GlcNAc6ST-1 regulates sulfation of N-glycans and myelination in the peripheral nervous system

Highly specialized glial cells wrap axons with a multilayered myelin membrane in vertebrates. Myelin serves essential roles in the functioning of the nervous system. Axonal degeneration is the major cause of permanent neurological disability in primary myelin diseases. Many glycoproteins have been identified in myelin, and a lack of one myelin glycoprotein results in abnormal myelin structures in many cases. However, the roles of glycans on myelin glycoproteins remain poorly understood. Here, we report that sulfated N-glycans are involved in peripheral nervous system (PNS) myelination. PNS myelin glycoproteins contain highly abundant sulfated N-glycans. Major sulfated N-glycans were identified in both porcine and mouse PNS myelin, demonstrating that the 6-O-sulfation of N-acetylglucosamine (GlcNAc-6-O-sulfation) is highly conserved in PNS myelin between these species. P0 protein, the most abundant glycoprotein in PNS myelin and mutations in which at the glycosylation site cause Charcot-Marie-Tooth neuropathy, has abundant GlcNAc-6-O-sulfated N-glycans. Mice deficient in N-acetylglucosamine-6-O-sulfotransferase-1 (GlcNAc6ST-1) failed to synthesize sulfated N-glycans and exhibited abnormal myelination and axonal degeneration in the PNS. Taken together, this study demonstrates that GlcNAc6ST-1 modulates PNS myelination and myelinated axonal survival through the GlcNAc-6-O-sulfation of N-glycans on glycoproteins. These findings may provide novel insights into the pathogenesis of peripheral neuropathy.


N-glycan analysis and separation by HPLC
Analyses of PA-N-glycans using HPLC were performed as described previously 5,6 . To separate neutral and anionic N-glycans, PA-N-glycans were passed through an anion-exchange DEAE column (TSKgel DEAE-5PW, Tosoh, Tokyo, Japan) using HPLC. Anion-exchange column HPLC purification was performed at a flow rate of 1.0 ml/min at room temperature. The mobile phase consisted of solvent A (distilled water adjusted to pH 9.0 with aqueous ammonia) and solvent B (0.5 M ammonium acetate titrated to pH 9.0 with aqueous ammonia). The column was equilibrated with solvent A. After a sample had been injected, solvent B was 0% in the first 5 min, and then increased linearly to 60% in the next 37 min. For analysis of porcine samples, P 0 -CNS and PLP-null mouse samples, solvent B was 0% in the first 2 min, and then increased linearly to 12% for 3 min and 100% in the next 44 min. PA-sugar chains were detected at excitation and emission wavelengths of 310 nm and 380 nm, respectively (FP-2025 Plus, Jasco Corporation, Hachioji, Japan). Peaks between neutral and anionic N-glycans are derived from contaminants that are not removed during N-glycan purification 6 . RP-HPLC for anionic PA-N-glycans was performed on a CAPCELL PAK C18 column (SG120, Cat. No. 12512; Shiseido, Tokyo, Japan) at a flow rate of 0.6 ml/min at 45ºC. Solvent C consisted of 0.6% acetic acid buffer (pH 4.0) containing 0.28% triethylamine, and solvent D consisted of solvent C containing 5% acetonitrile. The column was equilibrated with mixtures of solvent C and solvent D (initial ratio 95:5) that were increased linearly to 100% in 60 min. After that, the ratio of 100% of solvent D was kept for 1 min.
PA-sugar chains were detected at an excitation wavelength of 310 nm and an emission wavelength of 380 nm.
Neutral PA-N-glycans of varying sizes were separated by HPLC using a NP-column (Shodex Asahipak NH2P-50 4E, 4.6 x 250 mm; Showa Denko K.K., Tokyo, Japan) at a flow rate of 0.6 ml/min at 30ºC. The mobile phase consisted of solvent E (93% acetonitrile and 0.3% acetic acid titrated to pH 7.0 with 1 M aqueous ammonia) and solvent F (20% acetonitrile and 0.3% acetic acid titrated to pH 7.0 with 1 M aqueous ammonia). The column was equilibrated with mixtures of solvent E and solvent F (80:20) that were linearly increased to 49% in 240 min and then to 90% in 7 min. NP-HPLC analysis was performed using the Prominence HPLC system equipped with a fluorescence detector (excitation and emission wavelengths were 310 and 380 nm, respectively; Shimadzu, Kyoto, Japan). Each detected PA-N-glycan was further analyzed by RP-HPLC. RP-HPLC was performed on a Develosil C30-UG-5 column (4.6 x 150 mm; Nomura Chemical, Seto, Japan) at a flow rate of 0.5 ml/min at 30ºC. Solvent G consisted of 5 mM ammonium acetate buffer (pH 4.0), and solvent H consisted of solvent G containing 10% acetonitrile. The column was equilibrated with mixtures of solvent G and solvent H (initial ratio 75:25) that were increased linearly to 47% in 55 min and then to 70% in 5 min. PA-sugar chains were detected at excitation and emission wavelengths of 320 and 400 nm, respectively. N-glycan structures were identified by calculating the Mannose-Unit value from NP-HPLC, and the Glucose-Unit value from RP-HPLC, as described previously 7,8 , or by comparison with known standards and sequential exoglycosidase digestion.

Identification of the anionic N-glycans in mouse PNS myelin
The anionic fractions of the major peaks 5-9 in Fig. 2A were individually collected and further analyzed. The N-glycans from peaks 5 and 6 were desialylated by neuraminidase. After neuraminidase treatment, the neutral N-glycans were collected using DEAE HPLC (Supplementary Figs. S2A and S2B) and analyzed by NP-HPLC (Figs. 2B and 2C). After major peaks 5a, 5b, 5c (from peak 5) and 6' (from peak 6) were collected, the N-glycans of these peaks were further analyzed and identified by RP-HPLC using known standards α2,3-sialidase-resistant, but neuraminidase-sensitive. After the N-glycans from peaks 7+8 were treated with neuraminidase, the N-glycan elution profiles coincided with GP3 and desialylated GP4 profiles on RP-HPLC.
Taken together, the presumed N-glycan structure from peaks 7+8 and 9 are shown in Fig. 2F. There are two possibilities for the N-glycan structure from peak 9.

Identification of the main neutral N-glycan in mouse PNS myelin
The neutral fraction in Fig. 2A was collected using DEAE HPLC and analyzed by NP-HPLC ( Supplementary   Fig. S4A). After the fraction of the main peak N1 was collected, the N-glycans from peak N1 were further separated by RP-HPLC ( Supplementary Fig. S4B). The main peak N2 was identified using a known standard.
This neutral N-glycan structure is shown in Supplementary Fig. S4C.

Data quantification and analysis of PA-N-glycans
PA-N-glycans were quantified and analyzed as described previously 5,6 . HPLC chromatogram data were analyzed using LC station software (Shimadzu) and Empower2 software (Waters, Milford, MA).

RT-PCR
Total RNA was extracted from sciatic nerves and the medulla oblongata of 12-week-old young adult mice with Sepasol-RNA I Super (Nacalai Tesque) according to the manufacturer's instructions. cDNA was synthesized from 1 µg of total RNA using ReverTra Ace reverse transcriptase (Toyobo, Osaka, Japan) with random 6-mer primers. The retrotranscription reaction was subjected to PCR amplification using the following primers and KAPA Taq EXtra (Nippon Genetics, Tokyo, Japan) in a thermal cycler. The primers

Immunofluorescence studies
Immunostaining was performed as described previously 9  analyzed using the Axio Imager (Zeiss). The g-ratio was measured by dividing the diameter of an axon (without myelin) by that of the total fiber diameter (axon + myelin sheath) using ImageJ. Diameters were normalized by their perimeters.

SBF-SEM imaging and analyses
The imaging and 3D ultrastructural analyses were performed as described previously 10,11 . Briefly, sciatic nerves of 6 wild-type and 6 GlcNAc6ST-1-KO mice were removed after transcardial perfusion using PBS and 0.1M PB (pH 7.4) containing 4% PFA and 0.5% glutaraldehyde. Tissues were immersed in the same fixative overnight. The samples were post-fixed with reduced osmium, stained en bloc with thiocarbohydrazide, osmium and lead, and embedded in conductive resin. Following trimming, samples were imaged with Sigma VP or Merlin (Zeiss) equipped with 3View2XP (Gatan, Pleasanton, CA). The serial images acquired were handled and processed for segmentation and 3D reconstruction with Fiji/ImageJ (http://fiji.sc/Fiji) and Amira (FEI Visualization Science Group, Hillsboro, OR). 2B-2E were analyzed by RP-HPLC. There was one main peak in each N-glycan elution profile. The N-glycan elution profiles of peaks 6'', (7+8)i', (7+8)ii' and 9'' coincided with those of the peaks 5c', 5a', 5b' and 5c', respectively. The N-glycans were identified using known standards (structures shown in Fig. 2F).