Jack bean α-mannosidase: Amino acid sequencing and N-glycosylation analysis of a valuable glycomics tool - PDF

Description
Glycobiology vol. 24 no. 3 pp , 2014 doi: /glycob/cwt106 Advance Access publication on December 1, 2013 Jack bean α-mannosidase: Amino acid sequencing and N-glycosylation analysis of a valuable

Please download to get full document.

View again

of 10
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Information
Category:

Devices & Hardware

Publish on:

Views: 59 | Pages: 10

Extension: PDF | Download: 0

Share
Transcript
Glycobiology vol. 24 no. 3 pp , 2014 doi: /glycob/cwt106 Advance Access publication on December 1, 2013 Jack bean α-mannosidase: Amino acid sequencing and N-glycosylation analysis of a valuable glycomics tool B S Gnanesh Kumar 2,, Gottfried Pohlentz 3,, Mona Schulte 3, Michael Mormann 1,3, and Nadimpalli Siva Kumar 1,2 2 Protein Biochemistry and Glycobiology Laboratory, Department of Biochemistry, University of Hyderabad, Hyderabad , India and 3 Institute for Hygiene, University of Münster, Münster D-48149, Germany Received on October 7, 2013; revised on November 23, 2013; accepted on November 24, 2013 Jack bean (Canavalia ensiformis) seeds contain several biologically important proteins among which α-mannosidase (EC ) has been purified, its biochemical properties studied and widely used in glycan analysis. In the present study, we have used the purified enzyme and derived its amino acid sequence covering both the known subunits (molecular mass of 66,000 and 44,000 Da) hitherto not known in its entirety. Peptide de novo sequencing and structural elucidation of N-glycopeptides obtained either directly from proteolytic digestion or after zwitterionic hydrophilic interaction liquid chromatography solid phase extractionbased separation were performed by use of nanoelectrospray ionization quadrupole time-of-flight mass spectrometry and low-energy collision-induced dissociation experiments. De novo sequencing provided new insights into the disulfide linkage organization, intersection of subunits and complete N-glycan structures along with site specificities. The primary sequence suggests that the enzyme belongs to glycosyl hydrolase family 38 and the N-glycan sequence analysis revealed high-mannose oligosaccharides, which were found to be heterogeneous with varying number of hexoses viz, Man 8 9 GlcNAc 2 and Glc 1 Man 9 GlcNAc 2 in an evolutionarily conserved N-glycosylation site. This site with two proximal cysteines is present in all the acidic α-mannosidases reported so far in eukaryotes. Further, a truncated paucimannose type was identified to be lacking terminal two mannose, Man 1 (Xyl)GlcNAc 2 (Fuc). Keywords: α-mannosidase / de novo sequencing / jack bean / mass spectrometry / N-glycans 1 To whom correspondence should be addressed: Tel: (N.S.K.)/ (M.M.); Fax: (N.S.K.)/ (M.M.); (N.S.K.); (M.M.) B.S.G.K. and G.P. contributed equally. Introduction α-mannosidases hydrolyze terminal α-mannosidic linkages from glycan moieties of various glycoconjugates. In eukaryotes, they are involved in the processing and degradation of N-glycans by hydrolyzing the terminal α1 2-, α1 3- and α1 6-linked mannose residues from high-mannose, hybrid and complex type N-glycans present in glycoproteins (Daniel et al. 1994; Herscovics 1999). α-mannosidases are categorized into two classes based on inhibition and sequence similarity as Class I α-mannosidase of glycosyl hydrolase (GH) family 47 present in the endoplasmic reticulum (ER α-man) and the Golgi apparatus (GM I) that hydrolyze specifically α1 2-mannosidic linkages involved in maturation of N-glycans. They are specifically inhibited by kifunensine and 1-deoxymannojirimycin (Daniel et al. 1994; Herscovics 1999). Another processing α-mannosidase present in Golgi bodies (GM II) hydrolyzes α1 3- and α1 6-mannosidic linkages and shows sequence similarity with lysosomal/acidic α-mannosidase, which is involved in the degradation of N-glycans. These are classified as Class II α-mannosidases of GH family 38 and are inhibited specifically by swainsonine (Daniel et al. 1994; Herscovics 1999). Unlike GM II, lysosomal/acidic α-mannosidases have broad specificity toward cleavage for α1 2-, α1 3- and α1 6-mannosidic linkages. The absence of this lysosomal enzyme in humans and cattle leads to a genetic disorder called α-mannosidosis (Daniel et al. 1994). In plants, the physiological role of the acidic α-mannosidase is still not clear and in recent findings silencing of α-mannosidase as well as β-n-acetyl hexosaminidase in tomato and capsicum delayed fruit ripening (Meli et al. 2010; Ghosh et al. 2011). Jack bean α-mannosidase (JBM) is a commercially available exo-glycosidase, widely used as a tool for glycan analysis that exhibits activity toward several different substrate, cleaving α1 2-, α1 3- and α1 6-linked mannose residues from various glycoprotein preparations (Li 1966). It is a zinc metalloenzyme comprising two pairs of subunits of molecular mass (M r ) 66,000 and 44,000 Da, respectively (Li 1967; Snaith 1975). The enzyme is initially synthesized as 110,000 Da precursor polypeptide which undergoes posttranslational cleavage and accumulates in protein storage vacuoles or protein bodies of bean cotyledons (Faye et al. 1988). The larger subunit is glycosylated containing high-mannose type oligosaccharides that are in part terminally glucosylated and truncated paucimannose type N-glycans (comprising β1,2-xylose and α1,3-fucose within the pentasaccharide core), whereas the smaller subunit shows no glycosylation (Araki et al. 1994; Kimura et al. 1999). The Author Published by Oxford University Press. All rights reserved. For permissions, please 252 Primary structure of Jack bean α-mannosidase Similar to other members of GH family 38, the JBM is a retaining enzyme (Howard et al. 1997) and like other Class II α-mannosidase is also inhibited by swainsonine (Kang and Elbein 1983), which has undergone phase I clinical trial as an anticancer agent (Goss et al. 1997). Due to its properties similar to those of Golgi α-mannosidase II (an important therapeutic target), JBM serves as a model enzyme for structural and mechanistic studies for inhibitors (Popowycz et al. 2001; Fiaux et al. 2005). Even though JBM is commercially available and has been well characterized biochemically and widely used for sugar analysis of glycoconjugates for decades, neither the complete primary sequence nor structural details have been elucidated so far. In the present study, the primary sequence of both subunits was deduced by means of MS using the bottom up approach of analysis of peptides generated by various proteases. Furthermore, the disulfide linkages in both the subunits have been elucidated and the highly conserved N-glycosylation site bearing high-mannose oligosaccharide was identified. Results Primary sequence of JBM De novo sequencing of peptides derived from proteolytic cleavages (in-gel or in-solution) with different proteases gave rise to 98% of the protein sequence of JBM that corresponds to 959 amino acids. Each amino acid has been determined from at least two independent proteolytic peptides and the resulting sequences of the two subunits are shown in Figure 1A. A homology search of the complete sequence using Blastp demonstrated that JBM is a member of GH family 38 with N- and C-terminal domains that are conserved (Figure 1B). Disulfide linkage analysis The determination of disulfide linkages was carried out according to Mormann et al. (2008). Fragmentation of the triply charged precursor ions at m/z (Figure 2A) derived from a tryptic digest of nonreduced α-mannosidase revealed an intrapeptide disulfide link in the peptide aa , thus indicating the S S bridge C 800 C 807 (Figure 2B). Similarly, the identification of the intrapeptide disulfide bonds C 422 C 432 and C 442 C 450 resulted from collision-induced dissociation (CID) experiments on the triply charged precursor ions at m/z (chymotryptic digest, aa ) and at m/z (digest with thermolysin, aa , cf. Figure 1A) obtained from nonreduced JBM. The cysteines C 47 and C 107 obviously harbor free sulfhydryl groups as could be proved by fragmentation of corresponding peptides derived from tryptic digests of nonreduced α-mannosidase (m/z ,aa ; m/z ,aa ; cf. Figure 1A). Active site peptide analysis Fragmentation of the doubly charged precursor ion at m/z derived from a chymotryptic digest revealed the sequence QIDPFGHSAVQGYaa (Supplementary data, Figure S1), which was previously identified as being the active site of JBM (Howard et al. 1998). Similarly, collisional activation of the 4-fold and triply charged tryptic peptides with m/z and m/z , respectively, revealed the same sequence stretch confirming the presence of the aspartic acid. The only resolved crystal structure of an acidic α-mannosidase has been reported for bovine lysosomal α-mannosidase (blam) (Heikinheimo et al. 2003) and the sequence alignment of JBM showed 41% amino acid identities, so all active site amino acids involved in catalysis and substrate binding were identified by aligning JBM with blam which is depicted in Figure 3. Subunit sequence determination The sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS PAGE) of JBM (Supplementary data, Figure S2) revealed two bands (larger and smaller subunit 66,000 and 44,000 Da, respectively) when heated in the presence of dithiothreitol (DTT), and the appearance of a single band at 110,000 Da without heating, as reported earlier (Einhoff et al. 1987). The tryptic peptides obtained from the larger subunit covered amino acid sequences from the range aa 1 562, whereas the smaller subunit revealed peptides out of the region aa (total 419 amino acids). This was further confirmed by digesting the single band 110,000 Da, which showed the presence of peptides originating from both subunits. Fragmentation of doubly charged precursor ions at m/z derived from GluC digestion revealed the peptide sequence MKYNTGAGTVPE (aa 1 12 ), (Supplementary data, Figure S3) that corresponds to the N-terminal sequence of the larger JBM subunit (upon SDS PAGE). The chymotryptic peptide with amino acid sequence LSQKGETINIGPGDLKMSF (aa ) was deduced from fragmentation of precursor ions at m/z (Supplementary data, Figure S4) corresponding to N-terminus of the smaller subunit. Additionally, the tryptic peptide derived from in-gel digestion of the smaller subunit revealed the sequence GETINIGPDLK (aa ) confirming that it is indeed the N-terminus of the smaller subunit (data not shown). N-glycopeptide analysis The fragment ion spectrum of the triply charged N-glycopeptide precursor ions at m/z derived from in-solution digestion by thermolysin demonstrates the presence of a highmannose type glycan (Man 8 GlcNAc 2 ) showing sequential loss of eight hexoses (mannoses) and two N-acetylhexosamines (N-acetylglucosamine (GlcNAc)) (Figure 4A and B). The complete peptide sequence of the corresponding glycopeptide was, however, obtained only after endoglycosidase H treatment (Figure 4C) (aa ). In addition, two glycopeptides containing the same peptide moiety and the N 446 -linked glycans Man 9 GlcNAc 2 and Glc 1 Man 9 GlcNAc 2, respectively, were detected (data not shown). In another approach, N-glycopeptides derived from trypsin/ chymotrypsin double digestion followed by zwitterionic hydrophilic interaction liquid chromatography (ZIC-HILIC) enrichment give rise to intense signals for three precursor ions in the nanoesi mass spectrum (Supplementary data, Figure S5). The fragmentation of triply charged precursor ions at m/z leads to sequential loss of 9 hexoses (mannoses) and two N-acetylhexosamines (GlcNAcs) (aa ) confirming a highmannose type glycan structure. Similarly, triply charged 253 BS Gnanesh Kumar et al. Fig. 1. (A) Amino acid sequence of JBM deduced from fragment ion spectra of peptide ions derived from proteolytic digests. N-terminus of smaller subunit is shown in box. Two N-glycosylation sites with confirmed site occupancy are underlined. Dots serve as proxies for non-identified amino acids. (B) Conserved domain database result showing JBM as member of GH 38. precursor ion at m/z showed sequential elimination of 10 hexoses (most probably 1 glucose and 9 mannoses) and two N-acetylhexosamines (GlcNAcs) attached to the extended peptide sequence (aa ) resulting from a different cleavage site preferred by the proteases used (data not shown). The N-glycopeptide carrying a truncated paucimannose glycan Man(Xyl)GlcNAc 2 (Fuc) at N 312 was observed in tryptic, chymotryptic and thermolysin digests (m/z , m/z and m/z , respectively). Fragment ions attributed to losses of either a pentose (xylose) or a deoxyhexose (fucose) or both were observed. Prior to the liberation of the hexose, the pentose had to be cleaved off, since no loss with an increment of 162 Da starting from the precursor ion could be detected. This clearly demonstrates the linkage of the xylose to the mannose residue. Similarly, loss of the proximal N-acetylhexosamine (GlcNAc) leading to Y 0 arises exclusively subsequent to cleavage of the fucose, thus proving the expected core fucosylation. Almost complete b- and y- type ion series fragments derived from the peptide backbone allowed to deduce the amino acid sequence of the glycopeptide under inspection (Figure 5). Additionally, peptides originating in the smaller subunit comprising a potential N-glycosylation site (N 867 ) which did not harbor any N-glycan moiety were identified in trypsin and chemical in-solution digests by fragmenting the precursor ions m/z and m/z , respectively. No evidence for occupation of this glycosylation site was found. Multiple sequence alignment of JBM sequence with various acidic α-mannosidases is represented in Supplementary data, Figure S6. 254 Primary structure of Jack bean α-mannosidase Fig. 2. (A) NanoESI fragment ion spectrum obtained from the triply charged peptide precursor ions at m/z derived from a tryptic JBM digest. (B) Corresponding fragmentation scheme showing the intrapeptide disulfide linkage between C 800 and C 807. Discussion α-mannosidases are very important and are of great significance by virtue of their biological function in the processing and degradation of glycoproteins in eukaryotes (Herscovics 1999), and therefore, a large number of these enzymes have been well characterized from various organisms as two distinct classes of enzymes. Compared with reports available on mammalian and other animal species, only few plant α-mannosidases of both the classes are characterized biochemically and to lesser extent at molecular level. Among these the JBM from a legume is biochemically well characterized and widely used, but no reports on the detailed molecular characterization of the enzyme. In the present study, we describe the primary sequence, subunit intersection, disulfide linkage arrangements as well as N-glycosylation of JBM. The overall amino acids identified in JBM are 981 of which 959 amino acids were determined by manual interpretation of fragment ion spectra of peptides and glycopeptides derived from various proteases. The larger subunit contains 562 amino acids (22 amino acids were not identified) and the smaller one has 419 amino acids. Intrapeptide disulfide linkages are present in both the larger and smaller subunits (Figure 1A). The absence of disulfide bridges between the two subunits suggests that the subunit interaction is non-covalent. Consistent with earlier findings (Howard et al. 1998), we also identified peptide sequence QIDPFGHSAVQGY in tryptic and chymotryptic digestions, which contains aspartic acid (D 145 ), which has been demonstrated as catalytic nucleophile in JBM (Howard et al. 1998) and is characteristic of GH 38 α-mannosidases. A similar sequence stretch containing the catalytic aspartic acid was identified in Dolichos lablab α-mannosidase (Gnanesh Kumar et al. 2013). All active site amino acids in JBM were identified by aligning with blam. The amino acids in JBM are as follows, H 23,D 25,W 28,Y 35,D 145,R 170,Y 210,D 268,F 269,Y 325,W 333, H 385,H 386,D 387,T 392 and Y 625 (Figure 3). The CID spectrum of the peptide precursor ions comprising H 335,H 336,D 337, and T 342 together with the corresponding fragmentation scheme is depicted in Supplementary data, Figure S7. It is interesting to note that active site amino acid Y 625 is located in the smaller subunit (upon SDS PAGE), suggesting that the integrity of both subunits is essential for the enzyme to be active. 255 BS Gnanesh Kumar et al. Fig. 3. Alignment of selected regions of blam with JBM showing active sites region (gray shawdowed). The active site amino acids are in bold and underlined. An earlier study of the N-terminal sequence of the two SDS PAGE-separated JBM subunits (Burrows and Rastall 1998) yielded the consistent sequence MKYNTGAGTVPEQLN for the larger subunit, which was also found in our study MKYNTGAGTVPEQLN (aa 1 15 ). But the N-terminus analysis of the smaller subunit revealed the sequences SXTINIG; XQETINIGPDLKMSF, and we obtained full sequence stretch LSQKGETINIGPGDLKMSF (aa ) at least in two different digests, including in-gel tryptic digestion of the smaller subunit. This suggests that our results are consistent with the earlier findings and this sequence stretch plays an important role in the subunit intersection of the protein. However, it is still uncertain whether the native protein is a single polypeptide and heating prior to SDS PAGE causes the formation of two subunits or the subunits are hold each other by strong non-covalent interactions. To understand clearly more about this aspect, conformation studies in native form of the enzyme need to be carried out. The presence of identical subunits (homo dimer) has been reported for α-mannosidase from two closely related species Phaseolus vulgaris ( 110 kda) (Paus 1977) and D. lablab ( 116 kda) (Gnanesh Kumar et al. 2013). However, the tomato enzyme is a hetero tetramer with pair of 70 and 47 kda subunits (Hossain et al. 2010). The N-terminal sequence of smaller subunit identified in tomato α-mannosidase is different from that of JBM and in both active site amino acid tyrosine is located in the smaller subunit (Y 625 for JBM and Y 661 for tomato α-mannosidase) (Supplementary data, Figure S6). In mammals and other lower eukaryotes, the proteolytic processing of enzyme polypeptide into subunits is found to be consistent and cleavage occurs around the same sequence stretch (Tollersrud et al. 1997; Vandersall-Nairn et al. 1998). The use of non-specific proteases such as the thermolysin as well as mixture of trypsin/chymotrypsin followed by enrichment by ZIC-HILIC solid phase extraction (SPE) has proved to be very advantageous for the elucidation of the site-specific N-glycosylation of glycoproteins (Neue et al. 2011; Selman et al. 2011). In view of these recent observations, we have used a similar strategy in the present study for N-glycan analysis. Three high-mannose type N-glycan structure viz. Man 8 GlcNAc 2, Man 9 GlcNAc 2 and Glc 1 Man 9 GlcNAc 2 were observed for JBM (Supplementary data, Figure S5). Similar structures along with the corresponding peptide, QSADQSX 1 APASAFSQSHLFX 2 ISYSPPTESSIPPDK, in which X 2 is glycosylated asparagine residue, were reported by Kimura et al. (1999). In contrast to this sequence where two proximal serine residues were observed (underlined), the peptide sequence derived from N-glycopeptide of thermolysin following Endo-β-N acetylglucosaminidase H (Endo H) digestion revealed the sequence FSQCHLFNISYCPPTESSLPDDKS (aa ) (Figure 4B and C) confirming the presence of two proximal cysteines (C 442,C 450 ). The formation of a disulfide linkage around the N-glycosylation site bearing high-mannose type is an interesting structural feature which might be crucial for protecting N-glycan due to the fact that the enzyme itself is carrying its own substrate. It might also prevent the processing of high-mannose N-glycans by other enzymes present in endoplasmic reticulum and Golgi apparatus during maturation as well as accessibility to itself on these structures during transport and storage. The latter finding is corroborated by results reported by Kimura et al. (1999) who showed that treatment of JBM with reducing agents prior to Endo H-induced cleavage of the high-mannose oligosaccharide from the intact protein is required. Furthermore, a drastically reduced enzymatic activity was found aft
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks