Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
DOCTOR OF MEDICAL SCIENCE Congenital disorders of glycosylation type Ia and Ib Genetic, biochemical and clinical studies Susanne Kjærgaard This review has been accepted as a thesis together with eight previously published papers, by the University of Copenhagen, March 31, and defended on August 20, 2004. H:S Rigshospitalet, Department of Clinical Genetics, Copenhagen. Correspondence: Susanne Kjærgaard, John F. Kennedy Instituttet, Gl. Landevej 7, 2600 Glostrup, Denmark. Official opponents: Professor, PhD Niels A. Jensen and Associate Professor, Doctor of Dental Surgery Henrik Clausen. Dan Med Bull 2004;51:350-63. 1. INTRODUCTION Congenital disorders of glycosylation (CDG), formerly known as carbohydrate-deficient glycoprotein syndromes, are a rapidly expanding group of inherited multisystem disorders resulting from defects in N-linked carbohydrate side-chains (glycans) of proteins. The molecular basis of several types of CDG has been elucidated during the past few years, and 12 types have been recognized by February 2003. CDG type Ia (CDG-Ia, OMIM #212065), by far the most common, is the focus of this thesis, and less emphasis is placed on CDG type Ib (CDG-Ib, OMIM #602579). CDG-Ia is caused by mutations in PMM2 and the resulting deficiency of the enzyme phosphomannomutase (PMM). Clinical manifestations include psychomotor retardation, hypotonia, failure to thrive, ataxia, liver dysfunction and dysmorphic features. CDG-Ib is due to mutations in MPI, which lead to deficiency of the enzyme phosphomannoisomerase (PMI). Manifestations of CDG-Ib include hepatic fibrosis, protein-losing enteropathy, thrombosis and hypoglycemia, but no primary neurological impairment. PMM and PMI both play a role in maintaining the intracellular supply of mannose derivatives required for glycosylation of proteins, and deficiency of one of the enzymes leads to hypoglycosylation. lular compartments: the cytosol, the rough endoplasmic reticulum (rER) and the Golgi apparatus (Figure 1). In the first part of this pathway PMM and PMI participate in the synthesis of guanosine diphosphomannose (GDP-mannose) from mannose-6-P. The nucleotide sugars donate sugars attached to the dolichol-pyrophosphate to form the lipid-linked oligosaccharide (LLO) precursor. Two N-acetylglucosamines and five mannoses are added to the LLO precursor, which is anchored to the cytosolic membrane of the rER. Subsequently, the LLO precursor is flipped to the luminal side of rER, where more sugars are added sequentially using dolichol-Pmannose/glucose as donors. When the branched oligosaccharide moiety contains two N-acetylglucosamines, nine mannoses and three glucoses, it is linked to selected asparagine residues of the nascently translated polypeptide. Next, three glucoses and one mannose are removed sequentially, and the glycoprotein is transferred to the Golgi apparatus for final modification. Six additional mannoses are removed and replaced by two or more each of N-acetylglucosamine and galactose, and the biosynthesis of the oligosaccharide branches are terminated by the addition of sialic acid residues. Certain glycoproteins are further modified by fucosylation or phosphorylation. Finally, the glycoproteins are secreted from the Golgi to their sites of action. Secretory glycoproteins are secreted into the body fluids, membrane glycoproteins are incorporated in cell membranes, and lysosomal enzymes are taken up by lysosomes (Varki et al, 1999, Jaeken et al, 2001). Glycans participate in the folding of glycoproteins in the rER and modify their stability, action and clearance. Glycans are essential for the function of cell surface receptors, as signals for protein targeting, mediators of cell-to-cell interaction, mediators of protein turnover and as protection of polypeptides from proteases (Varki et al, 1999). 2.2. CDG TYPES AND NOMENCLATURE Several human disorders due to defects in the N-glycosylation pathway have recently been characterised. The disorders fall into two groups based on the localisation of the enzymatic defect (Participants, First International Workshop on CDGS 1999). Group I CDG (CDG-I) refers to defects in the initial steps of N-linked protein glycosylation. These defects affect the assembly of LLO and/or its transfer to asparagine residues on nascent polypeptides. Group II CDG (CDG-II) refers to defects in the processing of protein-bound glycans or the addition of other glycans to the polypeptides. Untyped patients are assigned to CDG-x until their genetic defect is established. Currently known CDG types, the corresponding enzyme deficiencies, and the genes are listed in Table 1. The aims of the studies were: 1. to characterise Danish CDG-Ia patients biochemically and molecularly 2. to study the effect of PMM2 mutants on phosphomannomutase function in vitro 3. to investigate the origin of the predominant PMM2 mutations R141H and F119L by haplotype analysis 4. to characterise the clinical phenotype of CDG-Ia due to the predominant R141H/F119L PMM2 genotype 5. to study the effect of mannose therapy in patients with CDG-Ia and CDG-Ib. 2. BACKGROUND 2.1 BIOSYNTHESIS AND FUNCTIONS OF N-GLYCANS The widespread post-translational glycosylation of proteins affect their properties and functions. Most secreted proteins and membrane proteins (receptors and transporters) as well as some intracellular proteins (e.g., lysosomal enzymes) have glycan side chains, which are added to asparagine (N-glycosylation) or serine/threonine polypeptide residues (O-glycosylation). Involving 30-40 enzymatic and transport steps, N-glycosylation runs through three cel350 2.3. ISOELECTRIC FOCUSING OF TRANSFERRIN – A SCREENING TEST FOR CDG An individual glycoprotein has a heterogeneous but consistent population of N-linked glycans, which can be separated and visualised by isoelectric focusing (IEF) and Western blotting/immunoprecipitation (Stibler et al, 1998, Niehues et al, 1998). Various glycoforms of the protein are each represented by a band in the IEF pattern. Transferrin is only one of many hypoglycosylated proteins in CDG, but its microheterogeneity on IEF has turned out to be a sensitive biochemical marker of most N-glycosylation defects (Stibler et al, 1998). Normally, the predominant glycoform of transferrin contains four oligosaccharide branches (tetrasialotransferrin) on two N-linked glycans. Hypoglycosylation of proteins results in partial deficiency of the terminal, negatively charged sialic acid, and hence in a cathodal shift of the IEF pattern. In patients with CDG-I the proportions of glycoforms with zero (asialo-) and two (disialotransferrin) oligosaccharide branches are increased with a corresponding decrease in tetrasialotransferrin – a “type I” pattern (Figure 2) (Stibler et al, 1991b). Dried blood spots and serum can be used for IEF (Stibler and Cederberg 1993). A false-positive IEF pattern may be seen when DANISH MEDICAL BULLETIN VOL. 51 NO. 4/NOVEMBER 2004 Man Man Glc Glc Man Man Glc Glc PMI Cytosol Ib Fru-6-P Fru -6 -P Man Man-6-P -6 -P Glc-6-P Glc -6 -P PMM Ia GDP-Man GDP - Man Man-1-P Man -1 -P Man Man P2 P2 P2 Man P P Ie If Man55 Man Man6 P2 P2 Man77 Man Man88 Man P2 P2 Id P2 Ig Man Man 55 Glc Glc Glc 3 3 Glc22 Glc Glc Man 99 Man Man 99 Man 9 9 Man Man 99 Man P2 P2 Ic Golgi IIc GDP-fuc rER P Man Man GDP-fuc P2 Glc Glc 33 Man Man 99 Glc Glc 22 Man Man 99 Glc Glc Man Man 99 Man 99 Man Man88 Man Man 33 Man Man 55 Man Man 55 Man Man88 Man P2 Ih IIb Sia Sia 22 Gal Gal 22 Gal 2 2 Man Man 33 Man 33 Man Man33 Man IId IIa Man: Mannose; Glc: Glucose; Fru: Fructose; Gal: Galactose; Fuc: Fucose; Sia: Sialic acid Dolichol N-Acetylglucosamine Transporter Figure 1. N-linked protein glycosylation pathway (simplified). The currently known defects are indicated by red. using EDTA-anticoagulated plasma, and abnormal IEF pattern not related to CDG may be seen in serum from patients with untreated galactosemia, untreated hereditary fructose intolerance or alcohol abuse (Jaeken et al, 1996, Stibler 1991a, Stibler et al, 1997). Amino acid polymorphisms of transferrin can alter its IEF pattern (Stibler et al, 1988). IEF of serum transferrin from parents of a CDG-suspected individual will differentiate between a transferrin protein variant and abnormal glycosylation. Normal IEF pattern has been reported in a few cases with enzymatically and genetically proven CDG-Ia (Fletcher et al, 2000, Dupre et al, 2001). Abnormal glycosylation of transferrin (and other glycoproteins) tends to improve slightly with increasing age in CDG-Ia, which might obscure the diagnosis in a minority of older patients (Dupre et al, 2001, Stibler et al, 1998). It is remarkable, and still unexplained, that abnormal glycosylation of serum transferrin (and other glycoproteins) is not detectable by IEF in CDG-Ia fetuses and preterm newborns until the 36th gestational week (Clayton et al, 1993, Stibler and Skovby 1994a). IEF analysis cannot differentiate types of CDG, and additional biochemical and genetic analyses are needed to reach a specific diagnosis. Quantitative determination of Carbohydrate-Deficient Transferrin (CDT) is an alternative screening method, which measures the amount of di-, mono- and asialotransferrin (Stibler et al, 1991a). However, elevated values need to be confirmed by IEF. 2.4. PATIENTS By February 2003, 32 patients from 26 unrelated Danish families DANISH MEDICAL BULLETIN VOL. 51 NO. 4/NOVEMBER 2004 had been diagnosed with CDG-Ia, 17 of whom were born in 19901998. This corresponds to an incidence of 1:41 000 newborns. We know of two additional, deceased patients with clinical manifestations highly suggestive of CDG-Ia, but biological materials were not available for confirmation. So far only two sibs with CDG-Ib and one patient with CDG-Ig have been identified. All these patients had a “type I“ pattern IEF of serum transferrin (Figure 2) (Kjaergaard et al, 1998a, Westphal et al, 2001b, Grubenmann et al, 2002). 3. CONGENITAL DISORDER OF GLYCOSYLATION TYPE IA 3.1. PHOSPHOMANNOMUTASE DEFICIENCY Phosphomannomutase deficiency was established as the cause of CDG-Ia in 1995 by the detection of PMM activities ≤ 10% in fibroblasts/leukocytes/liver tissue from patients and of intermediate activities in their parents (Van Schaftingen and Jaeken 1995). PMM is a cytosolic enzyme that catalyses the conversion of mannose-6-P to mannose-1-P (Figure 1). PMM requires mannose-1,6-biphosphate or glucose-1,6-biphosphate as a cofactor, which serves to phosphorylate the catalytic site of PMM. Mannose-1-P is required for the synthesis of GDP-mannose, the donor of mannose to the growing LLO precursor. PMM deficiency results in depletion of the mannose-1-P and GDP-mannose pools in fibroblasts (Körner et al, 1998a, Rush et al, 2000). Hence, the incorporation of mannose into LLO is reduced, and glycoproteins lacking complete N-linked glycans are synthesised (Wada et al, 1992, Yamashita et al, 1993, Powell et al, 1994, Krasnewich et al, 1995). Fibroblasts with PMM deficiency do not accumulate mannose-6-P, which suggests that 351 Table 1. Currently known CDG types, corresponding enzyme deficiencies, and genes. Type Deficiency Gene Reference CDG-Ia Phosphomannomutase PMM2 Van Schaftingen 1995 Matthijs 1997 CDG-Ib Phosphomannoisomerase MPI Niehues 1998 Jaeken 1998 CDG-Ic Dolichyl-P-Glc:Man9GlcNAc2-PP-dolichyl α-1,3-glucosyltransferase ALG6 Körner 1998 Imbach 1999 CDG-Id Dolichyl-P-Man:Man5GlcNAc2-PP-dolichyl α-1,3-mannosyltransferase ALG3 Körner 1999 CDG-Ie Dolichol-P-Man synthase 1 DPM1 Imbach 2000 Kim 2000 CDG-If Dolichol-P-Man utilisation defect MPDU1 Schenk 2001 Kranz 2001 CDG-Ig Dolichyl-P-Man:Man7GlcNAc2-PP-dolichyl α-1,6-mannosyltransferase. ALG12 Chantret 2002a Grubenmann 2002 CDG-Ih Dolichyl-P-Glc:Glc1Man9GlcNAc2-PP-dolichyl α-3-glucosyltransferase ALG8 Chantret 2003 CDG-IIa UDP-GlcNAc:α-6-D-mannoside β-1,2-Nacetylglucosaminyltransferase II (GlcNAcT II) MGAT2 Jaeken 1994 Tan 1996 CDG-IIb α-1,2-glucosidase I GCS1 De Praeter 2000 CDG-IIc GDP-fucose transporter FUCT1 Lübke 2001 CDG-IId N-acetylglucosamine β-1,4-galactosyltransferase I B4GALT1 Hansske 2002 CDG-x Genetic basis unknown PMI maintains an equilibrium between mannose-6-P and fructose6-P (Figure 1) (Körner et al, 1998a). PMM activities in extracts of cultured fibroblasts from Danish CDG patient have been measured by the assay described by Van Schaftingen and Jaeken (1995): the reduction of NADP to NADPH was measured spectrophotometrically following coupled enzymatic reactions catalysed by PMI, phosphoglucoisomerase and glucose-6-P dehydrogenase, and mannose-1-P was used as substrate (Figure 3). PMM activities in cultured fibroblasts ranged from undetectable to 15% of controls, thereby documenting PMM deficiency in all patients (Kjaergaard et al, 1998a). Grünewald et al, (2001) pointed out a possible pitfall in the diagnosis of PMM deficiency after observing CDG-Ia patients with high residual PMM activity overlapping control values in cultured fibroblasts. This problem was not encountered in assays of fresh leukocytes and liver tissue. The variability of PMM activity among different cell types points to a cell-specific effect, as yet unexplained. One possibility is that mutant PMM or the isozyme (see below) become Control overexpressed in rapidly dividing fibroblasts with high protein synthesis as opposed to fully differentiated leukocytes with little protein synthesis (Grünewald et al, 2001). The use of fresh leukocytes does not always circumvene difficulties of interpretation of results since PMM values of some carriers overlap with control or patient values (Grünewald et al, 2001). PMM is a homodimer of 30 kDa subunits, whose tertiary structure is not yet known (Kepes and Schekman 1988, Pirard et al, 1997). Two isozymes of PMM with 66% sequence homology have been identified: PMM1 and PMM2. They catalyse the same reaction at approximately equal rates, but PMM2 is more substrate-specific (Pirard et al, 1999a). PMM2 is the defective enzyme in CDG-Ia. The role of PMM1 is still a puzzle, and no disorder has so far been associated with defects of PMM1. Studies in normal rat on the tissue distribution/expression of PMM1 and PMM2 by Western/Northern blot analyses showed that PMM2 is the only isozyme detectable in most tissues except brain and lung, where PMM1 accounts for approximately 66% and 13% of the total activities, respectively (Pirard CDG Ia CDG Ib CDG Ig + Tetrasialotransferrin Figure 2. IEF pattern of serum transferrin from a normal control, CDG-Ia-, CDG-Ib- and CDG-Ig patient. The number of sialic acid residues (negative charges) is given at the right. The anode and catode are given at the left. The duplicate bands in the CDG-Ig lane are probably due to a genetic variant of transferrin at the polypeptide level. 352 Disialotransferrin Asialotransferrin – DANISH MEDICAL BULLETIN VOL. 51 NO. 4/NOVEMBER 2004 Man Man Glc Glc Hexokinase Hexokinase Man-6-P Man -6 -P Phosphoglucoisomerase PMM 66-P-gluconate - P - gluconate GlcGlc-6-P -6 -P FruFru-6-P -6 -P PMI NADP NADPH Glucose-6-Pdehydrogenase Man-1-P Figure 3. Enzymatic reactions involved in the assays for determination of PMM activity in fibroblasts and of serum mannose. Man: Mannose; Glc: Glucose; Fru: Fructose et al, 1999a). In contrast, Northern blot analysis of normal human tissues showed widespread expression of both PMM1 and PMM2 (Matthijs et al, 1997a and b). The highest expression of PMM2 was seen in pancreas and liver, organs with a predominant role in glycoprotein synthesis. PMM2 was weakly expressed in brain, unlike PMM1. How these observations correlate with the severe CNS involvement of CDG-Ia is not known. However, the low PMM activity measured in liver, fibroblasts and leukocytes of CDG-Ia patients indicates that PMM1 contributes little to the total PMM activity in these cell types, and that it does not rescue PMM2 deficiency. 3.2. THE PMM2 GENE CDG-Ia is inherited as an autosomal recessive trait, and a locus was assigned to chromosome 16p13 by linkage analysis (Jaeken et al, 1991, Martinsson et al, 1994). Two genes both encoding a PMM have been cloned based on their homology to yeast PMM (SEC53) (Kepes and Schekman 1988). The first gene, PMM1, is localised on chromosome 22q13, which excludes its involvement in CDG-Ia (Matthijs et al, 1997a, Wada et al, 1997). The second gene, PMM2, is localised on 16p13, and the detection of mutations in this gene showed conclusively that PMM deficiency is responsible for CDG-Ia (Matthijs et al, 1997b). PMM2 contains 8 exons, and its cDNA has an open reading frame of 738 base pairs encoding a protein of 246 amino acids. The high homology (57%) at both the nucleotide and the amino acid level between yeast PMM (SEC53) and PMM2 (and PMM1) indicates a strong conservation from yeast to man, reflecting a fundamental role of PMM (Matthijs et al, 1997b). A processed pseudogene, PMM2ψ, located on chromosome 18p, has also been identified. Caution is therefore required when designing primers for mutation analysis, as identical mutations have been reported in PMM2 and PMM2ψ (Schollen et al, 1998). Figure 4. Position of the reported PMM2 mutations. Mutations underlined have been identified in Danish patients with CDG-Ia. DANISH MEDICAL BULLETIN VOL. 51 NO. 5’ 4/NOVEMBER 2004 24delC C9Y F11C G15R G15E 1 3.3. PMM2 MUTATIONS Molecular analysis has revealed a large number of mutations in CDG-Ia patients worldwide. Seventy-one PMM2 mutations have been reported in patients from 24 countries, including Denmark (Figure 4) (Kjaergaard et al, 1998a and 1999, Kondo et al, 1999, Matthijs et al, 2000, Grünewald et al, 2001, http://www.med.kuleuven.ac.be/cdg/Mutation/cdg-Ia.htm). The mutations are scattered over the gene with some clustering in exon 5 and 8. Sixty-four of 71 different mutations are of the missense type, the others being three single base pair deletions, one four base pair deletion, two splice site mutations and one nonsense mutation. The mutations in the Danish CDG-Ia patients were identified by Single Strand Conformation Analysis (SSCA) on genomic DNA after PCR amplification with intronic primers followed by direct DNA sequencing and confirmation by restriction analysis (Kjaergaard et al, 1998a and 1999). The seven mutations identified so far are all of the missense type and result in replacement of amino acids that are strictly conserved among phosphomannomutases of yeast (SEC53), Candida albicans and human phosphomannomutases (PMM1 and PMM2), suggesting that these amino acids are important for the catalytic activity of the enzyme. Two mutations accounted for 88% of the mutations: 44% of the CDG-Ia alleles harboured the R141H mutation, 44% of the CDG-Ia alleles harboured the F119L mutation, and 83% of the patients were compound heterozygous for these mutations (Kjaergaard et al, 1998a). Surprisingly, we did not observe homozygosity for either of the two predominant mutations. The absence of homozygotes expected on the basis of Hardy-Weinberg equilibrium was statistically highly significant. Subsequently, we and others have detected homozygosity for F119L, but homozygosity for R141H has been conspicuously absent in results of genotype analysis (Matthijs et al, 1998a and 2000, Kjaergaard et al, 1998a and 1999, Bjursell et al, 2000, Imtiaz et al, 2000, De Lonlay et al, 2001). These observations IVS1-1G>A L32R G42R V44A Y64C D65Y V67M P69S Y76C IVS3+2T>C E93A N101K C103F L104V Y106C A108V 324delG P113L 2 3 4 G117R F119L I120T R123Q R123X V129M P131A I132T 451 -454delGAAA E151G I132N I153T 389delC F157S E139K R141H R162W F144L F172V D148N G175R 5 6 F183S D185G D188G C192G H195R E197A F206T G208A G214S N216I N216S D217E H218L D223E D223N T226S G228R G228C Y229S V231M A233T T237R T237M R238G R238P R239W C241S 7 8 3’ 353 suggest that homozygosity for R141H is not compatible with life or, less likely, leads to another clinical phenotype. 3.4. RELATIONSHIP BETWEEN PMM2 GENOTYPE AND RESIDUAL PMM ACTIVITY The extensive allelic heterogeneity worldwide and the somewhat variable residual PMM activity in different cell types from the same patient complicate the study of the correlation between genotype and residual PMM activity. Furthermore, PMM activity is low in extracts of control fibroblasts, and differentiation of PMM activities below 0.5 nmol/mg/min is not reliable, which make it difficult to compare residual activities of various mutant fibroblast cell lines. We and others observed profoundly decreased PMM activity in fibroblasts from all patients (Kjaergaard et al, 1998a, Imtiaz et al, 2000). In general, there was no clear correlation between genotype and residual PMM activity except for the R141H/F119L genotype, which invariably was associated with very low PMM activity (Kjaergaard et al, 1998a, Grünewald et al, 2001). Certain genotypes (R141H/C241S, F157S/C241S, R141H/T237M) were associated with a relatively high residual PMM activity in fibroblasts (30-70% of controls) (Grünewald et al, 2001). 3.5. IN VITRO EXPRESSION STUDIES To investigate the impact of PMM2 mutations on the catalytic activity of the enzyme, wild type and mutant PMM2 were expressed in E. coli (Kjaergaard et al, 1999, Pirard et al, 1999b). We cloned PMM2cDNA into a pET3a expression vector, and the six different mutations identified at the time (R141H, F119L, C192G, G117R, T237R and D223E) were each introduced into PMM2-cDNA by site-directed mutagenesis. Specific activities of recombinant PMM2 suggested the presence of two types of mutations. One type resulted in virtually null activity, and the other type resulted in PMM2 with decreased activity and/or altered properties such as affinity for the substrate and stability. The R141H protein had no residual PMM2 activity, and the F119L protein had approximately 25% of normal activity, higher Km for mannose-1-P and less thermostability (Kjaergaard et al, 1999, Pirard et al, 1999b). The G117R and T237R proteins also had virtually no residual PMM2 activity, whilst the D223E and C192G proteins had activities in the normal range but higher Km’s for mannose-1-P and less thermostability (Kjaergaard et al, 1999). The recombinant V231M protein has been shown to retain residual PMM2 activity (Pirard et al, 1999b). Each of our patients has at least one mutation that retains residual PMM2 activity when expressed in E. coli (Table 2). However, this activity is not always reflected by activities depending on two mutant PMM2 alleles found in cultured fibroblasts from the patients (Kjaergaard et al, 1999, Pirard et al, 1999b). The activity in cultured fibroblast of some but not all genotypes was less than expected from the activity of the recombinant protein, which might indicate poor stability of the mutant PMM2 and/or a dominantnegative effect of the null activity polypeptide. Even fibroblasts homozygous for the F119L had very low or undetectable PMM activ- Table 2. PMM2 genotypes identified in 26 unrelated Danish CDG-Ia patients. PMM2 genotype No. of patients R141H/F119L . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 R141H/C192G . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 R141H/V231M . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 F119L/F119L . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 F119L/G117R . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 F119L/V231M . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 T237R/D223E . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 The recombinant proteins corresponding to the mutations underlined have no detectable PMM activity. 354 ity (Kjaergaard et al, 1999). The lack of correlation may also be explained by difficulties in comparing activities due to the relatively low sensitivity of the PMM assay and the low PMM activity in control fibroblasts. So far, no CDG-Ia patient has been found to have two PMM2 mutations conveying virtually no PMM2 activity when expressed in vitro (Kjaergaard et al, 1999, Pirard et al, 1999b, Vuillaumier-Barrot et al, 1999 and 2000, Westphal et al, 2001a). Even though one should be cautious applying the results of in vitro studies in E. coli to the in vivo situation in man, the expression studies support the hypothesis that homozygosity for R141H (or another “severe” mutation) impairs PMM2 activity to a degree incompatible with life, whilst F119L (or another “mild” mutation) conveys a phenotype compatible with survival due to residual PMM2 activity. The fate of zygotes homozygous for R141H is unknown. The absence of patients harbouring two null alleles points to an essential role of PMM2 in normal cellular functions, and suggests that such zygotes are lost in early pregnancy. An absolute requirement for residual activity is compatible with the glycosylation, however deficient, of circulating proteins in CDG-Ia patients and with the apparent lack of other biopathways to synthesize mannose-1-P and GDP-mannose. Within one of the three extremely conserved protein sequences (motifs) delineated in PMM, the phosphorylation site (see p. 7) has been assigned to an aspartate residue at position 12 (Aravind et al, 1998, Collet et al, 1998). The motifs may be involved in the catalytic site of PMM, but none of the mutations identified in Danish patients affects the amino acids within these motifs. Indeed, mutations distant to the presumed catalytic site may influence conformation and thereby activity of the enzyme. More knowledge of the structure of PMM is needed to predict the impact of a mutation on the enzymatic function of PMM2. 3.6. GENETIC EPIDEMIOLOGY CDG-Ia occurs worldwide. The reported incidence is highest in Denmark (1:41 000 newborns) followed by Sweden (1:80 000), the Netherlands and Belgium (Matthijs et al, 2000, Kjaergaard et al, 2001). R141H, by far the most frequent PMM2 mutation, is present in 73% of Caucasian patients (Matthijs et al, 2000). R141H and F119L make up the vast majority of mutations of Scandinavian patients (Kjaergaard et al, 1998a, Bjursell et al, 2000). Genotypes are much more heterogeneous in other Western European countries (Matthijs et al, 2000). The predominance of two mutations among Danish patients reflects the genetic homogeneity of the Danish population also seen in other diseases, e.g., cystic fibrosis (Schwartz et al, 1993) and galactosemia (M. Schwartz, unpublished data). Prior to cloning of the PMM2 gene one specific haplotype was shown to be markedly overrepresented among patients from Southwestern Sweden, Southern Norway and Eastern Denmark, indicating one shared ancestral mutation (Bjursell et al, 1997). In order to investigate the origin of the two predominant mutations, haplotype analysis of the Danish CDG-Ia families was performed using markers defining the Scandinavian founder haplotype and one additional closely linked marker (D16S3020, located within 20kb of PMM2) (Kjaergaard et al, 1998a). R141H and F119L were each associated with a specific haplotype. Our suggestion that the Scandinavian founder mutation was F119L was subsequently confirmed (Kjaergaard et al, 1998a, Bjursell et al, 1998). Haplotype analysis of CDGIa patients with the R141H mutation from other Western European countries also revealed a clear linkage disequilibrium for the same haplotype as seen in the Danish patients (Schollen et al, 2000a). These haplotypes indicated that R141H and F119L originated from two ancestral mutations. R141H showed more recombinations with some of the more distant markers than F119L, suggesting that R141H is older than F119L. The frequency of F119L was highest among Scandinavian patients, and it gradually decreased when moving southward in Europe (Matthijs et al, 2000). The F119L mutation may well have arisen in a Scandinavian ancestor. R141H was DANISH MEDICAL BULLETIN VOL. 51 NO. 4/NOVEMBER 2004 frequent among patients from all Western European countries and the US compatible with an ancient origin (Matthijs et al, 2000). The lack of patients homozygous for R141H was further explored by determining the frequency of R141H in two normal populations. The carrier frequency in Dutch neonates and in Danish blood donors was 1:79 and 1:60, respectively (Schollen et al, 2000a). These frequencies were not significantly different, and pooled data gave a frequency of 1:72. Under Hardy-Weinberg equilibrium an incidence of R141H homozygotes of approximately 1:20 000 would be expected. These observations further support that R141H in the homozygous state is not compatible with life. Both populations were also analysed for the second most frequent mutation, F119L. No F119L carriers were identified, indicating a much lower carrier frequency of this mutation than of R141H (Schollen et al, 2000a). 3.7. PRENATAL DIAGNOSIS Prior to the identification of PMM2 as the gene involved in CDG-Ia, prenatal diagnosis was based on linkage analysis and measurement of PMM activity in amniotic and chorionic villus cells (Bjursell et al, 1998, Charlwood et al, 1998, Matthijs et al, 1998b). Problems with the risk of error due to recombination and with interpretation of enzyme activity were avoided when mutation analysis became available. Prenatal diagnosis by mutation analysis of chorionic villus cells is now routinely performed and widely requested by affected families in Denmark. So far, we have performed 18 prenatal analyses: seven fetuses were compound heterozygotes, and these pregnancies were terminated. Seven fetuses were heterozygote carriers, and four were homozygous normal (unpublished data). Seven compound heterozygotes out of 18 at-risk pregnancies is a high proportion of affected fetuses compared to the expected 25%, but statistically not significant. We are in the process of pooling results of prenatal mutation analyses with other diagnostic centres to estimate the recurrence risk in a larger sample of CDG-Ia families. 3.8. CLINICAL PRESENTATION CDG-Ia was first described by Jaeken et al, in 1980. The clinical features vary with age, and usually go through four stages: the infantile alarming multisystem stage, the childhood ataxia-mental retardation stage, the teenage leg atrophy stage and the adult hypogonadic stage (Table 3) (Jaeken et al, 1991). The first stage is variable in severity, but often dramatic, requiring frequent hospitalisations. The child becomes more stable with increasing age, and loss of acquired skills is unusual. Most patients are wheelchair-bound, but motor and communicative skills vary. The overall mortality rate is approximately 20%, mostly occurring in the first years of life due to infection, multiple organ failure or heart tamponade (Jaeken et al, 1991 and 2001). The clinical spectrum of CDG-Ia has expanded since the initial description. Moderate to severe neurological dysfunction, variable dysmorphic features and variable involvement of other organs are now recognised, and a phenotype with only borderline cognitive impairment has been reported recently (Jaeken et al, 2001, Van Ommen et al, 2000, Drouin-Garraud et al, 2001, Enns et al, 2002). Rare presenting features, such as hyperinsulinemic hypoglycemia, congenital nephrotic syndrome and obstructive cardiomyopathy, have also been reported (Van der Knaap et al, 1996, Imtiaz et al, 2000, Böhles et al, 2001). 3.9. BIOCHEMICAL FEATURES A large number of serum glycoproteins have been shown to have abnormal IEF pattern and/or altered concentration or enzymatic activity, including transport proteins, glycoprotein hormones, coagulation and anticoagulation factors, lysosomal enzymes and enzyme inhibitors (Jaeken et al, 2001). Clinically important biochemical features include: DANISH MEDICAL BULLETIN VOL. 51 NO. 4/NOVEMBER 2004 Table 3. Clinical stages of CDG-Ia. Infantile alarming multisystem stage 1 Failure to thrive Floppiness Psychomotor retardation Abnormal subcutaneous fat pads Inverted nipples Hepatomegaly and liver dysfunction Cerebellar atrophy Esotropia Pericardial effusions Multisystemic failure Childhood ataxia mental retardation stage 2 Mental retardation (IQ: 40-60) Motor disability Peripheral neuropathy Cerebellar ataxia Tapetoretinal degeneration Stroke-like episodes Seizures Teenage leg atrophy stage 3 Atrophy of lower limbs Stable cerebellar ataxia Deformity of thorax and spine Osteopenia Absence of puberty in females (hypergonadotropic hypogonadism) Adult stable disability stage 4 Stable neurological impairment Short stature Kyphoscoliosis and long thin extremities Premature aging of the skin in females Retinitis pigmentosa 1. Decreased activity of factor XI and decreased activities of the coagulation inhibitors ATIII and protein C are present in the vast majority of CDG-Ia patients, and abnormal glycoforms of the latter two have been shown (Van Geet and Jaeken 1993, Stibler et al, 1996, Kjaergaard et al, 2001). A complex coagulopathy may be responsible for stroke-like episodes, thromboembolic events and hemorrhages seen in some patients. 2. Adolescent females have elevated levels of luteinising hormone and follicle-stimulating hormone (De Zegher and Jaeken 1995). Absence of puberty and hypergonadotropic hypogonadism are consistent features of CDG-Ia girls with few exceptions. Kristiansson et al, (1995) were not able to demonstrate hypoglycosylation of gonadotropins. Oestradiol secretion in response to exogenous gonadotropins has suggested biological inactivity of FSH, but the lack of response in one Danish patient confirmed that primary ovarian failure might also be responsible for absence of puberty (De Zegher and Jaeken 1995, Kristiansson et al, 1995, Kjaergaard et al, 2001). Oestradiol therapy induced growth of breast tissue and pubic hair in two Danish females (Kjaergaard et al, 2001). 3. Lysosomal enzyme activities in serum are frequently elevated (e.g., arylsulphatase A, β-galactosidase, β-glucuronidase and βhexosaminidase). In contrast, lysosomal enzyme activities are frequently low in leukocytes (e.g., α-fucosidase, β-glucuronidase, α-iduronidase, α- and β-mannosidase) (Barone et al, 1998). These observations may be explained by deficiency of the mannose-6-P targeting signal necessary for their uptake in lysosomes. Abnormal glycoforms of N-acetylglucosaminidase and fucosidase have been demonstrated (Jaeken et al, 1991). Findings of slightly to moderately increased lysosomal enzyme activities in plasma and/or moderately decreased activities in leukocytes from a patient with neurological impairment should alert the clinician to the possibility of CDG. 4. Abnormal glycosylation of serum transferrin, thyroid-binding globulin and α1-antitrypsin are consistently observed (Stibler et 355 al, 1998). These glycoproteins were used as parameters in our mannose therapy study (see p. 12). 5. Consistent biochemical abnormalities of non-glycosylated proteins include raised liver transaminases and hypoalbuminemia (Jaeken et al, 1991, Kjaergaard et al, 2001). Liver transaminases usually rise during infections, but tend to normalise with increasing age. Hypoalbuminemia may contribute to the extravascular fluid accumulations, e.g., pericardial effusion and ascites, seen in some patients. 3.10. CLINICAL PHENOTYPE IN RELATION TO GENOTYPE Most reports on the clinical manifestations of CDG-Ia predate the availability of PMM2 mutation analysis (Kristiansson et al, 1989, Jaeken et al, 1991, Petersen et al, 1993, Stibler et al, 1994b). A recent mutation update revealed 73 different PMM2 genotypes among 249 CDG-Ia patients, including Danish patients, and the R141H/F119L genotype accounted for 27% of the patients (Matthijs et al, 2000). The extensive genetic heterogeneity has complicated attempts to study genotype-phenotype correlation (Imtiaz et al, 2000, De Lonlay et al, 2001). The predominance of the R141H/F119L genotype in our large series of CDG-Ia patients allowed us to study its clinical phenotype and outcome (Kjaergaard et al, 2001). The features of 23 patients are listed in Table 4. All patients had an early, uniform presentation with severe feeding problems, hypotonia, inverted nipples, abnormal subcutaneous fat pads and developmental delay obvious before age 6 months. Motor ability ranged from none to walking with a rolator, and vocabulary ranged from none to comprehensible speech. The overall mortality ascribed to CDG was 18%. Brain imaging showed cerebellar atrophy in all but two patients. Cerebral atrophy was frequently present, and serial imaging indicated that the cerebellar and cerebral atrophy ran a progressive course in some patients, as previously observed (Jensen et al, 1995). Table 4. Clinical features of CDG-Ia patients with R141H/F119L genotype. Consistent features (present in >90% of the patients) Neonatal presentation Severe feeding difficulties Severe failure to thrive Severe hypotonia Developmental delay obvious before age 6 months Hepatic dysfunction (raised liver transaminases and coagulation abnormalities) Inverted nipples Abnormal subcutaneous fat pads Esotropia Delayed visual maturation Mental retardation Cerebellar atrophy Ataxia Muscular atrophy Peripheral neuropathy Febrile seizures Absence of puberty in females Frequent manifestations (present in ≥ 50% of the patients) Pericardial effusions (asymptomatic) Afebrile seizures Stroke-like episode Acquired microcephaly Congenital joint restriction Hernia/hydrocele/retentio testis in males Thorax deformity Myopia Decreased vision Less frequent (present in < 50% of the patients) Neurogenic hearing impairment Thromboembolic event Episode of prolonged bleeding Episode of ascites Kyphoscoliosis Fracture due to minimal trauma 356 The mean head circumference standard deviation score declined gradually from normal at birth to –1.9 at age 5 (Kjaergaard et al, 2002). Eye manifestations were evaluated during another Danish survey (Jensen et al, 2003). Patients with the R141H/F119L genotype had congenital esotropia, delayed visual maturation and a reduced rod response by electroretinography. The vast majority also had low visual acuity and progressive myopia. Failure to thrive is a frequent feature of CDG-Ia, although the onset and degree in individual patients have been variable (Jaeken et al, 1991, Petersen et al, 1993, Kristiansson et al, 1998). Feeding problems and failure to thrive were prominent in all our patients, and we analysed longitudinal data of weight and length to delineate the pattern of prepubertal growth (Kjaergaard et al, 2002). Anthropometric data of patients with identical CDG genotypes were remarkably similar. Measurements of weight and length were within normal range at birth. During the first 7 months of life, standard deviation scores of mean weight and length declined markedly, 2.7 and 2.4, respectively, and there was no prepubertal catch up. Our patients fed poorly, and the majority had frequent vomiting and diarrhoea during the first year of life. Although data on their protein and caloric intake and gastrointestinal losses were not available, inadequate nutrition is the most likely major explanation of the severe failure to thrive. Hormones and other factors involved in growth are likely to be affected by defective glycosylation, but previous studies have not been able to hold growth hormone or thyroid hormones responsible (Macchia et al, 1995, De Zegher and Jaeken. 1995). Controlled dietary and hormonal interventional studies as well as studies of intestinal protein loss are required to further explore the causes of growth failure. Patients with the R141H/F119L genotype have an early “classical” presentation, including severe failure to thrive, but with variable functional outcome, and they may well represent the severe end of the broad clinical spectrum (Kjaergaard et al, 2001, Grünewald et al, 2001). Although few patients share other genotypes, some of these seem to be associated with severe disease and high mortality (R141H/D188G, R141H/V231M) or mild disease (R141H/C251S, R141H/F183S, F157S/C241S) (Matthijs et al, 2000, Grünewald et al, 2001, Erlandson et al, 2001, Drouin-Garraud et al, 2001). In general, the PMM2 genotype does not seem to be a good predictor of clinical/functional outcome in individual CDG-Ia patients. Clinical variability, especially of functional outcome, was evident among patients with identical PMM2 genotypes and even among affected sibs and monozygotic twins (Kjaergaard et al, 2001). The lack of any clear genotype-phenotype correlation parallels other diseases, e.g., Gaucher disease (Dipple and McCabe 2000). The interand intrafamilial variability in CDG-Ia suggests that additional environmental factors and/or genes that are inherited independently of PMM2 have an impact on the clinical phenotype. Apart from unknown environmental stresses, polymorphisms/mutations in other genes, e.g., of the N-glycosylation pathway, may contribute to the severity of the disease. In fact, severely affected patients have an increased frequency of a polymorphism (F304S) in the ALG6 gene compared to moderately and mildly affected CDG-Ia patients (Westphal et al, 2002). ALG6 encodes the glycosyltransferase that adds the first glucose to the growing LLO in the rER (Figure 1) (Körner et al, 1998c). ALG6 is the gene mutated in CDG type Ic (Table 1), and the polymorphism compromises glycosylation when expressed in yeast under certain conditions (Westphal et al, 2002). Clearly, this is not an explanation of clinical variability in a highly complex disorder. Several biochemical steps with possible modifying influences occur between a mutated gene and its clinical ourcome. The importance of the genetic background has been demonstrated in animal models in which phenotypic differences among strains were obvious, e.g., the survival of the CDG type IIa (CDGIIa) mice. Wang et al, (2001) introduced a deletion into the mouse germline that reproduced the glycan biosynthetic defect responsible DANISH MEDICAL BULLETIN VOL. 51 NO. 4/NOVEMBER 2004 for human CDG-IIa (see p. 13). Mice lacking the Mgat2 gene had deficiency of N-acetylglucosaminyltransferase II activity and severe gastrointestinal, hematologic, and skeletal abnormalities. All Mgat2null mice died in early postnatal life. However, crossing the Mgat2 mutation into another genetic background (strain) resulted in Mgat2-null mice surviving for several months. 3.11. MANNOSE THERAPY Mannose is presumed to enter human fibroblasts by a mannose-specific transporter, supplying most of the mannose for N-glycosylation of proteins (Figure 1). Previously, glucose was thought to be the major source of mannose in this process (Pannerselvam and Freeze 1996b, Panneerselvam et al, 1997a). In vitro studies have demonstrated that addition of exogenous mannose (≥ 0.25 mM) to the culture medium of CDG-Ia fibroblasts can correct both the deficient mannose incorporation and the size of lipid- and protein-linked oligosaccharides (Panneerselvam and Freeze 1996a, Körner et al, 1998b). Furthermore, CDG-Ia patients had low blood mannose levels (average: 25 µM) compared to those of normal subjects (average: 55 µM) (Panneerselvam et al, 1997b). These observations encouraged us to perform a short-term study of dietary mannose therapy in CDG-Ia patients. Information about mannose absorption, clearance and tolerance in man was sparse. We demonstrated that both normal controls and CDG-Ia patients had at least a two- or threefold elevation in blood mannose at a dose of 0.1 g/kg body weight given at 3 h intervals (Alton et al, 1997). Mannose in serum was determined by an enzymatic assay described by Etchison and Freeze (1997): glucose was selectively removed from the sample by glucokinase and glucose-6-P-dehydrogenase, whereupon anionic reaction products and excess substrates were removed by anion-exchange chromatography. Mannose in the effluent was then assayed spectrophotometrically by the reduction of NADP to NADPH following coupled enzymatic reactions catalysed by hexokinase, phosphoglucoisomerase and glucose-6-P-dehydrogenase in the presence or absence of PMI (Figure 3). Our study established the feasibility of testing mannose as a therapy of CDG-Ia patients. Next step was to determine whether shortterm supplementation of mannose in CDG-Ia patients could improve their abnormal glycosylation of circulating glycoproteins (Kjaergaard et al, 1998b). Five CDG-Ia patients completed the protocol of enteral supplementation with mannose 0.1 g/kg body weight every 3 h for 9 days. A higher dose had caused gastrointestinal distress (Alton et al, 1997). Unfortunately, the mannose levels obtained in the patients were lower than the concentration shown to correct abnormal glycosylation in fibroblasts. However, since the in vitro correction was dose-dependent, a lower concentration therefore could be at least partial beneficial. Surprisingly, the mean concentrations of the glycoproteins tended to decline during continued mannose supplementation, and CDT increased as did the abnormal glycoforms, with the most pronounced changes observed after cessation of mannose supplementation. Physiological fluctuation of blood levels of glycoproteins or methodological variations could explain these observations. However, it seemed unlikely that the same trend in glycosylation of all glycoproteins analysed in five patients should be caused by random variation. Since the aim of our study was to test for correction of glycosylation in order to justify longterm mannose supplementation, we did not attempt to include a hard-to-obtain group of normal children or a group of CDG-Ia patients given another substance than mannose, which might have been able to illuminate the unexpected observations. Our results suggest an effect of mannose on the glycosylation of circulating glycoproteins in CDG-Ia, but it does not seem to be a beneficial one in the short-term. More knowledge of mannose metabolism in man is required to interpret the results correctly. Mannose supplementation has not been effective biochemically or clinically in other studies of CDG-Ia. Continuous intravenous DANISH MEDICAL BULLETIN VOL. 51 NO. 4/NOVEMBER 2004 mannose infusion for three weeks with serum levels up to 2 mM followed by oral mannose for six months did not improve the glycosylation of various glycoproteins (Mayatepek et al, 1997, Mayatepek and Kohlmüller 1998). Apparently, low PMM activity cannot be rescued by increasing the amount of its substrate in CDG-Ia patients in vivo. The biochemical basis for the in vitro correction of protein glycosylation is still not clear. So far the trials on mannose supplementation in CDG-Ia have been small and uncontrolled, and a long-term trial has not been reported. Providing PMM-deficient cells with mannose-1-P may be a way to increase the GDP- mannose pool, but mannose-1-P is not able to penetrate cell membranes due to its high polarity. Attempts to develop membrane-permeant derivatives of mannose-1-P are in process (Rutschow et al, 2002). However, therapy of CDG-Ia remains problematic. One reason is the prenatal onset of CDG-Ia demonstrated by the presence of dysmorphic features and neurological dysfunction at birth. On the other hand, the normal fetal growth and the failure to detect hypoglycosylation of transferrin in CDG-Ia prenatally (see p. 6) suggest that maternal compensation and/or a developmentally regulated alternate pathway may bypass PMM2 deficiency in utero. Presently, treatment offered patients with CDG-Ia remains supportive. 4. CONGENITAL DISORDER OF GLYCOSYLATION TYPE IB 4.1. PHOSPHOMANNOISOMERASE DEFICIENCY CDG-Ib is due to mutations in the MPI gene resulting in deficiency of PMI (Niehues et al, 1998, Jaeken et al, 1998, De Koning et al, 1998a). PMI is a cytosolic Zn2+-dependent enzyme catalysing the interconversion of fructose-6-P and mannose-6-P (Figure 1). As previously outlined, mannose-6-P is the precursor of GDP-mannose, which is required for glycosylation. PMI deficiency leads to a reduction of the GDP-mannose pool, and hence to incompletely glycosylated glycoproteins. PMI is widely distributed in human tissues, and low PMI activities, ranging from 3% to 18% of controls, have been demonstrated in leukocytes, cultured fibroblasts and liver tissue from CDG-Ib patients (Niehues et al, 1998, Jaeken et al, 1998, De Koning et al, 1998a, De Lonlay et al, 1999, Babovic-Vuksanovic et al, 1999, Westphal et al, 2001b). Human MPI has been cloned based on its homology to the gene in yeast. MPI, located on chromosome 15q22, consists of eight exons encoding a protein of 423 amino acids (Proudfoot et al, 1994, Schollen et al, 2000b). Fourteen different MPI mutations in 13 families have been reported, and 12 of these mutations are of the missense type, mostly affecting conserved amino acids (Schollen et al, 2000b, Westphal et al, 2001b, Vuillaumier-Barrot et al, 2002). 4.2. CLINICAL PRESENTATION Clinical manifestations of CDG-Ib include hepatic fibrosis, proteinlosing enteropathy, cyclic vomiting, thromboses, bleeding episodes and hypoglycemia, the latter sometimes due to hyperinsulinemia (Niehues et al, 1998, Jaeken et al, 1998, De Koning et al, 1998a, De Lonlay et al, 1999, Babovic-Vuksanovic et al, 1999, Westphal et al, 2001b). The absence of both neurological involvement and dysmorphic features sets the clinical phenotype quite apart from those of other types of CDG. Transferrin IEF pattern is identical to that of patients with other type I CDG (Figure 2). Albumin is low, liver transaminases are frequently raised, and the coagulopathy seen in CDG-Ib is indistinguishable from that of CDG-Ia with low levels of factor XI, ATIII and protein C. Two Danish sibs with CDG-Ib have been diagnosed by detection of “type I” transferrin IEF pattern, decreased PMI activity in cultured fibroblasts and identification of missense mutations in the MPI gene (Westphal et al, 2001b). The patients were originally reported in 1980 by Pedersen and Tygstrup because of their unusual clinical manifestations of recurrent venous thromboses, protein-losing enteropathy and hepatic fibrosis. No diagnosis was reached at the 357 time, and one sib died due to complications of the disease at 5 years of age. The good long-term outcome of the surviving sib, now age 35, demonstrates that serious manifestations in childhood may subside and allow for a normal adult life. Higher metabolic demands in infancy and childhood may overwhelm the cellular capacity for glycosylation, but with increasing age the patients become more stable as observed in other metabolic diseases, e.g., defects in branched chain amino acid catabolism. Symptoms, age at onset, severity of disease and outcome of CDG-Ib vary among reported patients. The Danish sibs presented later (2 1/2 y) than previously described CDG-Ib patients (0-12 months), and their clinical manifestations were clearly episodic and occurred in association with fever. Environmental stresses (e.g., infections) and genetic background may be important factors in severity and outcome of the disease. 4.3. MANNOSE THERAPY Mannose-6-P is normally derived from two sources: from fructose6-P via PMI or from free mannose entering the fibroblasts by a specific transporter, whereupon mannose is converted to mannose-6-P by a hexokinase (Figure 1). The latter pathway bypasses PMI deficiency, and oral mannose supplementation has been beneficial both clinically and biochemically to PMI-deficient patients. Daily oral mannose supplements have substantially reversed their proteinlosing enteropathy, coagulopathy, and hypoglycemia as well as improved their glycosylation of transferrin (Niehues et al, 1998, De Lonlay et al, 1999, Babovic-Vuksanovic et al, 1999). Mannose supplementation in our patient normalised the ATIII level and improved her glycosylation of transferrin, which, as expected, returned to the presupplementation state after cessation (Westphal et al, 2001b). Cellular import of mannose appeared to be much more efficient in our patient than in controls since her base-line serum mannose was lower and blood clearance of oral mannose threefold higher. These observations may be explained by upregulation of the mannose-specific transporter. Mannose clearance studies have not been reported in other CDG-Ib patients. Free mannose in man presumably comes from endogenous glycoprotein catabolism and dietary sources, but little is known about the availability of mannose in the human diet. Differences in dietary mannose intake may also contribute to the variable and episodic manifestations of CDG-Ib. Although the metabolic defect in CDG-Ib is localised in the same biosynthetic pathway and in the step immediately preceding the enzymatic step blocked in CDG-Ia, the clinical manifestations of the two disorders are quite dissimilar. Why is the clinical presentation of PMI deficiency mainly hepato-intestinal? Why is the central nervous system spared in PMI deficiency as opposed to other types of CDG? One possible explanation is that hexokinase I expressed in brain has a high affinity for mannose (Km 5 mM), thereby bypassing PMI and its deficiency. In contrast, glucokinase, the major hexokinase of hepatocytes, has a very low affinity for mannose (Km 33 mM), which may explain the liver involvement in CDG-Ib (Jaeken et al, 1998). Other possible explanations include the existence of different PMI isozymes and differences in expression of the mannose-specific transporter. 5. CONGENITAL DISORDERS OF GLYCOSYLATION – OTHER TYPES Several other defects in N-glycosylation have recently been elucidated (Table 1 and Figure 1). Comprehensive descriptions are not yet available as few patients of each type have been reported. In general, clinical manifestations are less specific than in CDG-Ia and Ib and include variable psychomotor retardation, hypotonia, epilepsy, variable dysmorphic features, and, frequently, coagulopathy and raised liver transaminases. It is important to emphasize that patients with other CDG types do not have the characteristics of CDG-Ia, i.e., inverted nipples, abnormal subcutaneous fat pads or cerebellar atrophy. Descriptions beyond the following summary of other CDG types are outside the scope of this thesis. 358 CDG-Ic is due to deficiency of the glycosyltransferase, which adds the first glucose to the growing LLO in the rER (Figure 1) (Körner et al, 1998c, Imbach et al, 1999). The clinical presentation includes moderate psychomotor retardation, hypotonia, epilepsy, feeding problems, esotropia and coagulopathy (Grünewald et al, 2000). The IEF of serum transferrin shows a “type I” pattern. CDG-Id is due to deficiency of the mannosyltransferase, which adds the sixth mannose to the growing LLO in the rER (Figure 1) (Körner et al, 1999). Clinical features are spastic paresis, severe psychomotor retardation, epilepsy, microcephaly and dysmorphism. ATIII may be reduced, and serum transferrin IEF shows a “type I” pattern. CDG-Ie is due to deficiency of the cytosolic subunit of dolichol-Pmannose synthetase, the enzyme required to generate dolichol-Pmannose, which is the donor of mannose for the growing LLO on the luminal side of the rER (Figure 1) (Imbach et al, 2000, Kim et al, 2000). Clinical manifestations include severe psychomotor retardation, hypotonia, epilepsy, microcephaly, cortical blindness, hepatosplenomegaly, dysmorphic features and coagulopathy. Liver transaminases are raised, and serum transferrin IEF pattern shows a “type I” pattern. CDG-If is due to defects in the MPDU1 gene, which is required for utilisation of the donor substrates, dolichol-P-mannose and dolichol-P-glucose, but not their synthesis (Figure 1) (Kranz et al, 2001, Schenk et al, 2001). The exact function of the gene product is not known. The clinical phenotype includes hypotonia, severe psychomotor retardation, seizures and variable skin affection. ATIII may be reduced, but liver transaminases are normal. The serum transferrin IEF shows a “type I” pattern. CDG-Ig. Recently, transferrin IEF screening (“type I” pattern, Figure 2) of a Danish patient with psychomotor retardation, hypotonia, failure to thrive, low immunoglobulins and low ATIII led to the identification of a novel type of CDG, type Ig (Grubenmann et al, 2002). CDG-Ig is due to mutations in the ALG12 gene encoding the mannosyltransferase, which adds the eighth mannose to the growing LLO in the rER (Figure 1) (Chantret et al, 2002a, Grubenmann et al, 2002). CDG-Ih, the most recently characterised type of CDG, is due to deficiency of the glycosyltransferase adding the second glucose onto the growing LLO in the rER (Figure 1) (Chantret et al, 2003). The only reported patient had hypoalbuminemia, protein-losing enteropathy, hepatomegaly and coagulopathy, but no CNS involvement – a clinical presentation close to that of CDG-Ib. Serum transferrin IEF showed a “type I” pattern. CDG-IIa is caused by deficiency of N-acetylglucosaminyltransferase II, the enzyme adding a glucosamine residue to the oligosaccharide of the newly synthesised glycoprotein in the Golgi apparatus (Figure 1) (Jaeken at al. 1994). Clinical features are severe psychomotor retardation, dysmorphism, hypotonia, epilepsy and stereotyped gestures. Raised liver transaminases and reduced ATIII activity are present. The serum transferrin IEF pattern is different from that seen in CDG-I with markedly decreased or absent tetrasialotransferrin and increased mono-, di-and trisialotransferrin (“type II” pattern). CDG-IIb is caused by deficiency of glucosidase I, the enzyme removing the terminal glucose from the oligosaccharide after its transfer to the polypeptide in the rER (Figure 1) (De Praeter et al, 2000). The only reported patient had severe hypotonia, dysmorphic features, hypoventilation, seizures, severe infections, hepatomegaly and coagulopathy. Serum transferrin IEF pattern was normal. The diagnosis was based on structural analysis of an abnormal tetrasaccharide found by thin layer chromatography of urine from the patient in question. CDG-IIc, usually called Leukocyte Adhesion Deficiency type II, is caused by reduced transport of GDP-fucose into the Golgi, resulting in hypofucosylation of glycoconjugates, including selectin ligands (Lübke et al, 2001). Features of CDG-IIc are psychomotor retardation DANISH MEDICAL BULLETIN VOL. 51 NO. 4/NOVEMBER 2004 and dysmorphic features combined with recurrent severe non-suppurative bacterial infections and high blood leukocyte counts. The immunodeficiency is due to defective selectin-dependent leukocyte trafficking to the sites of infection. IEF of serum transferrin is normal. Notably, oral fucose supplementation has proven beneficial biochemically and clinically in one CDG-IIc patient (Marquardt et al, 1999). CDG-IId is due to deficiency of galactosyltransferase, the enzyme adding galactose to the oligosaccharide of the newly synthesised glycoprotein in the Golgi apparatus (Figure 1) (Hansske et al, 2002). The only reported patient had myopathy, Dandy-Walker malformation, psychomotor retardation, raised liver transaminases and coagulopathy. IEF of serum transferrin showed a “type II” pattern. CDG-x. Several cases with not yet characterised glycosylation defects detected by transferrin IEF have been reported. Their diverse clinical presentations included psychomotor retardation, hypotonia, dysmorphic features and seizures as well as hydrops fetalis, severe neonatal diarrhoea and Budd-Chiari syndrome (Huemer at al. 2000, Mention et al, 2001, De Koning et al, 1998b). Within the last few years defects in proteoglycan synthesis and αdystroglycan synthesis have been identified, and the range of phenotypes related to glycosylation defects has been extended to previously known clinical diagnoses such as hereditary multiple exostoses, a progeroid variant of Ehlers-Danlos syndrome, muscleeye-brain disease and Walker-Warburg syndrome (Duncan et al, 2001, Okajima et al, 1999, Yoshida et al, 2001, Beltran-Valero De Bernabe et al, 2002). I-cell disease (or mucolipidosis II), which is due to deficiency of N-acetylglucosamine-1-phosphotransferase, required for synthesis of the mannose-6-P targeting signal, is usually assigned to the group of lysosomal storage diseases, but it is in fact a glycosylation defect. These disorders are not detected by IEF screening of serum transferrin. 6. DIAGNOSTIC STRATEGY CDG (except type Ib) should be considered in a patient with unexplained psychomotor retardation, isolated or associated with the clinical and biochemical features mentioned in Table 3 and p. 10. Inverted nipples, abnormal subcutaneous fat pads, cerebellar atrophy, failure to thrive, pericardial effusions and liver dysfunction suggest CDG-Ia. Early diagnosis of CDG-Ib is crucial as the disorder is potentially lethal, but curable with mannose supplementation. CDG-Ib should be considered in unexplained liver disease and in protein-losing enteropathy, isolated or in association with thrombosis, recurrent vomiting and hypoglycemia in otherwise normal children. Low ATIII activity and raised liver transaminases are common but not consistent findings in CDG. IEF of serum transferrin is a useful screening test, and it will identify the vast majority of patients with CDG (N-glycosylation defects), but not all (e.g., CDG-IIb and CDG-IIc). Protein variants of transferrin and causes not related to CDG should be excluded when an abnormal IEF pattern is detected (see p. 6). IEF analysis cannot distinguish between different types of CDG, and further analyses are needed to reach a specific diagnosis (Figure 5). Patients of Danish origin with a “type I” transferrin IEF pattern are analysed for the two most frequent PMM2 mutations, R141H and F119L. If not found, all exons of PMM2 are sequenced. If normal, the next step is to assay PMM activity in cultured fibroblasts or leukocytes, since PMM deficiency could be due to large deletions or mutations in the promoter not detected by routine sequencing. When CDG-Ia is excluded, an oligosaccharide profile of cultured fibroblasts is done to detect accumulation and/or depletion of oligosaccharide structures suggesting possible enzymatic defects and responsible genes. Profiling of oligosaccharides is based on HPLC analysis of LLO extracted from cultured fibroblasts following incubation with [3H] mannose (Powell et al, 1994). Next, the suspected enzyme is assayed and the responsible gene screened for mutations. The N-glycosylation pathway in the rER is highly conserved DANISH MEDICAL BULLETIN VOL. 51 NO. 4/NOVEMBER 2004 IEF of serum transferrin Abnormal pattern “Type I” Abnormal pattern “Type II” PMM2, MPI analysis PMM, PMI assay N-glycan structure analysis LLO analysis Enzymatic assay Mutation analysis Normal pattern CDG not excluded Enzymatic assay Mutation analysis Figure 5. Diagnostic strategy for CDG (N-glycosylation defects). among eukaryotes, and yeast mutants have been valuable in characterisation of novel types of CDG-I. For example, fibroblasts from the Danish patient who led to the identification of CDG-Ig accumulated lipid-linked GlcNA2Man7 oligosaccharides, as do alg12 mutant yeast cells protein (Grubenmann et al, 2002). The sequence information obtained from the yeast alg12 locus made it possible to identify the orthologous human ALG12 gene, and mutations at the locus were subsequently detected in patient DNA. In order to know if the mutations compromised the function of the ALG12-encoded mannosyltransferase, normal and mutant cDNA were expressed in alg12-deficient yeast. Indeed the mutant human ALG12 protein was unable to complement alg12 deficiency in yeast, as opposed to the normal ALG12. The same approach has been used to characterise CDG-Ic, CDG-Id and CDG-Ie. The absence of psychomotor retardation in CDG-Ib sets the clinical phenotype apart from those of other CDG types. The diagnosis is confirmed by demonstration of PMI deficiency in fibroblasts or leukocytes and identification of MPI mutations. “Type II” serum transferrin IEF pattern points to a N-glycosylation defect localised after the transfer of oligosaccharides to the protein, and a structural analysis of protein-bound glycans is performed to suggest an affected enzymatic step. Then, the suspected enzyme is assayed and the responsible gene screened for mutations. CDG-IIb may be picked up by analysis of urine oligosaccharides as described in the only case reported so far (De Praeter et al, 2000). CDG-IIc is detected by determination of the blood groups as the fucose-containing blood groups A, B, O and Lewis A are absent in these patients, followed by mutation screening of the responsible gene (Lübke et al, 2001). 7. CONCLUSIONS CDG-Ia was diagnosed in 32 Danish patients by detection of abnormal serum transferrin IEF pattern, profoundly decreased PMM activity in fibroblasts and mutations in the PMM2 gene. The incidence was 1:41 000 newborns, the highest reported so far. Two of seven missense mutations predominated, R141H and F119L, and 83% of the patients were compound heterozygous for these two mutations. Unexpectedly, we did not observe homozygosity for the R141H mutation. In vitro expression studies on the impact of mutations on PMM activity suggested the presence of two types of mutations. One type resulted in virtually null activity, and the other type resulted in PMM2 with decreased activity and/or altered properties. Each of our patients had at least one mutation that retained residual PMM2 activity when expressed in E. coli. The studies supported that homozygosity for R141H (or another “severe” mutation) impairs PMM2 activity to a degree incompatible with life, whilst F119L (or another 359 “mild” mutation) conveys a phenotype compatible with survival due to residual PMM2 activity. R141H and F119L were associated with separate specific haplotypes suggesting that each is derived from an ancestral mutation. R141H is by far the most frequent mutation in Caucasian patients with a carrier frequency of 1:60 in Denmark. F119L is a Scandinavian founder mutation with a much lower carrier frequency. Patients with the R141H/F119L genotype have an early uniform presentation with severe feeding problems, severe growth failure, hypotonia, inverted nipples, abnormal subcutaneous fat pads and developmental delay obvious before age 6 months but variable functional outcome. Patients with this genotype present with the “classical” phenotype and belong to the severe end of the broad clinical spectrum of CDG-Ia. Short-term mannose supplementation was not able to improve glycosylation in CDG-Ia patients. In contrast, mannose supplementation in an adult CDG-Ib patient proved beneficial biochemically. The reasons for the clinical variability seen among both CDG-Ia and CDG-Ib patients with identical molecular defect are unknown, but environmental factors and genetic background may have an impact on the clinical phenotype. The genotype does not seem to be a good predictor of clinical outcome in the individual patient. Prenatal diagnosis by mutation analysis of chorion villus sample has been performed in 18 at-risk pregnancies. 8. FUTURE PERSPECTIVES Efforts should be made to raise awareness among physicians of CDG as a large group of highly variable multi-system disorders, some of which are treatable. Increased awareness and broader serum transferrin IEF screening will increase the number of CDG patients with less typical presentations, and novel CDG types will be identified. The many biochemical steps/genes involved in the glycosylation pathways suggest a large number of potential defects leading to pathological conditions. More detailed clinical characterisations by compiling clinical data of novel CDG types is another way to promote their diagnosis and to improve their management. How abnormal glycosylation leads to the plethora of symptoms and signs in CDG-Ia and Ib is currently unexplained. Animal models may be helpful in unravelling the pathophysiology and, ultimately, in exploring possible treatments. The availability of the sequence and structure of each of the mouse Pmm1, Pmm2 and Mpi genes is the initial step in development of mouse models with deficiency of one of the enzymes (Heykants et al, 2001, Davis et al, 2002). Based on the lethality of R141H homozygotes in humans, it is unlikely that mice with complete ablation of PMM2 will produce viable offspring. Introducing appropriate mutations in the gene conveying residual PMM2 activity is more likely to have a viable outcome. Patients and the pharmaceutical industries will continue to benefit from increased knowledge of the impact of glycans on the stability, targeting, action and clearance of bioactive molecules. One example is Gaucher disease due to deficiency of the lysosomal enzyme glucocerebrosidase resulting in accumulation of glucocerebroside. During normal biosynthesis of lysosomal enzymes Nglycans become modified with man-6-P residues to target them to the man-6-P receptor of lysosomes. Consequently, recombinant glucocerebrosidase produced for enzyme replacement therapy in Gaucher disease is modified to contain the targeting signal. The recombinant enzyme is thereby targeted to the correct site of action. Collaboration with other centres dedicated to CDG is necessary for several reasons. The diagnosis of and basic research in CDG require diverse technologies, and gathering of patients is needed to increase the number required for clinical studies. EUROGLYCAN, a European collaboration financed by the EU, is an excellent forum for these purposes. Research in CDG will most likely provide more knowledge of the roles of N-glycans in physiology. 360 ABBREVIATIONS ATIII Antithrombin III CDG Congenital Disorders of Glycosylation CDG-Ia Congenital Disorder of Glycosylation type Ia CDG-Ib Congenital Disorder of Glycosylation type Ib GDP Guanosine diphosphate CDT Carbohydrate-Deficient Transferrin IEF Isoelectric focusing LLO Lipid-linked oligosaccharide MPI Gene encoding PMI NADPH Nicotinamide adenine dinucleotide phosphate PMI Phosphomannoisomerase PMM Phosphomannomutase (both isozymes) PMM1 Phosphomannomutase 1 (isozyme 1) PMM1 Gene encoding phosphomannomutase 1 PMM2 Phosphomannomutase 2 (isozyme 2) PMM2 Gene encoding phosphomannomutase 2 rER rough endoplasmic reticulum. REFERENCES 1. Alton G, Kjaergaard S, Etchison JR, Skovby F, Freeze H. Oral Ingestion of Mannose Elevates Blood Mannose Levels: A First Step Toward a Potential Therapy for Carbohydrate-Deficient Glycoprotein Syndrome Type I. Biochem Molecul Med 60: 127-33. 1997. 2. Aravind L, Galperin MY, Koonin EV. The catalytic domain of the P-type ATPase has the haloacid dehalogenase fold. Trends Biochem Sci 23: 1279. 1998. 3. Babovic-Vuksanovic D, Patterson MC, Schwenk WF, O’Brien JF, Vockley J, Freeze HH, Mehta DP, Michels VV. Severe hypoglycemia as a presenting symptom of carbohydrate-deficient glycoprotein syndrome. J Pediatr 135: 775-81. 1999. 4. Barone R, Carchon H, Jansen E, Pavone L, Fiumara A, Bosshard NU, Gitzelmann R, Jaeken J. Lysosomal enzyme activities in serum and leukocytes from patients with carbohydrate-deficient glycoprotein syndrome type IA (phosphomannomutase deficiency). J Inherit Metab Dis 21: 167-72. 1998. 5. Beltran-Valero De Bernabe D, Currier S, Steinbrecher A, Celli J, Van Beusekom E, Van Der Zwaag B, Kayserili H, Merlini L, Chitayat D, Dobyns WB, Cormand B, Lehesjoki AE, Cruces J, Voit T, Walsh CA, Van Bokhoven H, Brunner HG. Mutations in the O-Mannosyltransferase Gene POMT1 Give Rise to the Severe Neuronal Migration Disorder Walker-Warburg Syndrome. Am J Hum Genet. 71: 1033-43. 2002. 6. Bjursell C, Stibler H, Wahlström J, Kristiansson B, Skovby F, Strömme P, Blennow G, Martinsson T. Fine mapping of the gene for carbohydratedeficient glycoprotein syndrome, type I (CDG1): linkage disequilibrium and founder effect in Scandinavian families. Genomics 39: 24753. 1997. 7. Bjursell C, Wahlström J, Berg K, et al: Detailed mapping of the phosphomannomutase 2 (PMM2) gene and mutation detection enable improved analysis for Scandinavian CDG type I families. Eur J Hum Genet 6: 603-11. 1998. 8. Bjursell C, Erlandson A, Nordling M, Nilsson S, Wahlström J, Stibler H, Kristiansson B, Martinsson T. PMM2 mutation spectrum, including 10 novel mutations, in a large CDG type 1A family material with a focus on Scandinavian families. Hum Mutat 16: 395-400. 2000. 9. Böhles H, Sewell AA, Gebhardt B, Reinecke-Lüthge A, Klöppel G, Marquardt T. Hyperinsulinaemic hypoglycaemia – leading symptom in a patient with congenital disorder of glycosylation Ia (phosphomannomutase deficiency). J Inherit Metab Dis 24: 858-62. 2001. 10. Chantret I, Dupre T, Delenda C, Bucher S, Dancourt J, Barnier A, Charollais A, Heron D, Bader-Meunier B, Danos O, Seta N, Durand G, Oriol R, Codogno P, Moore SE. Congenital disorders of glycosylation type Ig is defined by a deficiency in dolichyl-P-mannose: Man7Glc NAc2-PP-dolichyl mannosyltransferase. J Biol Chem 277: 25815-22. 2002a. 11. Chantret I, Dancourt J, Dupre T, Delenda C, Bucher S, VuillaumierBarrot S, Ogier De Baulny H, Peletan C, Danos O, Seta N, Durand G, Oriol R, Codogno P, Moore SE. A deficiency in dolichyl-P-glucose: Glc1Man9GlcNAc2-PP-dolichyl alpha 3-glucosyltransferase defines a new subtype of congenital disorders of glycosylation (CDG). J Biol Chem 278: 9962-9971. 2003. 12. Charlwood J, Clayton P, Keir G, Mian N, Young E, Winchester B. Prenatal diagnosis of the carbohydrate-deficient glycoprotein syndrome type 1A (CDG1A) by a combination of enzymology and genetic linkage analysis after amniocentesis or chorionic villus sampling. Prenat Diagn 18: 693-9. 1998. 13. Clayton P, Winchester B, Di Tomaso E, Young E, Keir G, Rodeck C. Carbohydrate-deficient glycoprotein syndrome: normal glycosylation in the fetus. Lancet 341: 956. 1993. DANISH MEDICAL BULLETIN VOL. 51 NO. 4/NOVEMBER 2004 14. Collet JF, Stroobant V, Pirard M, Delpierre G, Van Schaftingen E. A new class of phosphotransferases phosphorylated on an aspartate residue in an amino-terminal DXDX(T/V) motif. J Biol Chem 273: 14107-12. 1998. 15. Davis JA, Wu XH, Wang L, DeRossi C, Westphal V, Wu R, Alton G, Srikrishna G, Freeze HH. Molecular cloning, gene organization, and expression of mouse Mpi encoding phosphomannose isomerase. Glycobiology 12: 435-42. 2002. 16. De Koning TJ, Dorland L, Van Diggelen OP, Boonman AM, De Jong GJ, Van Noort WL, De Schryver J, Duran M, Van den Berg IE, Gerwig GJ, Berger R, Poll-The BT. A novel disorder of N-glycosylation due to phosphomannose isomerase deficiency. Biochem Biophys Res Commun 245: 38-42. 1998a. 17. De Koning TJ, Toet M, Dorland L, De Vries LS, Van den Berg IE, Duran M, Poll-The BT. Recurrent nonimmune hydrops fetalis associated with carbohydrate-deficient glycoprotein syndrome. J Inherit Metab Dis 21: 681-2. 1998b. 18. De Lonlay P, Cuer M, Vuillaumier-Barrot S, Beaune G, Castelnau P, Kretz M, Durand G, Saudubray JM, Seta N. Hyperinsulinemic hypoglycemia as a presenting sign in phosphomannose isomerase deficiency: A new manifestation of carbohydrate-deficient glycoprotein syndrome treatable with mannose. J Pediatr 135: 379-83. 1999. 19. De Lonlay P, Seta N, Barrot S, Chabrol B, Drouin V, Gabriel BM, et al, A broad spectrum of clinical presentations in congenital disorders of glycosylation I: A series of 26 cases. J Med Genet 38: 14-9. 2001. 20. De Praeter CM, Gerwig GJ, Bause E, Nuytinck LK, Vliegenthart JF, Breuer W, Kamerling JP, Espeel MF, Martin JJ, De Paepe AM, Chan NW, Dacremont GA, Van Coster RN. A novel disorder caused by defective biosynthesis of N-linked oligosaccharides due to glucosidase I deficiency. Am J Hum Genet 66: 1744-56. 2000. 21. De Zegher F, Jaeken J. Endocrinology of the carbohydrate-deficient glycoprotein syndrome type 1 from birth through adolescense. Pediatr Res 37: 396-401. 1995. 22. Dipple KM, McCabe ER. Phenotypes of patients with “simple” Mendelian disorders are complex traits: thresholds, modifiers, and systems dynamics. Am J Hum Genet 66: 1729-35. 2000. 23. Drouin-Garraud V, Belgrand M, Grunewald S, Seta N, Dacher JN, Henocq A, Matthijs G, Cormier-Daire V, Frebourg T, Saugier-Veber P. Neurological presentation of a congenital disorder of glycosylation CDG-Ia: implications for diagnosis and genetic counseling. Am J Med Genet 101: 46-9. 2001. 24. Duncan G, McCormick C, Tufaro F. The link between heparan sulfate and hereditary bone disease: finding a function for the EXT family of putative tumor suppressor proteins. J Clin Invest 108: 511-6. 2001. 25. Dupre T, Cuer M, Barrot S, Barnier A, Cormier-Daire V, Munnich A, Durand G, Seta N. Congenital disorder of glycosylation Ia with deficient phosphomannomutase activity but normal plasma glycoprotein pattern. Clin Chem 47: 132-4. 2001. 26. Enns GM, Steiner RD, Buist N, Cowan C, Leppig KA, McCracken MF, Westphal V, Freeze HH, O'Brien JF, Jaeken J, Matthijs G, Behera S, Hudgins L. Clinical and molecular features of congenital disorder of glycosylation in patients with type 1 sialotransferrin pattern and diverse ethnic origins.J Pediatr 141: 695-700. 2002. 27. Erlandson A, Bjursell C, Stibler H, Kristiansson B, Wahlström J, Martinsson T. Scandinavian CDG-Ia patients: genotype/phenotype correlation and geographic origin of founder mutations. Hum Genet 108: 35967. 2001. 28. Etchison JR, Freeze HH. Enzymatic assay of D-mannose in serum. Clin Chem 43: 533-8. 1997. 29. Fletcher JM, Matthijs G, Jaeken J, Van Schaftingen, Nelson PV. Carbohydrate-deficient glycoprotein syndrome: beyond the screen. J Inher Metab Dis 23: 396-8. 2000. 30. Grubenmann CE, Frank CG, Kjaergaard S, Berger EG, Aebi M, Hennet T. ALG12 mannosyltransferase defect in congenital disorder of glycosylation type lg. Hum Mol Genet 11: 2331-9. 2002. 31. Grünewald S, Imbach T, Huijben K, Rubio-Gozalbo ME, Verrips A, de Klerk JB, Stroink H, de Rijk-van Andel JF, Van Hove JL, Wendel U, Matthijs G, Hennet T, Jaeken J, Wevers RA. Clinical and biochemical characteristics of congenital disorder of glycosylation type Ic, the first recognized endoplasmic reticulum defect in N-glycan synthesis. Ann Neurol 47: 776-81. 2000. 32. Grünewald S, Schollen E, Van Schaftingen E, Jaeken J, Matthijs G. High residual activity of PMM2 in patients’ fibroblasts: possible pitfall in the diagnosis of CDG-Ia (phosphomannomutase deficiency). Am J Hum Genet 68: 347-54. 2001. 33. Hansske B, Thiel C, Lübke T, Hasilik M, Honing S, Peters V, Heidemann PH, Hoffmann GF, Berger EG, Von Figura K, Korner C. Deficiency of UDP-galactose: N-acetylglucosamine beta-1,4-galactosyltransferase I causes the congenital disorder of glycosylation type IId. J Clin Invest 109: 725-33. 2002. 34. Heykants L, Schollen E, Grünewald S, Matthijs G. Identification and localization of two mouse phosphomannomutase genes, Pmm1 and Pmm2. Gene 270: 53-9. 2001. DANISH MEDICAL BULLETIN VOL. 51 NO. 4/NOVEMBER 2004 35. Huemer M, Huber WD, Schima W, Moeslinger D, Holzbach U, Wevers R, Wank H, Stoeckler-Ipsiroglu S. Budd-Chiari syndrome associated with coagulation abnormalities in a child with carbohydrate deficient glycoprotein syndrome type Ix. J Pediatr 36: 691-5. 2000. 36. Imbach T, Burda P, Kuhnert P, Wevers RA, Aebi M, Berger EG, Hennet T. A mutation in the human ortholog of the Saccharomyces cerevisiae ALG6 gene causes carbohydrate-deficient glycoprotein syndrome type Ic. Proc Natl Acad Sci U S A 96: 6982-7. 1999. 37. Imbach T, Schenk B, Schollen E, Burda P, Stutz A, Grünewald S, Bailie NM, King MD, Jaeken J, Matthijs G, Berger EG, Aebi M, Hennet T. Deficiency of dolichol-phosphate-mannose synthase-1 causes congenital disorder of glycosylation type Ie. J Clin Invest 105: 233-9. 2000. 38. Imtiaz F, Worthington V, Champion M, et al, Genotypes and phenotypes of patients in the UK with carbohydrate-deficient glycoprotein syndrome type 1. J Inherit Metab Dis 23: 162-74. 2000. 39. Jaeken J, Vanderschueren-Lodeweyckx M, Casaer P, et al, Familial psychomotor retardation with markedly fluctuating serum prolactin, FSH and GH levels, partial TBG-deficiency, increased serum arylsulphatase A and increased CSF protein: a new syndrome? Pediatr Res 14: 179. 1980. 40. Jaeken J, Stibler H, Hagberg B (eds). The Carbohydrate-Deficient Glycoprotein Syndrome: A new inherited multisystemic disease with severe nervous system involvement. Acta Paediatr Scand 375 Suppl: 171. 1991. 41. Jaeken J, Schachter H, Carchon H, De Cock P, Coddeville B, Spik G. Carbohydrate deficient glycoprotein syndrome type II: a deficiency in Golgi localised N-acetyl-glucosaminyltransferase II. Arch Dis Child 71: 123-7. 1994. 42. Jaeken J, Pirard M, Adamowicz M, Pronicka E, Van Schaftingen E. Inhibition of phosphomannose isomerase by fructose 1-phosphate: an explanation for defective N-glycosylation in hereditary fructose intolerance. Pediatr Res 40: 764-6. 1996. 43. Jaeken J, Matthijs G, Saudubray JM, Dionisi-Vici C, Bertini E, De Lonlay P, Henri H, Carchon H, Schollen E, Van Schaftingen E. Phosphomannose isomerase deficiency: a carbohydrate-deficient glycoprotein syndrome with hepatic-intestinal presentation. Am J Hum Genet 62: 1535-9. 1998. 44. Jaeken J, Matthijs G, Carchon H and Van Schaftingen E. Defects of Nglycan synthesis. In Scriver CR, Beaudet AL, Sly WS and Valle D (eds.), The metabolic and molecular bases of inherited disease 8th ed., McGraw-Hill, New York 2001. 45. Jensen H, Kjaergaard S, Klie F, Møller HU. Ophthalmic manifestations of congenital disorder of glycosylation type Ia. Ophthalmic Genet 24: 81-88. 2003. 46. Jensen PR, Hansen FJ, Skovby F. Cerebellar hypoplasia in children with the carbohydrate-deficient glycoprotein syndrome. Neuroradiology 37: 328-30. 1995. 47. Kepes F, Schekman R. The yeast SEC53 gene encodes phosphomannomutase. J Biol Chem 263: 9155-61. 1988. 48. Kim S, Westphal V, Srikrishna G, Mehta DP, Peterson S, Filiano J, Karnes PS, Patterson MC, Freeze HH. Dolichol phosphate mannose synthase (DPM1) mutations define congenital disorder of glycosylation Ie (CDG-Ie). J Clin Invest 105: 191-8. 2000. 49. Kjaergaard S, Skovby F, Schwartz M. Absence of homozygosity for predominant mutations in PMM2 in Danish patients with carbohydratedeficient glycoprotein syndrome type 1. Eur J Hum Gen 6: 331-6. 1998a. 50. Kjaergaard S, Kristiansson B, Stibler H, Freeze H, Schwartz M, Martinsson T, Skovby F. Failure of Short-Term Mannose Therapy of Patients with Carbohydrate-Deficient Glycoprotein Syndrome Type 1A. Acta Paediatr 87: 884-8. 1998b. 51. Kjaergaard S, Skovby F, Schwartz M. Carbohydrate-deficient glycoprotein syndrome type 1A: Expression and characterisation of wild type and mutant PMM2 in E. coli. Eur J Hum Genet 7: 884-8. 1999. 52. Kjaergaard S, Schwartz M, Skovby F. Congenital disorder of glycosylation type Ia (CDG-Ia): phenotypic spectrum of the R141H/F119L genotype. Arch Dis Child 85: 236-9. 2001. 53. Kjaergaard S, Müller J, Skovby F. Prepubertal growth in congenital disorder of glycosylation type Ia. Arch Dis Child 87: 324-7. 2002. 54. Kondo I, Mizugishi K, Yoneda Y, Hashimoto T, Kuwajima K, Yuasa I, Shigemoto K, Kuroda Y. Missense mutations in phosphomannomutase 2 gene in two Japanese families with carbohydrate-deficient glycoprotein syndrome type 1. Clin Genet 55: 50-4. 1999. 55. Körner C, Lehle L, Von Figura K. Abnormal synthesis of mannose-1phosphate derived carbohydrates in carbohydrate-deficient glycoprotein syndrome type I fibroblasts with phosphomannomutase deficiency. Glycobiology 8: 165-71. 1998a. 56. Körner C, Lehle L, Von Figura K. Carbohydrate-deficient glycoprotein syndrome type 1: correction of the glycosylation defect by deprivation of glucose or supplementation of mannose. Glycoconj J 15: 499-505. 1998b. 57. Körner C, Knauer R, Holzbach U, Hanefeld F, Lehle L, Von Figura K. Carbohydrate-deficient glycoprotein syndrome type V: deficiency of do- 361 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 362 lichyl-P-Glc:Man9GlcNAc2-PP-dolichyl glucosyltransferase. Proc Natl Acad Sci U S A 95: 13200-5. 1998c. Körner C, Knauer R, Stephani U, Marquardt T, Lehle L, Von Figura K. Carbohydrate deficient glycoprotein syndrome type IV: deficiency of dolichyl-P-Man:Man(5)GlcNAc(2)-PP-dolichyl mannosyltransferase. EMBO J 18: 6816-22. 1999. Kranz C, Denecke J, Lehrman MA, Ray S, Kienz P, Kreissel G, Sagi D, Peter-Katalinic J, Freeze HH, Schmid T, Jackowski-Dohrmann S, Harms E, Marquardt T. A mutation in the human MPDU1 gene causes congenital disorder of glycosylation type If (CDG-If). J Clin Invest 108: 1613-9. 2001. Krasnewich DM, Holt GD, Brantly M, Skovby F, Redwine J, Gahl WA. Abnormal synthesis of dolichol-linked oligosaccharides in carbohydrate-deficient glycoprotein syndrome. Glycobiology 5: 503-10. 1995. Kristiansson B, Andersson M, Tonnby B, Hagberg B. Disialotransferrin developmental deficiency syndrome. Arch Dis Child 64: 71-6. 1989. Kristiansson B, Stibler H, Wide L. Gonadal function and glycoprotein hormones in the carbohydrate-deficient glycoprotein (CDG) syndrome. Acta Paediatr 84: 655-60. 1995. Kristiansson B, Borulf S, Conradi N, Erlanson-Albertsson C, Ryd W, Stibler H. Intestinal, Pancreatic and Hepatic Involvement in Carbohydrate-Deficient Glycoprotein Syndrome Type I. J Pediatr Gastroenterol Nutr 27: 23-9. 1998. Lübke T, Marquardt T, Etzioni A, Hartmann E, Von Figura K, Körner C. Complementation cloning identifies CDG-IIc, a new type of congenital disorders of glycosylation, as a GDP-fucose transporter deficiency. Nat Genet 28: 73-6. 2001 Macchia PE, Harrison HH, Scherberg NH, Sunthornthepvarakul T, Jaeken J, Refetoff S. Thyroid function tests and characterization of thyroxine-binding globulin in the carbohydrate-deficient glycoprotein syndrome type I. J Clin Endocrinol Metab 80: 3744-9. 1995. Marquardt T, Luhn K, Srikrishna G, Freeze HH, Harms E, Vestweber D. Correction of leukocyte adhesion deficiency type II with oral fucose. Blood 94: 3976-85. 1999. Martinsson T, Bjursell C, Stibler H, Kristiansson B, Skovby F, Jaeken J, Blennow G, Strömme P, Hanefeld F, Wahlström J. Linkage of a locus for carbohydrate-deficient glycoprotein syndrome type I (CDG1) to chromosome 16p, and linkage disequilibrium to microsatellite marker D16S406. Hum Mol Genet 3: 2037-42. 1994. Matthijs G, Schollen E, Pirard M, Budarf ML, Van Schaftingen E, Cassiman JJ. PMM (PMM1), the human homologue of SEC53 or yeast phosphomannomutase, is localized on chromosome 22q13. Genomics 40: 41-7. 1997a. Matthijs G, Schollen E, Pardon E, Veiga-Da-Cunha M, Jaeken J, Cassiman JJ, Van Schaftingen E. Mutations in PMM2, a phosphomannomutase gene on chromosome 16p13, in carbohydrate-deficient glycoprotein type I syndrome (Jaeken syndrome). Nat Genet 16: 88-92. 1997b. Matthijs G, Schollen E, Van Schaftingen E, Cassiman JJ, Jaeken J. Lack of homozygotes for the most frequent disease allele in carbohydratedeficient glycoprotein syndrome type 1A. Am J Hum Genet 62: 542-50. 1998a. Matthijs G, Schollen E, Cassiman JJ, Cormier-Daire V, Jaeken J, Van Schaftingen E. Prenatal diagnosis in CDG1 families: beware of heterogeneity. Eur J Hum Genet 6: 99-104. 1998b. Matthijs G, Schollen E, Bjursell C, Erlandson A, Freeze H, Imtiaz F, Kjaergaard S, Martinsson T, Schwartz M, Seta N, Vuillaumier-Barrot S, Westphal V, Winchester B. Mutations in PMM2 that cause congenital disorders of glycosylation, type Ia (CDG-Ia). Hum Mutat 16: 386-94. 2000. Mayatepek E, Schröder M, Kohlmüller D, Bieger WP, Nützenadel W. Continuous mannose infusion in carbohydrate-deficient glycoprotein syndrome type I. Acta Paediatr 86: 1138-40. 1997. Mayatepek E, Kohlmüller D. Mannose supplementation in carbohydrate-deficient glycoprotein syndrome type I and phosphomannomutase deficiency. Eur J Pediatr 157: 605-6. 1998. Mention K, Michaud L, Dobbelaere D, Guimber D, Gottrand F, Turck D. Neonatal severe intractable diarrhoea as the presenting manifestation of an unclassified congenital disorder of glycosylation (CDG-x). Arch Dis Child Fetal Neonatal Ed. 85: F217-9. 2001. Niehues R, Hasilik M, Alton G, Körner C, Schiebe-Sukumar M, Koch HG, Zimmer KP, Wu R, Harms E, Reiter K, Von Figura K, Freeze HH, Harms HK, Marquardt T. Carbohydrate-deficient glycoprotein syndrome type Ib. Phosphomannose isomerase deficiency and mannose therapy. J Clin Invest 101: 1414-20. 1998. Okajima T, Fukumoto S, Furukawa K, Urano T. Molecular basis for the progeroid variant of Ehlers-Danlos syndrome. Identification and characterization of two mutations in galactosyltransferase I gene. J Biol Chem 274: 28841-4. 1999. Panneerselvam K, Freeze HH. Mannose corrects altered N-glycosylation in carbohydrate-deficient glycoprotein syndrome fibroblasts. J Clin Invest 97: 1478-87. 1996a. 79. Panneerselvam K, Freeze HH. Mannose enters mammalian cells using a specific transporter that is insensitive to glucose. J Biol Chem 271: 941721. 1996b. 80. Panneerselvam K, Etchison JR, Freeze HH. Human fibroblasts prefer mannose over glucose as a source of mannose for N-glycosylation. Evidence for the functional importance of transported mannose. J Biol Chem 272: 23123-9. 1997a. 81. Panneerselvam K, Etchison JR, Skovby F, Freeze HH. Abnormal metabolism of mannose in families with carbohydrate-deficient glycoprotein syndrome type 1. Biochem Mol Med 61: 161-7. 1997b. 82. Participants in the First International Workshop on CDGS. Carbohydrate-deficient glycoprotein syndromes become congenital disorders of glycosylation: an updated nomenclature for CDG. Glycoconj J 16: 669-71. 1999. 83. Pedersen PS, Tygstrup I. Congenital hepatic fibrosis combined with protein-losing enteropathy and recurrent thrombosis. Acta Paediatr Scand 69: 571-4. 1980. 84. Petersen MB, Brostrøm K, Stibler H, Skovby F. Early manifestations of the carbohydrate-deficient glycoprotein syndrome. J Pediatr 122: 66-70. 1993. 85. Pirard M, Collet JF, Matthijs G, Van Schaftingen E. Comparison of PMM1 with the phosphomannomutases expressed in rat liver and in human cells. FEBS Lett 411: 251-4. 1997. 86. Pirard M, Achouri Y, Collet JF, Schollen E, Matthijs G, Van Schaftingen E. Kinetic properties and tissular distribution of mammalian phosphomannomutase isozymes. Biochem J 339: 201-7. 1999a. 87. Pirard M, Matthijs G, Heykants L, Schollen E, Grünewald S, Jaeken J, Van Schaftingen E. Effect of mutations found in carbohydrate-deficient glycoprotein syndrome type IA on the activity of phosphomannomutase 2. FEBS Lett 452: 319-22. 1999b. 88. Powell LD, Panneerselvam K, Vij R, Diaz S, Manzi A, Buist N, Freeze H, Varki A. Carbohydrate-deficient glycoprotein syndrome: not an Nlinked oligosaccharide processing defect, but an abnormality in lipidlinked oligosaccharide biosynthesis? J Clin Invest 94: 1901-9. 1994. 89. Proudfoot AE, Turcatti G, Wells TN, Payton MA, Smith DJ. Purification, cDNA cloning and heterologous expression of human phosphomannose isomerase. Eur J Biochem 219: 415-23. 1994. 90. Rush JS, Panneerselvam K, Waechter CJ, Freeze HH. Mannose supplementation corrects GDP-mannose deficiency in cultured fibroblasts from some patients with Congenital Disorders of Glycosylation (CDG). Glycobiology 10: 829-35. 2000. 91. Rutschow S, Thiem J, Kranz C, Marquardt T. Membrane-permeant derivatives of mannose-1-phosphate. Bioorg Med Chem. 10: 4043-9. 2002. 92. Schenk B, Imbach T, Frank CG, Grubenmann CE, Raymond GV, Hurvitz H, Raas-Rotschild A, Luder AS, Jaeken J, Berger EG, Matthijs G, Hennet T, Aebi M. MPDU1 mutations underlie a novel human congenital disorder of glycosylation, designated type If. J Clin Invest 108: 168795. 2001. 93. Schollen E, Pardon E, Heykants L, Renard J, Doggett NA, Callen DF, Cassiman JJ, Matthijs G. Comparative analysis of the phosphomannomutase genes PMM1, PMM2 and PMM2ψ: the sequence variation in the processed pseudogene is a reflection of the mutations found in the functional gene. Hum Mol Genet 7: 157-64. 1998. 94. Schollen E, Kjaergaard S, Legius E, Schwartz M and Matthijs G. Lack of Hardy-Weinberg equilibrium for the most prevalent PMM2 mutation in CDG-Ia (congenital disorders of glycosylation type Ia). Eur J Hum Genet 8: 367-71. 2000a. 95. Schollen E, Dorland L, De Koning TJ, Van Diggelen OP, Huijmans JG, Marquardt T, Babovic-Vuksanovic D, Patterson M, Imtiaz F, Winchester B, Adamowicz M, Pronicka E, Freeze H, Matthijs G. Genomic organization of the human phosphomannose isomerase (MPI) gene and mutation analysis in patients with congenital disorders of glycosylation type Ib (CDG-Ib). Hum Mut 16: 247-52. 2000b. 96. Schwartz M, Brandt NJ, Skovby F. Screening for carriers of cystic fibrosis among pregnant women: a pilot study. Eur J Hum Genet 1: 239-44. 1993. 97. Stibler H, Borg S, Beckman G. Transferrin phenotype and level of carbohydrate-deficient transferrin in healthy individuals. Alcohol Clin Exp Res 12: 450-3. 1988. 98. Stibler H. Carbohydrate-deficient transferrin in serum: a new marker of potentially harmful alcohol consumption reviewed. Clin Chem 37: 2029-37. 1991a. 99. Stibler H, Jaeken J, Kristiansson B. Biochemical characteristics and diagnosis of the carbohydrate-deficient glycoprotein syndrome. Acta Paediatr Scand 375 Suppl: 21-31. 1991b. 100. Stibler H, Cederberg B. Diagnosis of the carbohydrate-deficient glycoprotein syndrome by analysis of transferrin in filter paper blood spots. Acta Paediatr 82: 55-9. 1993. 101. Stibler H, Skovby F. Failure to diagnose carbohydrate-deficient glycoprotein syndrome prenatally. Pediatr Neurol 11: 71. 1994a. 102. Stibler H, Blennow G, Kristiansson B, Lindehammer H, Hagberg B. Carbohydrate-deficient glycoprotein syndrome: clinical expression in DANISH MEDICAL BULLETIN VOL. 51 NO. 4/NOVEMBER 2004 adults with a new metabolic disease. J Neurol Neurosurg Psychiatry 57: 552-6. 1994b. 103. Stibler H, Holzbach U, Tengborn L, Kristiansson B. Complex functional and structural coagulation abnormalities in the carbohydrate-deficient glycoprotein syndrome type I. Blood Coagul Fibrinolysis 7: 118-26. 1996. 104. Stibler H, Von Döbeln U, Kristiansson B, Guthenberg C. Carbohydratedeficient transferrin in galactosaemia. Acta Paediatr 86: 1377-8. 1997. 105. Stibler H, Holzbach U, Kristiansson B. Isoforms and levels of transferrin, antithrombin, alpha(1)-antitrypsin and thyroxine-binding globulin in 48 patients with carbohydrate-deficient glycoprotein syndrome type I. Scand J Clin Lab Invest 58: 55-61. 1998. 106. Tan J, Dunn J, Jaeken J, Schachter H. Mutations in the MGAT2 gene controlling complex N-glycan synthesis cause carbohydrate-deficient glycoprotein syndrome type II, an autosomal recessive disease with defective brain development. Am J Hum Genet 59: 810-7. 1996. 107. Van der Knaap MS, Wevers RA, Monnens L, Jakobs C, Jaeken J, Van Wijk JA. Congenital nephrotic syndrome: a novel phenotype of type I carbohydrate-deficient glycoprotein syndrome. J Inherit Metab Dis 19: 787-91. 1996. 108. Van Geet C, Jaeken J. A unique pattern of coagulation abnormalities in carbohydrate-deficient glycoprotein syndrome. Pediatr Res 33: 540-1. 1993. 109. Van Ommen CH, Peters M, Barth PG, Vreken P, Wanders RJA, Jaeken J. Carbohydrate-deficient glycoprotein syndrome type 1a: a variant phenotype with borderline cognitive dysfunction, cerebellar hypoplasia, and coagulation abnormalities. J Pediatr 136: 400-3. 2000. 110. Van Schaftingen E, Jaeken J. Phosphomannomutase deficiency is a cause of carbohydrate-deficient glycoprotein syndrome type I. FEBS Lett 377: 318-20. 1995. 111. Varki A, Cummings R, Esko J, Freeze H, Hart G, Marth J (eds). Essentials of Glycobiology. Cold Spring Harbor Laboratory Press, New York 1999. 112. Vuillaumier-Barrot S, Barnier A, Cuer M, Durand G, Grandchamp B, Seta N. Characterization of the 415G>A (E139K) PMM2 mutation in carbohydrate-deficient glycoprotein syndrome type Ia disrupting a splicing enhancer resulting in exon 5 skipping. Hum Mutat 14: 543-4. 1999. 113. Vuillaumier-Barrot S, Hetet G, Barnier A, Dupre T, Cuer M, De Lonlay P, Cormier-Daire V, Durand G, Grandchamp B, Seta N. Identification of four novel PMM2 mutations in congenital disorders of glycosylation (CDG) Ia French patients. J Med Genet 37: 579-80. 2000. 114. Vuillaumier-Barrot S, Le Bizec C, de Lonlay P, Barnier A, Mitchell G, Pelletier V, Prevost C, Saudubray JM, Durand G, Seta N. Protein losing enteropathy-hepatic fibrosis syndrome in Saguenay-Lac St-Jean, Quebec is a congenital disorder of glycosylation type Ib. J Med Genet 39: 849-51. 2002. 115. Wada Y, Nishikawa A, Okamoto N, Inui K, Tsukamoto H, Okada S, Taniguchi N. Structure of serum transferrin in carbohydrate-deficient glycoprotein syndrome. Biochem Biophys Res Commun 189: 832-6. 1992. 116. Wada Y, Sakamoto M. Isolation of the human phosphomannomutase gene (PMM1) and assignment to chromosome 22q13. Genomics 39: 416-7. 1997. 117. Wang Y, Tan J, Sutton-Smith M, Ditto D, Panico M, Campbell RM, Varki NM, Long JM, Jaeken J, Levinson SR, Wynshaw-Boris A, Morris HR, Le D, Dell A, Schachter H, Marth JD. Modeling human congenital disorder of glycosylation type IIa in the mouse: conservation of asparagine-linked glycan-dependent functions in mammalian physiology and insights into disease pathogenesis. Glycobiology 11: 1051-70. 2001. 118. Westphal V, Peterson S, Patterson M, Tournay A, Blumenthal A, Treacy EP, Freeze HH. Functional significance of PMM2 mutations in mildly affected patients with congenital disorders of glycosylation Ia. Genet Med 3: 393-8. 2001a. 119. Westphal V, Kjaergaard S, Davis JA, Peterson SM, Skovby F, Freeze HH. Genetic and metabolic analysis of the first adult with congenital disorder of glycosylation type Ib: long-term outcome and effects of mannose supplementation. Mol Genet Metab 73: 77-85. 2001b. 120. Westphal V, Kjaergaard S, Schollen E, Martens K, Grünewald S, Schwartz M, Matthijs G, Freeze HH. A frequent mild mutation in ALG6 may exacerbate the clinical severity of patients with congenital disorder of glycosylation Ia (CDG-Ia) caused by phosphomannomutase deficiency. Hum Mol Genet 11: 599-604. 2002. 121. Yamashita K, Ideo H, Ohkura T, Fukushima K, Yuasa I, Ohno K, Takeshita K. Sugar chains of serum transferrin from patients with carbohydrate deficient glycoprotein syndrome. Evidence of asparagine-Nlinked oligosaccharide transfer deficiency. J Biol Chem 268: 5783-9. 1993. 122. Yoshida A, Kobayashi K, Manya H, Taniguchi K, Kano H, Mizuno M, Inazu T, Mitsuhashi H, Takahashi S, Takeuchi M, Herrmann R, Straub V, Talim B, Voit T, Topaloglu H, Toda T, Endo T. Muscular dystrophy and neuronal migration disorder caused by mutations in a glycosyltransferase, POMGnT1. Dev Cell 1: 717-24. 2001. DANISH MEDICAL BULLETIN VOL. 51 NO. 4/NOVEMBER 2004 363