Masakazu Hachisu

Both Isoforms of Human UDP-glucose:glycoprotein Glucosyltransferase are Enzymatically Active

Being recognized as an important constituent of the glycoprotein folding cycle, uridine diphosphate-glucose:glycoprotein glucosyltransferase (UGGT) has been a subject of intense study. Up to now, it is two isoforms, UGGT1 and 2 have been identified, which share ∼ 50% amino acid identity. UGGT1 is a well-documented enzyme which functions as a folding sensor in the endoplasmic reticulum, by the virtue of its ability to transfer a glucose residue to non-glucosylated high-mannose-type glycans of immature glycoproteins exhibiting non-native conformation. On the other hand, direct evidence to support the glucosyltransferase activity of UGGT2 has been lacking, leaving it unclear as to whether it has any function in the glycoprotein folding process. This study aimed to reveal the property of human UGGT2 by using synthetic substrates such as fluorescently labeled glycans and N-glycosylated proteins. The analysis, for the first time, revealed the glucosyltransferase activity of UGGT2, whose specificity was shown to be quite similar to UGGT1, in terms of both glycan specificity and preferential recognition of proteins having non-native conformations. Finally, Sep15 was found to form the heterodimeric complex with both isoforms of UGGT and markedly enhanced its glucosyltransferase activity.

Top‐Down Chemoenzymatic Approach to a High‐Mannose‐Type Glycan Library: Synthesis of a Common Precursor and Its Enzymatic Trimming

From the stacks: A novel method for construction of a high-mannose-type glycan library by systematic enzymatic trimming of a single synthetic Man9-based precursor was developed. Efficient chemical synthesis of the tetradecasaccharide common precursor and orthogonal enzymatic trimming to obtain all M(8-9) and G(1)M(8-9) derivatives was demonstrated. G = glucose, M = mannose.

Construction of a high-mannose-type glycan library by a renewed top-down chemo-enzymatic approach.

A comprehensive method for the construction of a high-mannose-type glycan library by systematic chemo-enzymatic trimming of a single Man9-based precursor was developed. It consists of the chemical synthesis of a non-natural tridecasaccharide precursor, the orthogonal demasking of the non-reducing ends, and trimming by glycosidases, which enabled a comprehensive synthesis of high-mannose-type glycans in their mono- or non-glucosylated forms. It employed glucose, isopropylidene, and N-acetylglucosamine groups for blocking the A-, B-, and C-arms, respectively. After systematic trimming of the precursor, thirty-seven high-mannose-type glycans were obtained. The power of the methodology was demonstrated by the enzymatic activity of human recombinant N-acetylglucosaminyltransferase-I toward M7-M3 glycans, clarifying the substrate specificity in the context of high-mannose-type glycans.

Profiling Aglycon-Recognizing Sites of UDP-glucose:glycoprotein Glucosyltransferase by Means of Squarate-Mediated Labeling

Because of its ability to selectively glucosylate misfolded glycoproteins, UDP-glucose:glycoprotein glucosyltransferase (UGGT) functions as a folding sensor in the glycoprotein quality control system in the endoplasmic reticulum (ER). The unique property of UGGT derives from its ability to transfer a glucose residue to N-glycan moieties of incompletely folded glycoproteins. We have previously discovered nonproteinic synthetic substrates of this enzyme, allowing us to conduct its high-sensitivity assay in a quantitative manner. In this study, we aimed to conduct site-selective affinity labeling of UGGT using a functionalized oligosaccharide probe to identify domain(s) responsible for recognition of the aglycon moiety of substrates. To this end, a probe 1 was designed to selectively label nucleophilic amino acid residues in the proximity of the canonical aglycon-recognizing site of human UGGT1 (HUGT1) via squaramide formation. As expected, probe 1 was able to label HUGT1 in the presence of UDP. Analysis by nano-LC-ESI/MS(n) identified a unique lysine residue (K1424) that was modified by 1. Kyte-Doolittle analysis as well as homology modeling revealed a cluster of hydrophobic amino acids that may be functional in the folding sensing mechanism of HUGT1.

Glycan specificity of a testis-specific lectin chaperone calmegin and effects of hydrophobic interactions.

BACKGROUND Testis-specific chaperone calmegin is required for the generation of normal spermatozoa. Calmegin is known to be a homologue of endoplasmic reticulum (ER) residing lectin chaperone calnexin. Although functional similarity between calnexin and calmegin has been predicted, detailed information concerned with substrate recognition by calmegin, such as glycan specificity, chaperone function and binding affinity, are obscure. METHODS In this study, biochemical properties of calmegin and calnexin were compared using synthetic glycans and glycosylated or non-glycosylated proteins as substrates. RESULTS Whereas their amino acid sequences are quite similar to each other, a certain difference in secondary structures was indicated by circular dichroism (CD) spectrum. While both of them inhibited protein heat-aggregation to a similar extent, calnexin exhibited a higher ability to facilitate protein folding. Similarly to calnexin, calmegin preferentially recognizes monoglucosylated glycans such as Glc1Man9GlcNAc2 (G1M9). While the surface hydrophobicity of calmegin was higher than that of calnexin, calnexin showed stronger binding to substrate. We reasoned that lectin activity, in addition to hydrophobic interaction, contributes to this strong affinity between calnexin and substrate. CONCLUSIONS Although their similarity in carbohydrate binding specificities is high, there seems to be some differences in the mode of substrate recognition between calmegin and calnexin. GENERAL SIGNIFICANCE Properties of calmegin as a lectin-chaperone were revealed in comparison with calnexin.

Cationic derivative of electrospun non-woven cellulose-chitosan composite fabrics for immobilization of aminoacylase-I

The chemical modification of cellulose (Cs)-chitosan (Ch)-electrospun non-woven (ESNW) composite fabrics is described. Since the as-spun Cs/Ch-composite ESNW (Cs : Ch = 4 : 6) deforms in water, an insolublilization procedure using an alkaline-ethanol solution was developed to extend the applications in an aqueous environment. The fine fiber surface was modified with bifunctional isocyanate in order to introduce the -NCO groups on the ESNW surfaces. The -NCO groups were condensed with N,N-(diethyl)ethylene diamine to generate a cationic diethylaminoethyl (DEAE)-ESNW. An enzyme, aminoacylase-I, was immobilized onto the cationic matrix under mild conditions. The immobilized enzyme was subjected to the stereo-specific recognition of Nα-acetyl-methionines. The results suggest that the chemically modified Cs/Ch-ESNW is a useful support matrix for biological catalysts.

Synthesis of peptide–cellulose conjugate mediated by a soluble cellulose derivative having β-Ala esters

A derivatization of cellulose was investigated for gaining up the solubility for subsequent conjugation reaction. N(β)-Boc-β-Ala, was introduced to the parent cellulose using carbonyl diimidazole. Degree of substitution towards the cellulose hydroxyls was 49%. Subsequent removal of Boc in trifluoroacetic acid (TFA) yielded β-Ala-cellulose·TFA salt. A protected hexapeptide, Boc-Ser(Bzl)-Gly-Tyr(Bzl)-Ser(Bzl)-Gly-Lys(Z) was synthesized via 9 steps of peptide elongation, and the C-end carboxyl groups of the peptide was coupled with the β-Ala-cellulose in a homogenous dimethyl sulfoxide solution, using 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide monohydrochloride to ensure the N(β)-selective acylation. The degree of substitution of the protected peptide towards the β-amino groups of β-Ala-cellulose (DS%(/NH(2))) was 52% for 1.0 eq.mol of the protected peptide feed, and in the presence of N-hydroxysuccinimide, DS%(/NH(2)) was increased to 61%. At 4.0 eq.mol feed, almost quantitative conjugation was observed as DS%(/NH(2))=98-99%. Deprotection of the conjugate using thioanisole-TFA resulted in complete removal of Boc, Bzl on Tyr, and Z on Lys, while a very trace amount of Bzl on Ser seemed to be left uncleaved.

Synthesis of peptide-cellulose conjugate mediated by a soluble cellulose derivative having β-Ala esters (II): conjugates with O-phospho-l-serine-containing peptides

A synthetic route is described here for novel peptide-cellulose conjugates containing O-phospho-l-serine. First, Boc-Ser(PO3Ph2) and the related dipeptides, Boc-Ser(PO3Ph2)-Asp(OBzl) and Boc-Asp(OBzl)-Ser(PO3Ph2), were synthesized by adopting the phosphoryl-protection strategy. The condensation reaction between the α-carboxyl group of the protected Boc-Ser(PO3Ph2) and the β-amino groups of β-Ala-Cellulose using isobutyl chloroformate and N-methylmorpholine yielded the product conjugate, Nβ-[Boc-Ser(PO3Ph2)]-β-Ala-Cellulose. The degree of substitution of Boc-Ser(PO3Ph2) towards the β-amino groups of β-Ala-Cellulose was estimated as DSN = 0.75 (maximum, 1.0). Similar reactions between β-Ala-Cellulose and two kinds of protected dipeptides, Boc-Asp(OBzl)-Ser(PO3Ph2) and Boc-Ser(PO3Ph2)-Asp(OBzl), gave the corresponding conjugates, and the DSN was estimated to be 0.95 and 0.69, respectively. The phenyl, benzyl, and Boc groups were removed in one-pot using the Pt2O catalyst in 50 % trifluoroacetic acid/acetic acid. The 31P-NMR and UV-Visible spectra indicated the complete deprotection without any observable elimination of the phosphorylated peptides.

Design and synthesis of peptide-cellulose conjugate molecules —Aspects from energy/steric profiles—

A multipurpose chemical modification of cellulose to yield soluble cellulose derivatives was investigated on the basis of (i) O-acylation of the parent cellulose using N-Boc-protected Gly, αAla, and βAla, and subsequent Boc-removal, (ii) N-acylation of amino acid-modified cellulose in homogenous solution reaction system using Boc-oligopeptides with multiple types of the side-chain functionalities, and (iii) N-acylation of βAla-Cellulose using 10 kinds of simple Boc-oligopeptides. As for the (i), Thermochemical computation to estimate ΔG transition was well coincided with the observed yields in Oacylations. In the case of (ii), beside the similar ΔG-schemes, a strain energy profile, which is coupled with the preferred conformer transition of the amino acid-modified cellulose, was involved to rule out the N-acylation. The symmetric anhydrate in the N-acylating agents in the (iii) are tend to be in torsions due to the β-strand-like steric stabilizations in the peptides, resulting in the highest N-acylation yields as much as 98 %.

Chain Conformational Study on Underwater Silk Proteins from Caddisfly, Stenopsyche marmorata - Implication of a Fiber-Forming Mechanism

Larval silk/cement proteins from a caddisfly, Stenopsyche marmorata, were isolated as a protein mixture of Smsp-1, 2, 3 and 4. Smsp-1 is a giant phosphorylated protein, which occupies ca. 45%-mass of the silk gland content, and composed of a long-range periodic amino acid sequences, involving 8 kinds of characteristic segments. The silk protein film was prepared and drawn in water up to 9-folds of the initial axis length, then the drawn film was subjected to polarized FT-IR and WAXD. The results implied that the Smsp-1 backbone adopts two different conformations, one of which was the β-turn-like conformers. The molecular mechanic studies were separately performed to evaluate the solid-state chain structures of the hydrophobic/Pro-rich segments 3 and 4, which are enriched in the primary sequence of Smsp-1, and the results were coincident with those from the vibration spectra and WAXD. The molecular dynamic (MD) studies were also carried out in order to estimate their preferred chain conformations in a solution state. The MD trajectory suggests that the segments 3 and 4 tend to adopt a turn-like conformation, which is a potential precursor of the β-turn-like conformers. In conclusion, the underwater silk proteins have a fiber-forming mechanism, which is substantially different from a silkworm, Bombyx mori.