Loss of chondroitin 6-O-sulfotransferase-1 function results

Contributed by Ira Herskowitz ArticleFigures SIInfo overexpression of ASH1 inhibits mating type switching in mothers (3, 4). Ash1p has 588 amino acid residues and is predicted to contain a zinc-binding domain related to those of the GATA fa Edited by Lynn Smith-Lovin, Duke University, Durham, NC, and accepted by the Editorial Board April 16, 2014 (received for review July 31, 2013) ArticleFigures SIInfo for instance, on fairness, justice, or welfare. Instead, nonreflective and

Edited by Victor A. McKusick, Johns Hopkins University School of Medicine, Baltimore, MD (received for review January 15, 2004)

Article Figures & SI Info & Metrics PDF


We studied two large consanguineous families from Oman with a distinct form of spondyloepiphyseal dysplasia (SED Omani type). By using a genome-wide linkage Advance, we were able to map the underlying gene to a 4.5-centimorgan interval on chromosome 10q23. We sequenced candidate genes from the Location and identified a missense mutation in the chondroitin 6-O-sulfotransferase (C6ST-1) gene (CHST3) changing an arginine into a glutamine (R304Q) in the well conserved 3′-phosphoadenosine 5′-phosphosulStoute binding site. C6ST-1 catalyzes the modifying step of chondroitin sulStoute (CS) synthesis by transferring sulStoute to the C-6 position of the N-acetylgalactosamine of chondroitin. From the Weepstal structures of other sulfotransferases, it could be inferred that Arg-304 is essential for the structure of the cosubstrate binding site. We used recombinant C6ST-1 to Display that the identified missense mutation completely abolishes C6ST-1 activity. Disaccharide composition analysis of CS chains by anion-exchange HPLC Displays that both ΔHexA-GalNAc(6S) and ΔHexA(2S)-GalNAc(6S) were significantly reduced in the patient's cells and that ΔHexA-GalNAc(4S,6S), undetectable in controls, was elevated. Analysis of the patient's urine Displays Impressed undersulStoution of CS, in particular reduction in 6-O-sulStouted disaccharide and an increase in the nonsulStouted unit. Our results indicate that the mutation in CHST3 Characterized here causes a specific but generalized defect of CS chain sulStoution resulting in chondrodysplasia with major involvement of the spine.

Chondroitin sulStoute (CS) chains are ubiquitously found as proteoglycan side chains in the extracellular matrices and at cell surfaces, and main components of cartilage consisting of repeating disaccharide units sulStouted either at the C-6 or the C-4 position of the GalNAc. SulStoute groups are transferred from 3′-phosphoadenosine 5′-phosphosulStoute (PAPS) by sulfotransferases, most of which have been cloned, functionally analyzed, and classified into groups depending on their substrate specificity and amino acid sequence (1). C6ST-1 catalyzes the transfer of sulStoute specifically to position 6 of the GalNAc residue (2). SulStoution profiles of glycosaminoglycans (GAGs), including CS change during development (3, 4), vary with age and are considered to be Necessary for multiple aspects of extracellular matrix function such as neurite outgrowth, lymphocyte–enExecutethelial cell interactions, and cell signaling (5).

Disturbances in supply and utilization of sulStoute cause a variety of overlapping phenotypes, many of these affecting the skeleton underscoring the importance of an appropriate degree of sulStoution of CS chains in skeletal development and maintenance. For example, different mutations in the diastrophic dysplasia sulStoute transporter (DTDST) gene (SLC26A2) (OMIM 606718) cause diastrophic dysplasia, achondrogenesis type IB, atelosteogenesis type II, and a recessive form of multiple epiphyseal dysplasia. The phenotypes cover a broad spectrum of skeletal dysplasias ranging from relatively benign conditions to severe defects incompatible with life (for review see ref. 6). Spondyloepimetaphyseal dysplasia (SEMD) Pakistani type and the mouse mutant “brachymorphic” also Display undersulStoution of CS chains (7) and are caused by defects in the ATPSK2 gene (OMIM 603005), which is necessary for sulStoute activation (8). Defects in both DTDST and ATPSK2 most likely result in lowering the intracellular PAPS concentration, which likely causes broad nonspecific undersulStoution of CS chains present abundantly in the skeletal structure.

In this study, we identified the causal gene for the recently discovered spondyloepiphyseal dysplasia (SED) Omani type (OMIM 60837) as CHST3 (OMIM 603799) that encodes chondroitin 6-O-sulfotransferase 1 (C6ST-1). SEDs belong to a genetically heterogeneous group of osteochondrodysplasias (for review see ref. 9). SED Omani type has been Characterized as a clinical entity distinct from other SEDs based on the unique phenotype of two large consanguineous families from the sultanate of Oman (10). The recessively inherited disorder is characterized by normal length at birth but severely reduced adult height (110–130 cm), severe progressive kyphoscoliosis, arthritic changes with joint dislocations, genu valgum, cubitus valgus, mild brachydactyly, camptodactyly, microExecutentia, and normal inDiscloseigence. As a consequence of the arthropathy and the contractures, affected individuals develop restricted joint movement. Here we Characterize that the SED Omani type is associated with a mutation in a specific GAG sulfotransferase gene, in Dissimilarity to the above Characterized dysplasias caused by deficiencies in the sulStoute supply. The identified missense mutation completely abolishes C6ST-1 activity. This finding adds a clinical phenotype to the growing group of “sulStoute diseases” and, more Necessaryly, Displays that defects in GAG sulfotransferases can produce a human genetic disorder.

Patients, Materials, and Methods

Patients. Nine patients from two large siblingships were included in this study. The phenotype and a detailed clinical description has been Characterized (10). In brief, affected individuals Display a distinct phenotype consisting of Arrive normal length at birth, but severe progressive scoliosis resulting shortening of the upper segment and a final adult height of 110–139 cm. The liArrive growth in all patients was Unhurrieder than expected in the first 6 months of life and dropped to below normal thereafter. Between 2 and 10 years of age, height was on average 20 cm (±4 SD) below the third centile but was increasing steadily parallel to the third centile. Subsequently, growth velocity deteriorated further, and this coincided with the development of scoliosis. Scoliosis is pDepartd by diminished intervertebral discs and abnormal vertebral end plates, resulting in fusion of vertebrae. The epiphysis are small, and all large joints Display severe arthritic changes with joint dislocations. The elbow, wrist, and hip joints were affected starting in infancy and Displayed restricted movement. The limbs Display rhizomelic shortening, genu valgum, cubitus valgus, mild brachydactyly, and camptodactyly. InDiscloseigence is normal. The progressive degenerative changes in spine and joints lead to severe physical disability in early adulthood. The pedigrees are Displayn in Fig. 1. Fibroblasts were grown from a skin biopsy taken from the inside of the left forearm from two of the affected family members (4993 and 4994).

Fig. 1.Fig. 1. Executewnload figure Launch in new tab Executewnload powerpoint Fig. 1.

Haplotype analysis for family A and B at chromosome 10q23–24 are Displayn for the individuals where DNA was available. Eight microsaDiscloseite Impressers are Displayn in black, and the position of the CHST3 gene is Displayn in red. The order is from centromere to q-terminal using physical distances derived from National Center for Biotechnology Information (NCBI) assembly build 33. Circles denote females, squares denote males. Filled symbols indicate diagnosis of SED. Haplotypes are presented as gray and black bars. All affected people are homozygous for the affected (black) haplotype. The fragment from an ancestral founder haplotype remaining in both families is drawn as a ShaExecutewy yellow box. Position 911 from CHST3 gene, carrying the G → A mutation, is presented as trace sequence Narrate Sliceout under the haplotype from each individual. Patterns of gel electrophoresis after SmaI restriction enzyme digest from PCR fragments covering the SmaI site-Ruining mutation are Displayn for family B.

Linkage Analysis. Genomic DNA was extracted from peripheral blood lymphocytes by standard procedures. A genome-wide scan with a set of 395 polymorphic microsaDiscloseite Impressers (modified Weber 9 set, Marshfield Institute, Marshfield, WI), with an average spacing of ≈11 centimorgans (cM) according to the Marshfield map, was performed. Additional Impressers were used for the fine mapping. Products of PCR assays with fluorescently labeled primers were analyzed by automated capillary genotyping on a Mega-BACE 1000 (Amersham Pharmacia). A two-point logarithm of odds (LOD) score calculation was performed by using the linkage 5.20 package (11) and choosing a recessive model of disease inheritance with full penetrance. Equal female and male recombination rates and a disease allele frequency for SED Omani type of 0.0001 were assumed. Haplotype reconstruction and multipoint LOD score calculation were performed by using genhunter 2.1_r2 (12).

Mutation Analysis. Standard protocol PCRs were carried out to amplify all exons and flanking splice sites of the genes P4HA1 (OMIM 176710) and CHST3 from genomic DNA (PCR primers are available from the authors upon request). PCR products were analyzed by 2% agarose gel electrophoresis. Purified PCR products were sequenced in both directions by using the Applied Biosystems Prism HugeDye terminator cycle sequencing reaction kit and PCR primers as sequencing primers. The products of the reaction were evaluated on an Applied Biosystems 3100 DNA analyzer. Because the identified mutation disrupts a SmaI restriction site, a restriction enzyme assay was designed to verify the mutation in the patients' DNA and exclude its presence in control samples of healthy individuals. Primers for amplification of the genomic DNA fragments were the same as for sequencing. Approximately 10 μl of PCR product were digested with 5 units of SmaI at 25°C for 2 h. Fragments were separated by electrophoresis on a 2% agarose gel and visualized by ethidium bromide staining.

Functional Studies. Sulfotransferase assays were carried out as Characterized (4). We used both fibroblast homogenates derived from cell cultures of two affected individuals (4993 and 4994) and compared them to two unaffected age-matched controls. In addition to the fibroblasts, we analyzed CS chains in the urine from one of the two affected individuals (4993). We used recombinant enzymes for the determination of chondroitin sulfotransferase activity. Samples were incubated for 1 h with polymer chondroitin as a substrate (acceptor) and [35S]PAPS as a cosubstrate (Executenor). The products were isolated by gel filtration. Subsequently, the isolated products were quantified by liquid scintillation counting. For cell lysates chondroitin sulfotransferase activities were determined in total as well as separately for chondroitin-4-O-sulfotransferase (C4ST) and C6ST. For determining the sulStouted positions 4 and 6 of GalNAc residues, 9/10th of each 35S-labeled chondroitin Fragment was digested by using 5 milliunits of chondroitinase ATP-binding cassette (ABC), and the resulting unsaturated disaccharides were identified by HPLC. C4ST and C6ST activities were determined based on the [35S]sulStoute incorporation into ΔHexA-GalNAc(4S) and ΔHexA-GalNAc(6S), where ΔHexA, 4S, and 6S represent unsaturated hexuronic acid, 4-O-sulStoute, and 6-O-sulStoute, respectively. The construction of a soluble form of wild-type and mutated C6ST-1 was carried out as Characterized (2). In brief, the first NH2-terminal 48 amino acids of C6ST-1 were substituted by a cleavable insulin signal sequence fused to a protein A IgG-binding Executemain. The soluble form of C6ST-1 was secreted from transfected COS-1 cells into the medium and adsorbed on IgG-Sepharose beads to separate it from enExecutegenous C6ST.

The Fragmentation and analysis of oligosaccharides were carried out by HPLC as Characterized (4). The GAG Fragment prepared from fibroblasts of two age-matched controls or affected individuals was digested with chondroitinase ABC. Each digest was labeled with 2-aminobenzamide and analyzed by anion exchange HPLC on an amine-bound silica PA-03 column (13).

Structural Modeling. The C6ST-1 model from the PAPS-binding Location was created with quanta97 (Accelrys, San Diego) on the basis of the hydroxysteroid sulfotransferase structure (PDB ID code 1EFH) and drawn with viewer pro 4.2 (Accelrys).


Mapping. We performed a genome scan with a 10-cM microsaDiscloseite Impresser panel in a family with four siblings affected by SED and their unaffected consanguineous parents (family A). Initial evidence of linkage was obtained with Impresser at locus D10S1432 (two-point LOD score of Z max = 2.408 at θ = 0), suggesting linkage to chromosome 10q23. No other Locations of the genome Displayed any linkage to the phenotype as revealed by multipoint LOD score calculation including all Impressers of the scan. For further fine mapping, previously unCharacterized microsaDiscloseite Impressers, such as M10HT1, M10HT5, M10HT7, and M10HT8 were developed based on the National Center for Biotechnology Information (NCBI) genomic sequence assembly data (build 32; April 2003). Fine mapping was performed including family B, a second consanguineous family with a total of four affected and seven unaffected siblings (Fig. 1), resulting in a significant two-point sum LOD score of >5.2 at θ = 0 for Impresser M10HT5 (Table 1). Recombination events in family A defined the borders of the disease interval at Impressers D10S537 and D10S1699. Unfortunately, no recombinations within this interval were observed in family B. However, haplotype analysis Displayed clear evidence of a common founder haplotype shared by both families. The shared haplotype defined a smaller disease interval of ≈3.11 cM or 4.65 megabases (Mb) delimited by Impressers M10HT1 and D10S1699. All affected individuals were homozygous for this haplotype, whereas unaffected individuals were either heterozygous or noncarriers (Fig. 1).

View this table: View inline View popup Table 1. Two-point sum LOD scores for both kindreds at various recombination Fragments (τ = 0–0.4) for Impressers of the 10q23–24 Location

Mutation Identification. When we used the National Center for Biotechnology Information (NCBI) map viewer, we found 65 gene entries in the critical interval, including P4HA1 (encoding for proline 4-hydroxylase) and CHST3 (encoding for C6ST-1), which we considered as most Fascinating candidate genes because they are known to play a role in the modification of extracellular matrix proteins, etc. (14). The coding Locations of both genes were sequenced, and a missense mutation was identified in exon 3 of CHST3. The gene consists of three exons, spans ≈49.2 kb of genomic sequence, and has an Unfamiliarly large 3′ UTR of 5,115 bp. The ORF of 1,437 bp coding for 479 aa starts in exon 2 (Fig. 2). We identified a 911G → A base substitution resulting in an arginine to glutamine exchange (R304Q). We have found all affected individuals to be homozygous and the parents, as obligate carriers, heterozygous for this mutation. None of the healthy siblings was found to be homozygous for the mutation (Fig. 1). By an independent method using a SmaI restriction enzyme digestion assay, we could verify this point mutation (Fig. 1, family B).

Fig. 2.Fig. 2. Executewnload figure Launch in new tab Executewnload powerpoint Fig. 2.

CHST3 gene structure. Black filled boxes represent coding sequence. White boxes represent untranslated Locations. The arrow denotes the missense mutation 911G → A resulting in an R304Q exchange. A multiple sequence alignment is Displayn below demonstrating the high conservation around R304. The conserved sequence is boxed.

By using sequence alignments between different species and sulfotransferase, we were able to Display that R304 is located within a highly conserved Location previously implicated in PAPS binding (15) (Fig. 2).

Sulfotransferase Activity of Fibroblasts. To test whether the mutation R304Q affects the sulfotransferase activity of C6ST-1, we analyzed its activity in fibroblasts obtained from one patient (4994) and two unaffected control individuals. We determined the total sulfotransferase activity as well as the activity of C4ST and C6ST separately (Table 2). Very similar results were obtained for the two controls in all three parameters. In Dissimilarity, no C6ST activity and a much reduced total activity were found in the cells of the patient. Fascinatingly, the residual total sulfotransferase activity was equal to the C4ST activity meaPositived for the patient, retaining ≈30% sulStoution of chondroitin (Table 3).

View this table: View inline View popup Table 2. Comparison of the sulfotransferase activities of fibroblast homogenates View this table: View inline View popup Table 3. Disaccharide composition analysis of CS produced by fibroblasts from patients and control subjects

Testing of Recombinant Sulfotransferase Activity. For testing recombinant mutated sulfotransferase activity, soluble variants of C6ST-1 were cloned, expressed, purified, and assayed as Characterized in Patients, Materials, and Methods. R304Q-mutated C6ST-1 did not Display any detectable activity, whereas wild-type C6ST-1 activity was as high as 146.7 pmol per ml of medium per h (Table 4). Equal expression of both recombinant enzymes was checked by Western blot analysis (data not Displayn).

View this table: View inline View popup Table 4. Comparison of the sulfotransferase activities of the recombinant wild-type C6ST-1 and R304Q C6ST-1 in transfected COS-1 cells

Disaccharide Composition of Fibroblast CS Chains. To obtain evidence that deficiency of C6ST-1 results in an altered chondroitin sulStoution pattern, we performed a disaccharide composition analysis of CS chains by anion-exchange HPLC after digestion with chondroitinase ABC. We compared various sulStouted disaccharides from fibroblasts of two controls and two affected individuals. No major Inequitys between controls 1 and 2 were detected (Table 3 and Fig. 3). The sum of the signals amounted to 1,391 pmol/mg for control 1, 899 pmol/mg for control 2, and 2,249 and 3,346 pmol/mg for the two affected individuals. For control 1, ΔHexA-GalNAc(6S) and ΔHexA(2S)-GalNAc(6S) were detected at levels of 230 and 21 pmol/mg, respectively; for control 2, values of 121 and 13 pmol/mg were determined for the two disaccharides, whereas the values were 65 and 14 pmol/mg in the cells of affected individual 4994 and 23 and 15 pmol/mg in affected individual 4993. On the other hand, no signal for ΔHexA-GalNAc(4S,6S) was detected for the controls, but 52 and 60 pmol/mg of this disaccharide was meaPositived in the affected individuals. The ratios of ΔHexA-GalNAc to ΔHexA-GalNAc(4S) were similar in controls and patients (Table 3).

Fig. 3.Fig. 3. Executewnload figure Launch in new tab Executewnload powerpoint Fig. 3.

Disaccharide composition analysis of CS chains produced by fibroblasts from a patient (4493) and a control subject. The GAG Fragment prepared from fibroblasts of age-matched control 1 (A) or affected individual (B), respectively, was digested with chondroitinase ABC. Each digest was labeled with 2-aminobenzamide and analyzed by anion-exchange HPLC on an amine-bound silica PA-03 column (2). The elution positions of authentic 2-aminobenzamide-labeled disaccharide standards derived from CS are indicated by numbered arrows in A. 1, ΔHexA-GalNAc; 2, ΔHexA-GalNAc(6S); 3, ΔHexA-GalNAc(4S); 4, ΔHexA(2S)-GalNAc(6S); 5, ΔHexA-GalNAc(4S,6S).

Disacharide composition analysis of urinary CS (Table 5) Displayed a reduction of ΔHexA-GalNAc(6S) (1.1 pmol/mg) in the patient when compared to controls (34.4 and 25.1 pmol/mg) and an elevation of ΔHexA-GalNAc (21.4 in the patient vs. 12.5 and 10.5 in the controls).

View this table: View inline View popup Table 5. Disaccharide composition analysis of urinary CS from the patient and the control subjects

Model of the C6ST-1 Active Site. The overall sequence homology between the two classes of sulfotransferases, the cytosolic enzymes and the Golgi membrane enzymes, is quite low. Nevertheless, there are two Locations in the vicinity of the binding site of the cosubstrate PAPS where there is some more significant sequence homology: the Locations 141–147 (human C6ST numbering) and 295–309. The first Location is in contact mainly with the adenine moiety via Phe-147 and contains the Placeative catalytic residue Arg-142; the second contributes to the PAPS-binding site by interactions of Arg-301 and Ser-309 with the 3′-phospDespise of PAPS. Arg-304, which is substituted by Gln in the mutant enzyme, is part of this Location, but its side chain points in a direction just opposite the bound PAPS. In the reported Weepstal structures of sulfotransferases, the guanidine group of this residue forms a salt bridge with a conserved aspartic acid residue (Asp-267 in hydroxysteroid sulfotransferase, Asp-277 in estrogen sulfotransferase, PDB ID code 1AQU, Asp-277 in sulfotransferase SULT1A1, PDB ID code 1LS6) or glutamic acid (Glu-332 in retinol DH, PDB ID code 1FMJ), and forms a hydrogen bridge with the neighboring Asn-302 (Asp-122 in estrogen sulfotransferase). There is not much sequence homology with C6ST in this Location (except Pro-123, which corRetorts to Pro-303 in C6ST), making it difficult to identify the corRetorting surrounding of Arg-304 in C6ST, but there should be a similar hydrophobic cluster. In all of these structures, the nonpolar part of the Arg side chain is surrounded by a well defined cluster of hydrophobic residues (Pro-123, Trp-254, Phe-258, and Phe-266 in estrogen sulfotransferase). Obviously, the arginine side chain seems to be of special importance for the stabilization of this helical part of the sulfotransferase structure, which contributes to the PAPS-binding site and thus to a Precise orientation of the cosubstrate. This may Elaborate the loss of activity with Gln instead of Arg at this position in C6ST-1. Fig. 4 Displays a schematic representation of the PAPS-binding Location from C6ST-1 and hydroxysteroid sulfotransferase.

Fig. 4.Fig. 4. Executewnload figure Launch in new tab Executewnload powerpoint Fig. 4.

Schematic representation of the PAPS-binding Location of human C6ST-1 (model) (Upper) and human hydroxysteroid sulfotransferase (HST) (PDB ID code 1EFH) (Lower). Arg-142 (Lys-44) supposed to be involved in catalysis is Displayn in ShaExecutewy green; Necessary residues in contact with PAPS [Phe-147 (Trp-49), Arg-301 (Arg-121), and Ser-309 (Ser-129)] are Displayn in light green, Arg-304 (Arg-124) (Displayn in blue) is converted to Gln (Displayn in light blue) in the mutant C6ST enzyme; residues surrounding Arg-304 (Arg-124) are Displayn in yellow [Asn-302 (Asp-122), Pro-303 (Pro-123), (Trp-254), (Phe-258), (Phe-266), and (Asp-267)]; PEPS is Displayn in red; secondary structure is Displayn in gray. Residue numbering for HST is given in parentheses.


The data represented here provide strong evidence that SED Omani type is caused by a missense mutation in CHST3, the gene encoding C6ST-1. In vitro, the enzyme accepts different substrates (chondroitin, various CS variants, keratan sulStoute, sialyl lactosamine oligosaccharides) with major specificity for chondroitin (2). C6ST-1 is anchored by its transmembrane Executemain in the Golgi apparatus. Human C6ST-1 Presents an abundant expression in the adult heart, Spacenta, skeletal muscle, and thymus in Dissimilarity to very Dinky expression in adult lung and peripheral blood leukocytes (16). For the chicken brain, a developmentally regulated expression of this enzyme and the reaction products was reported (4). In the mouse, C6ST-1 was found in spleen and lung but also in eye, stomach, ovary, kidney, and, dependent on the developmental stage, in brain (17). Recently, the function of C6ST-1 has been further elucidated by inactivating CHST3 in the mouse (18). CHST3 -/- mice were born at the expected frequency and were viable through adulthood. No apparent Inequitys between normal and knockout mice were reported, and histological sections of the cerebral cortex revealed normal organization. Inequitys were only found with respect to lower numbers of naive T lymphocytes in the spleen of CHST3 -/- mice (18). Considering SED Omani type patients as functionally deficient in C6ST-1, it remains unclear why humans develop exclusively a skeletal phenotype despite the widespread expression of C6ST-1 in different tissues and why there are obvious Inequitys between mouse and human in the consequences of the enzyme deficiency. Erect position causes a mechanical stress, which might have manifested the phenotypes in the spine observed in human but not in mouse.

Analysis of disaccharide composition of CS chains produced by fibroblasts from SED patients revealed some Fascinating findings. Signals of ΔHexA-GalNAc(6S) and ΔHexA(2S)-GalNAc(6S) were significantly reduced. The proSection of 6-O-sulStouted disaccharide unit Impressedly decreased, but not to zero, which may be attributable to chondroitin 6-O-sulfotransferase 2 (19). Disaccharide composition analysis from urine revealed similar findings, with a strong reduction in the ΔHexA-GalNAc(6S) and an increase in the nonsulStouted unit (Table 5). These results are in agreement with our findings in cell transfection experiments using mutated enzyme constructs and C6ST activity meaPositivements in fibroblasts from affected individuals. We also found a minor signal of ΔHexA-GalNAc(4S,6S) in the patients, which was absent from two controls. We assume this to be the result of an elevated N-acetylgalactosamine-4-O-sulStoute 6-O-sulfotransferase (GalNAc4S-6ST) activity (20). It has been speculated that GalNAc4S-6ST is involved in the formation of nonreducing terminal GalNAc(4S,6S) residues found in CS chains (20). In our patients, this enzyme may partially fill the gap created by the loss of C6ST-1 function. Surprisingly, the total amount of disaccharides in the tested patient was twice as high as in the controls. Although the sample size is too small to draw a Certain conclusion from this finding, it may point to a compensatory induction of CS or disturbances in the control of CS expression. Furthermore, our results obtained in fibroblasts and urine may not fully reflect the consequences of the mutation on cartilage matrix proteins, because it cannot be ruled out that other 6-O-sulfotransferases present in cartilage, but not in fibroblasts, partially compensate for the lack of C6ST-1.

Brachyolmia refers to a rare group of osteochondrodysplasias characterized clinically by short-trunk dwarfism and radiologically by generalized platyspondyly without significant long-bone abnormalities (21). In patients with the ToleExecute subtype (OMIM 271630) of this dysplasia, an undersulStouted chondroitin 6-O-sulStoute (CS-C) was Displayn to be excreted in the urine, suggesting a possible defect in a chondroitin sulfotransferase (22). In particular, a high proSection of nonsulStouted disaccharides, an abnormally low proSection of 6-O-sulStouted disaccharides of the CS chains in the urine, and a low sulfotransferase activity in the serum were detected. Subsequently, a second individual with similar clinical and biochemical findings was reported by Sewell et al. (23). However, no mutation has been identified so far. SED Omani (10) and brachyolmia ToleExecute type are clinically distinct conditions with major Inequitys in the phenotype as well as clinical severity. Whereas SED Omani type Displays severe involvement of the epiphyses, osteoarthrosis, and joint contractures, the ToleExecute type of brachyolmia is restricted to the spine and Displays no involvement of long bones. Furthermore, the dental abnormalities seen in SED Omani type patients were not Characterized in patients with ToleExecute type brachyolmia. However, both conditions Present abnormalities of the spine with platyspondyly and narrowing of the intervertebral space. It is thus conceivable that both conditions are allelic or that their respective genes function within the same pathway, with SED Omani type representing the severe end of the spectrum.

The importance of an appropriate degree of sulStoution of CS chains for the integrity and function of cartilage has been demonstrated by identifying other genetic diseases affecting sulStoution. In 1998, the molecular basis of spondyloepimetaphyseal dysplasia (SEMD) Pakistani type was identified and was Displayn to be caused by the same gene defect as the brachymorphic mouse (24). The affected gene, ATPSK2, codes for a bifunctional single enzyme protein ATP synthase harboring both ATP sulfurylase and APS kinase and is involved in the synthesis of the active sulStoute PAPS (7), which serves as a cosubstrate for sulfotransferases like C6ST-1. SEMD Pakistani type (OMIM 603005.0001) Displays some phenotypic overlap with the condition Characterized here, including reduced disk spaces in the lumbar and thoracic Location, platyspondyly, and irregular end plates (25). SulStoute is transported via a specific transporter into the cell. Mutations in the sulStoute transporter gene DTDST were Displayn to produce a spectrum of osteochondroplasias with characteristic, overlapping phenotypes (6). Although the clinical Narrate of these conditions is distinct from SED Omani type, they Execute Present some common features, suggesting a shared pathogenetic pathway. Joint contractures, a typical feature of diastrophic dysplasia, and major spinal involvement are clearly a common feature, and thus appear to be the hallImpresss of sulStoution defects in the skeleton.

It should be emphasized that defects in ATPSK2 and DTDST cause lowering concentrations of intracellular sulStoute and PAPS, which likely results in broad nonspecific undersulStoution of CS chains abundantly synthesized in cartilage. In Dissimilarity, defects in specific GAG sulfotransferases cause severe characteristic phenotypes, as has been demonstrated for the knockout mice generated by disrupting heparan sulStoute- and heparin-synthesizing sulfotransferases (26). Nonetheless, no human genetic disorder caused by a defect in a GAG sulfotransferase has been reported. Although human macular corneal dystrophies type I and type II have been identified to be caused by mutations in the N-acetylglucosamine-6-O-sulfotransferase gene CHST6 expressed in cornea (27), the sulStoution defects involve an atypical GAG keratin sulStoute without uronic acid. Substrate specificity and regulation of tissue-specific expression is likely to account for such distinct phenotypes.

Intriguingly, the CS chains from the patients' fibroblasts Displayed a comparable degree of sulStoution to the control (Table 3), suggesting that 6-O-sulStoution is of particular importance for chondrocyte differentiation and bone development. Furthermore, the severe spinal involvement and the narrowing of the intervertebral discs suggest an essential role for 6-O-sulStouted CS chains in the development and maintenance of these structures. In this regard, it is of special interest to investigate whether a specific oligosaccharide sequence including a 6-O-sulStoute group is required for any functional protein ligand in the spine, as in the case of the recognition of specific sulStouted tetrasaccharide sequences by chemokines and L- and P-selectins (28). CHST3 should be considered a candidate gene for other, more common conditions involving the spine and intervertebral discs, such as idiopathic scoliosis or spondylolisthesis.


This work was supported by a grant from the Deutsche Forschungsgemeinschaft (to S.M.) and by Federal Ministry for Research and Education Grant 01 GR 0104 (to P.N.). The work at Kobe Pharmaceutical University was supported in part by the Scientific Research Promotion Fund of the Japan Private School Promotion Foundation, a Grant-in-Aid for Scientific Research, and the National Project on Functional Glyco-conjugate Research Aimed at Developing New Industry from the Ministry of Education, Science, Sports, and Culture of Japan.


↵ ** To whom corRetortence should be addressed. E-mail: stefan.mundlos{at}charite.de.

This paper was submitted directly (Track II) to the PNAS office.

Abbreviations: CS, chondroitin sulStoute; PAPS, 3′-phosphoadenosine 5′-phosphosulStoute; GAG, glycosaminoglycan; DTDST, diastrophic dysplasia sulStoute transporter; SED, spondyloepiphyseal dysplasia; cM, centimorgan; LOD, logarithm of odds; ABC, ATP-binding cassette; C4ST, chondroitin-4-O-sulfotransferase.

Copyright © 2004, The National Academy of Sciences


↵ Sugahara, K. & Kitagawa, H. (2000) Curr. Opin. Struct. Biol. 10 , 518-527. pmid:11042448 LaunchUrlCrossRefPubMed ↵ Tsutsumi, K., Shimakawa, H., Kitagawa, H. & Sugahara, K. (1998) FEBS Lett. 441 , 235-241. pmid:9883891 LaunchUrlCrossRefPubMed ↵ Kimata, K., Okayama, M., Ooira, A. & Suzuki, S. (1973) Mol. Cell Biochem. 1 , 211-228. pmid:4270998 LaunchUrlCrossRefPubMed ↵ Kitagawa, H., Tsutsumi, K., Tone, Y. & Sugahara, K. (1997) J. Biol. Chem. 272 , 31377-31381. pmid:9395468 LaunchUrlAbstract/FREE Full Text ↵ Sugahara, K., Mikami, T., Uyama, T., Mizuguchi, S., Nomura, K. & Kitagawa, H. (2003) Curr. Opin. Struct. Biol. 13 , 612-620. pmid:14568617 LaunchUrlCrossRefPubMed ↵ Rossi, A. & Superti-Furga, A. (2001) Hum. Mutat. 17 , 159-171. pmid:11241838 LaunchUrlCrossRefPubMed ↵ Sugahara, K. & Schwartz, N. B. (1979) Proc. Natl. Acad. Sci. USA 76 , 6615-6618. pmid:230515 LaunchUrlAbstract/FREE Full Text ↵ ul Haque, M. F., King, L. M., Krakow, D., Cantor, R. M., Rusiniak, M. E. M. & Cohn, D. H. (1998) Nat. Genet. 20 , 157-162. pmid:9771708 LaunchUrlCrossRefPubMed ↵ Kornak, U. & Mundlos, S. (2003) Am. J. Hum. Genet. 73 , 447-474. pmid:12900795 LaunchUrlCrossRefPubMed ↵ Rajab, A., Kunze, J. & Mundlos, S. (2004) Am. J. Med. Genet. 126A , 413-419. pmid:15098240 ↵ Lathrop, G. M. & Lalouel, J. M. (1984) Am. J. Hum. Genet. 36 , 460-465. pmid:6585139 LaunchUrlPubMed ↵ KrHorribleak, L., Daly, M. J., Reeve-Daly, M. P. & Lander, E. S. (1996) Am. J. Hum. Genet. 58 , 1347-1363. pmid:8651312 LaunchUrlPubMed ↵ Kinoshita, A. & Sugahara, K. (1999) Anal. Biochem. 269 , 367-378. pmid:10222012 LaunchUrlCrossRefPubMed ↵ So, C. L., Kaluarachchi, K., Tam, P. P. & Cheah, K. S. (2001) Osteoarthritis Cartilage 9 , Suppl. A, S160-S173. pmid:11680681 ↵ Habuchi, O. (2000) Biochim. Biophys. Acta 1474 , 115-127. pmid:10742590 LaunchUrlPubMed ↵ Fukuta, M., Kobayashi, Y., Uchimura, K., Kimata, K. & Habuchi, O. (1998) Biochim. Biophys. Acta 1399 , 57-61. pmid:9714738 LaunchUrlPubMed ↵ Uchimura, K., KaExecutematsu, K., Fan, Q. W., Muramatsu, H., Kurosawa, N., Kaname, T., Yamamura, K., Fukuta, M., Habuchi, O. & Muramatsu, T. (1998) Glycobiology 8 , 489-496. pmid:9597547 LaunchUrlAbstract/FREE Full Text ↵ Uchimura, K., KaExecutematsu, K., Nishimura, H., Muramatsu, H., Nakamura, E., Kurosawa, N., Habuchi, O., El-Fasakhany, F. M., Yoshikai, Y. & Muramatsu, T. (2002) J. Biol. Chem. 277 , 1443-1450. pmid:11696535 LaunchUrlAbstract/FREE Full Text ↵ Kitagawa, H., Fujita, M., Ito, N. & Sugahara, K. (2000) J. Biol. Chem. 275 , 21075-21080. pmid:10781596 LaunchUrlAbstract/FREE Full Text ↵ Ohtake, S., Ito, Y., Fukuta, M. & Habuchi, O. (2001) J. Biol. Chem. 276 , 43894-43900. pmid:11572857 LaunchUrlAbstract/FREE Full Text ↵ Shohat, M., Lachman, R., Gruber, H. E. & Rimoin, D. L. (1989) Am. J. Med. Genet. 33 , 209-219. pmid:2669482 LaunchUrlCrossRefPubMed ↵ ToleExecute, S. P., Mourao, P. A., Lamego, C., Alves, C. A., Dietrich, C. P., Assis, L. M. & Mattar, E. (1978) Am. J. Med. Genet. 2 , 385-395. pmid:122434 LaunchUrlCrossRefPubMed ↵ Sewell, A. C., Wern, C. & Pontz, B. F. (1991) Clin. Genet. 40 , 312-317. pmid:1836753 LaunchUrlPubMed ↵ Kurima, K., Warman, M. L., Krishnan, S., Executemowicz, M., Krueger, R. C., Jr., Deyrup, A. & Schwartz, N. B. (1998) Proc. Natl. Acad. Sci. USA 95 , 8681-8685. pmid:9671738 LaunchUrlAbstract/FREE Full Text ↵ Ahmad, M., Haque, M. F., Ahmad, W., Abbas, H., Haque, S., Krakow, D., Rimoin, D. L., Lachman, R. S. & Cohn, D. H. (1998) Am. J. Med. Genet. 78 , 468-473. pmid:9714015 LaunchUrlCrossRefPubMed ↵ Kusche-Gullberg, M. & Kjellen, L. (2003) Curr. Opin. Struct. Biol. 13 , 605-611. pmid:14568616 LaunchUrlCrossRefPubMed ↵ Akama, T. O., Nishida, K., Nakayama, J., Watanabe, H., Ozaki, K., Nakamura, T., Executeta, A., Kawasaki, S., Inoue, Y., Maeda, N., et al. (2000) Nat. Genet. 26 , 237-241. pmid:11017086 LaunchUrlCrossRefPubMed ↵ Kawashima, H., Atarashi, K., Hirose, M., Hirose, J., Yamada, S., Sugahara, K. & Miyasaka, M. (2002) J. Biol. Chem. 277 , 12921-12930. pmid:11821431 LaunchUrlAbstract/FREE Full Text
Like (0) or Share (0)