocrine glands displaying high signals. This argues for a function of Pmm1 both in embryonic development and in adult neuroendocrine tissues.
Surprisingly, knocking out the Pmm1 gene did not affect normal mouse development. Pmm1-deficient mice did not reveal any obvious abnormality and developed normally.
Pmm1 is ubiquitously expressed in the E17 embryo. We found high expression levels of the Pmm1 protein in the developing nervous system. Pmm1 is abundantly expressed in cells of proliferative zones such as the neuroblasts in the VZs and cerebellar germinal trigone. The epithelial layers of different organs such as the gastrointestinal tract and the respiratory and urinary systems of E17 embryos are highly immunoreactive as well. Moreover, the embryonic glands revealed considerable Pmm1 expression.
Pmm1 expression levels are downregulated in the adult mouse tissues. During adulthood, the high developmental Pmm1 expression was maintained in the brain, where Pmm1 signals were found in the neural cell bodies. In adult non-neural tissues the signals were downregulated, even below the detection level in the gut. Immunohistochemical analysis revealed Pmm1 mainly in the crypts, i.e., in or close to proliferating cells. Also, throughout the endocrine glands the high expression levels were maintained. This could argue for an important general function for Pmm1 in developing epithelial and neuronal tissues, whereas in the adult mouse Pmm1 is expressed in self-renewing stem cell populations such as the crypts of the adult intestine and cells with endocrine function.
Pmm1 expression and correlation with the CDG-Ia phenotype. Pmm1 is highly expressed in the tissues that are most severely affected in CDG-Ia, such as liver, kidney, gut, and brain (8, 29).
Of course, it is not clear yet whether the expression of Pmm1 in the CDG-Ia affected tissues corresponds to the localization of Pmm2 within these tissues. This will have to await the full documentation of the Pmm2 expression pattern.
In the brain, however, a complete expression pattern analysis was already presented for both Pmms (7). This revealed Pmm1 expression in the cerebellar granule and Purkinje cells. These cells are almost completely lost in CDG-Ia cerebellum (3, 5, 14, 15), probably due to a developmental defect progressing throughout early childhood (2, 10). The presence of Pmm1 in the external granular layer throughout cerebellar development is surprising, especially since Pmm2 is expressed in the exact same cell type and subcellular compartment (7).
The presence of Pmm1 in the most severely affected tissues in CDG-Ia and its exact colocalization with Pmm2 in the affected cell-types in the CDG-Ia brain confirm that Pmm1 cannot compensate for Pmm2 under in vivo conditions.
Pmm1-deficient mice are phenotypically normal. As judged by the expression data, loss of Pmm1 should have an impact on the developing brain and peripheral organs and possibly cell renewal in selected postneural tissues, i.e., follow an overall developmental motif of function. One could expect Pmm1-deficient mice to display deficiencies such as neurologic abnormalities, dysregulation of endocrine function, and fertility problems. However, we did not observe any of these abnormalities in our knockout animals. The Pmm1-deficient mice did not present a detectable phenotype and were indistinguishable from their wild-type littermates. They performed well on neurological tests, including tests for strength, balance, motor coordination, and learning. In addition, a detailed histological and glycohistochemical analysis did not reveal any abnormalities.
Lack of functional compensation by Pmm2. A possible explanation for the apparent normal condition of the Pmm1-deficient mice is redundancy, which is an evident outcome of gene duplication. In view of their similar in vitro activities, the most likely candidate as a compensating enzyme would be PMM2. However, a mutual compensation is unlikely since PMM1 cannot replace PMM2 in patients with congenital disorders of glycosylation (CDG) despite a broad overlap of their expression patterns (7). This suggests that the inability of Pmm1 to take over the Pmm2 function in brain is not due to a difference in expression pattern but more probably to a difference in physiological function. In line with this, Pmm2 expression levels were not upregulated in the Pmm1 knockout mice. Furthermore, our glycohistochemical and serum protein analysis in the Pmm1-deficient mice suggests that Pmm1 is not involved in protein glycosylation. To fully exclude this option, Pmm1 deficiency in a Pmm2-null background would have to be analyzed. Due to the early embryonic lethality of mice lacking Pmm2 (see accompanying study [25a]), this question cannot be directly answered and awaits the generation of mice with a partial loss of Pmm2 function surviving long enough for a meaningful analysis of an additional Pmm1 deficiency.
The genomic architecture of the human PMM and murine Pmm genes is remarkably well conserved, indicating that the original duplication event has occurred some 75 to 110 million years ago, well before mammalian radiation (25). Thus, the PMM genes would only be maintained in the genome for 100 million years if their duplication and divergence had a functional significance: without further selection one duplicate is inactivated with a half-life of 4 million years.
Indeed, compared to Pmm2, there are some regional differences in the expression pattern. In addition, Pmm1 encodes a phosphomannomutase that shows different substrate specificity and kinetics in in vitro assays (21). These observations argue for a different physiological role for PMM1 yet to be elucidated. From this standpoint, an alternative for the analysis of Pmm1 function might be subjecting the Pmm1-deficient mice to environmental challenge or the generation of mice overexpressing Pmm1.
Conclusion. Pmm1 is an enzyme that is widely but specifically expressed in embryonic and adult brain and endocrine glands. In most peripheral organs, high expression levels in the embryonic stage are downregulated at adulthood.
Despite the high expression levels in brain and endocrine glands, Pmm1 knockouts were viable and fertile and no histological abnormalities were observed in any of the major organ systems, including the nervous system and endocrine tissues that were studied in detail. Behavioral assays did not reveal any differences between Pmm1-null mice and wild-type mice. Glycan analysis of serum proteins and lectin glycohistochemistry did not reveal aberrant glycosylation patterns in the knockout.
The biological significance of Pmm1 may be revealed by crossing these mice with Pmm2-deficient mice or by challenging the Pmm1-null mice with different environmental factors.
. . . . .
|
ACKNOWLEDGMENTS |
|---|
We thank Stephanie Grünewald and Natalie De Geest for help with antibody generation, testing of the antibodies, and histology.
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
2. Akaboshi, S., K. Ohno, and K. Takeshita. 1995. Neuroradiological findings in the carbohydrate-deficient glycoprotein syndrome. Neuroradiology 37:491-495.[Medline]
3. Antoun, H., N. Villeneuve, A. Gelot, S. Panisset, and C. Adamsbaum. 1999. Cerebellar atrophy: an important feature of carbohydrate deficient glycoprotein syndrome type 1. Pediatr. Radiol. 29:194-198.[CrossRef][Medline]
4. Artuch, R., I. Ferrer, J. Pineda, J. Moreno, C. Busquets, P. Briones, and M. A. Vilaseca. 2003. Western blotting with diaminobenzidine detection for the diagnosis of congenital disorders of glycosylation. J. Neurosci. Methods 125:167-171.[CrossRef][Medline]
5. Barone, R., L. Pavone, A. Fiumara, R. Bianchini, and J. Jaeken. 1999. Developmental patterns and neuropsychological assessment in patients with carbohydrate-deficient glycoconjugate syndrome type IA (phosphomannomutase deficiency). Brain Dev. 21:260-263.[CrossRef][Medline]
6. Crawley, J. N., J. K. Belknap, A. Collins, J. C. Crabbe, W. Frankel, N. Henderson, R. J. Hitzemann, S. C. Maxson, L. L. Miner, A. J. Silva, J. M. Wehner, A. Wynshaw-Boris, and R. Paylor. 1997. Behavioral phenotypes of inbred mouse strains: implications and recommendations for molecular studies. Psychopharmacology 132:107-124.[CrossRef][Medline]
7. Cromphout, K., L. Keldermans, A. Snellinx, J. F. Collet, S. Grunewald, N. De Geest, R. Sciot, E. Vanschaftingen, J. Jaeken, G. Matthijs, and D. Hartmann. 2005. Tissue distribution of the murine phosphomannomutases Pmm1 and Pmm2 during brain development. Eur. J. Neurosci. 22:991-996.[CrossRef][Medline]
8. de Lonlay, P., N. Seta, S. Barrot, B. Chabrol, V. Drouin, B. M. Gabriel, H. Journel, M. Kretz, J. Laurent, M. Le Merrer, A. Leroy, D. Pedespan, P. Sarda, N. Villeneuve, J. Schmitz, E. van Schaftingen, G. Matthijs, J. Jaeken, C. Korner, A. Munnich, J. M. Saudubray, and V. Cormier-Daire. 2001. A broad spectrum of clinical presentations in congenital disorders of glycosylation I: a series of 26 cases. J. Med. Genet. 38:14-19.
9. Doucey, M. A., D. Hess, R. Cacan, and J. Hofsteenge. 1998. Protein C mannosylation is enzyme-catalyzed and uses dolichyl-phosphate-mannose as a precursor. Mol. Biol. Cell 9:291-300.
10. Drouin-Garraud, V., M. Belgrand, S. Grunewald, N. Seta, J. N. Dacher, A. Henocq, G. Matthijs, V. Cormier-Daire, T. Frebourg, and P. Saugier-Veber. 2001. Neurological presentation of a congenital disorder of glycosylation CDG-Ia: implications for diagnosis and genetic counseling. Am. J. Med. Genet. 101:46-49.[Medline]
11. Grunewald, S., K. Huyben, J. G. de Jong, J. A. Smeitink, E. Rubio, G. H. Boers, H. S. Conradt, U. Wendel, and R. A. Wevers. 1999. beta-Trace protein in human cerebrospinal fluid: a diagnostic marker for N glycosylation defects in brain. Biochim. Biophys. Acta 1455:54-60.[Medline]
12. Hansen, S. H., S. R. Frank, and J. E. Casanova. 1997. Cloning and characterization of human phosphomannomutase, a mammalian homologue of yeast SEC53. Glycobiology 7:829-834.
13. Heykants, L., E. Schollen, S. Grunewald, and G. Matthijs. 2001. Identification and localization of two mouse phosphomannomutase genes, Pmm1 and Pmm2. Gene 270:53-59.[CrossRef][Medline]
14. Horoupian, D. 1995. Olivopontocerebellar atrophy in carbohydrate-deficient glycoprotein syndrome. Pediatr. Pathol. Lab. Med. 15:175-179.[Medline]
15. Horslen, S. P., P. T. Clayton, B. N. Harding, N. A. Hall, G. Keir, and B. Winchester. 1991. Olivopontocerebellar atrophy of neonatal onset and disialotransferrin developmental deficiency syndrome. Arch. Dis. Child 66:1027-1032.[Abstract]
16. Jaeken, J., and H. Carchon. 2001. Congenital disorders of glycosylation: the rapidly growing tip of the iceberg. Curr. Opin. Neurol. 14:811-815.[CrossRef][Medline]
17. Jaeken, J., G. Matthijs, H. Carchon, and E. Van Schaftingen. 2001. Defects of N-glycan synthesis, p. 1601-1622. In A. L. B. C. R. Scriver, W. S. Sly, and D. Valle (ed.), The metabolic and molecular bases of inherited diseases. McGraw-Hill Book Co., New York, N.Y.
18. Matthijs, G., E. Schollen, C. Bjursell, A. Erlandson, H. Freeze, F. Imtiaz, S. Kjaergaard, T. Martinsson, M. Schwartz, N. Seta, S. Vuillaumier-Barrot, V. Westphal, and B. Winchester. 2000. Mutations in PMM2 that cause congenital disorders of glycosylation, type Ia (CDG-Ia). Hum. Mutat. 16:386-394.[CrossRef][Medline]
19. Matthijs, G., E. Schollen, E. Pardon, M. Veiga-Da-Cunha, J. Jaeken, J. J. Cassiman, and E. Van Schaftingen. 1997. Mutations in PMM2, a phosphomannomutase gene on chromosome 16p13, in carbohydrate-deficient glycoprotein type I syndrome (Jaeken syndrome). Nat. Genet. 16:88-92.[CrossRef][Medline]
20. Matthijs, G., E. Schollen, M. Pirard, M. L. Budarf, E. Van Schaftingen, and J. J. Cassiman. 1997. PMM (PMM1), the human homologue of SEC53 or yeast phosphomannomutase, is localized on chromosome 22q13. Genomics 40:41-47.[CrossRef][Medline]
21. Pirard, M., Y. Achouri, J. F. Collet, E. Schollen, G. Matthijs, and E. Van Schaftingen. 1999. Kinetic properties and tissular distribution of mammalian phosphomannomutase isozymes. Biochem. J. 339(Pt. 1):201-207.[Medline]
22. Pirard, M., J. F. Collet, G. Matthijs, and E. Van Schaftingen. 1997. Comparison of PMM1 with the phosphomannomutases expressed in rat liver and in human cells. FEBS Lett. 411:251-254.[CrossRef][Medline]
23. Pirard, M., G. Matthijs, L. Heykants, E. Schollen, S. Grunewald, J. Jaeken, and E. van Schaftingen. 1999. Effect of mutations found in carbohydrate-deficient glycoprotein syndrome type IA on the activity of phosphomannomutase 2. FEBS Lett. 452:319-322.[CrossRef][Medline]
24. Pohl, S., A. Hoffmann, A. Rudiger, M. Nimtz, J. Jaeken, and H. S. Conradt. 1997. Hypoglycosylation of a brain glycoprotein (beta-trace protein) in CDG syndromes due to phosphomannomutase deficiency and N-acetylglucosaminyl-transferase II deficiency. Glycobiology 7:1077-1084.
25. Schollen, E., E. Pardon, L. Heykants, J. Renard, N. A. Doggett, D. F. Callen, J. J. Cassiman, and G. Matthijs. 1998. Comparative analysis of the phosphomannomutase genes PMM1, PMM2, and PMM2psi: the sequence variation in the processed pseudogene is a reflection of the mutations found in the functional gene. Hum. Mol. Genet. 7:157-164.
25. Thiel, C., T. Lübke, G. Matthijs, K. von Figura, and C. Körner. 2006. Targeted disruption of the mouse phosphomannomutase 2 gene causes early embryonic lethality. Mol. Cell. Biol. 26:5615-5620.
26. Wada, Y., A. Nishikawa, N. Okamoto, K. Inui, H. Tsukamoto, S. Okada, and N. Taniguchi. 1992. Structure of serum transferrin in carbohydrate-deficient glycoprotein syndrome. Biochem. Biophys. Res. Commun. 189:832-836.[CrossRef][Medline]
27. Wada, Y., and M. Sakamoto. 1997. Isolation of the human phosphomannomutase gene (PMM1) and assignment to chromosome 22q13. Genomics 39:416-417.[CrossRef][Medline]
28. Yamashita, K., H. Ideo, T. Ohkura, K. Fukushima, I. Yuasa, K. Ohno, and K. Takeshita. 1993. Sugar chains of serum transferrin from patients with carbohydrate deficient glycoprotein syndrome: evidence of asparagine-N-linked oligosaccharide transfer deficiency. J. Biol. Chem. 268:5783-5789.
29. Zentilin Boyer, M., P. de Lonlay, N. Seta, M. Besnard, C. Pelatan, H. Ogier, J. P. Hugot, C. Faure, J. M. Saudubray, J. Navarro, and J. P. Cezard. 2003. Failure to thrive and intestinal diseases in congenital disorders of glycosylation. Arch. Pediatr. 10:590-595.[CrossRef][Medline]
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| J. Bacteriol. | J. Virol. | Eukaryot. Cell |
|---|
| Microbiol. Mol. Biol. Rev. | Clin. Vaccine Immunol. | All ASM Journals |
|---|
