Jianfei Li1 & Jiangye Zhang1 & Bi Lai1 & Ying Zhao1 &

Qinfan Li1

Received: 13 May 2015 /Accepted: 13 August 2015 /Published online: 26 August 2015# Springer Science+Business Media New York 2015

Abstract Golgi α-mannosidase II (GMII), a key glycosyl hydrolase in the N-linked glyco-sylation pathway, has been demonstrated to be closely associated with the genesis anddevelopment of cancer. In this study, we cloned cDNA-encoding Capra hircus GMII(chGMII) and expressed it in Pichia pastoris expression system. The chGMII cDNA containsan open reading frame of 3432 bp encoding a polypeptide of 1144 amino acids. The deducedmolecular mass and pI of chGMII was 130.5 kDa and 8.04, respectively. The geneexpression profile analysis showed GMII was the highest expressed gene in the spleen. Therecombinant chGMII showed maximum activity at pH 5.4 and 42 °C and was activated byFe2+, Zn2+, Ca2+, and Mn2+ and strongly inhibited by Co2+, Cu2+, and EDTA. By homologymodeling and molecular docking, we obtained the predicted 3D structure of chGMII and theprobable binding modes of chGMII-GnMan5Gn, chGMII-SW. A small cavity containingTyr355 and zinc ion fixed by residues Asp290, His176, Asp178, and His570 was identifiedas the active center of chGMII. These results not only provide a clue for clarifying the catalyticmechanism of chGMII but also lay a theoretical foundation for subsequent investigations in thefield of anticancer therapy for mammals.

Keywords Golgiα-mannosidase II .Capra hircus .Pichia pastoris expression . Active site .

Anticancer

Introduction

The invasion and metastasis of malignant tumor cells are essentially caused by thedisorder of interaction between the cells and their surrounding environment, including

Appl Biochem Biotechnol (2015) 177:1241–1251DOI 10.1007/s12010-015-1810-0

Electronic supplementary material The online version of this article (doi:10.1007/s12010-015-1810-0)contains supplementary material, which is available to authorized users.

* Qinfan Liliqinfan@nwsuaf.edu.cn

1 College of Veterinary Medicine, Northwest A&F University, Xian, China

Cloning, Expression, and Characterization of Capra hircusGolgi α-Mannosidase II

adjacent cells and extracellular bioactive molecules. The interactions are mediated bycell-surface glycoproteins, oligosaccharide of which participates in several biologicalrecognition, affecting cell survival or apoptosis, proliferation and differentiation,migration and invasion, etc. Aberrant glycosylation of glycoproteins is a typicalmolecular change of malignant transformations. Glycosyltransferase N-acetylglucosaminyl transferase V (GlcNAc-TV), often overexpressed during cancer,promotes the introduction of oligosaccharide chains on integrins and cadherins oradhesion receptors, which facilitate the focal adhesion turnover, cell migration, andtumor metastasis. Inhibition of GlcNAc-TV has been demonstrated useful for thetreatment of malignancies. However, it was difficult to design and synthesize cell-permeable inhibitors of glycosyltransferases [1]. Therefore, more efforts have beengiven to design inhibitors of glycosidase that act earlier in the biosynthesis of N-glycans in order to prevent the formation of the biosynthetic pathway primer ofGlcNAc-TV. Many research have been focused on inhibiting Golgi α-mannosidaseII (GMII), which belongs to glycosyl hydrolases family 38, mainly exit in trans andmedial Golgi cisternae [2, 3], and specifically recognize and hydrolyze the α1,6- andα1,3-linked mannose residues of GlcNAcMan5GlcNAc2(GnMan5Gn2) [4]. In clinicaltrials, swainsonine (SW), a natural inhibitor of GMII, has been shown to reducecertain tumors and hematological dysfunctions [5, 6]. Unfortunately, this potentinhibitor often co-inhibits lysosomal α-mannosidases (LM), inducing symptoms sim-ilar to lysosomal storage disease α-mannosidosis. Hence, it is crucial to find specificinhibitors of GMII with anticancer activity.

Understanding the structural basis for the catalytic and inhibition process of GMIImight provide unique opportunities for designing the selective inhibitors. Drosophilamelanogaster GMII (dGMII) has been used as a model for structural study [4]. On thebasis of its structure obtained by X-ray crystallography, numerous studies have beencarried out to reveal the catalytic mechanism and main active sites binding with itssubstrate and inhibitors [4, 7, 8]. Moreover, some selective inhibitors of GMII havebeen designed and synthesized based on the speculation, such as α-D-mannosederivatives [9], 5-substituted swainsonine [10], and sulfonium-ion analogues of di-epi-swainsonine [11].

Besides, the GMII cDNA and gene have been isolated from murine [12, 13], human [14],Arabidopsis thaliana [15], D. melanogaster [16], Spodoptera frugiperda [17], andCaenorhabditis elegans [18] and successfully expressed in COS cells [13], CHOP cell[19], Sf9 cell [17], and Drosophila cells [4]. Furthermore, the characterization of GMII hasbeen carried out [13, 19]. Although GMII from various organisms have been reported, thereis no detailed report on GMII of any mammalian species, which have high homology withhuman.

This study focused on a common mammal, goat. With the objective to obtain thestructural and functional properties of Capra hircus GMII (chGMII), we cloned cDNA-encoding chGMII and expressed it using Pichia pastoris expression system. Using thetechnology of quantitative real-time PCR, we obtained chGMII gene expression profile.Furthermore, enzymatic characterization was carried out by various enzyme assays. The 3Dstructure and the binding sites of chGMII substrate and chGMII inhibitor were predictedwith the help of homology modeling and molecular docking to find out the active site andpotential mutant site for selective inhibition. These make preparations for follow-up studiesof chGMII.

1242 Appl Biochem Biotechnol (2015) 177:1241–1251

Materials and Methods

RNA Isolation and Reverse Transcription PCR

Fresh tissue samples of the heart, liver, spleen, lung, kidney, ovary, and intestine were collectedfrom a 7-month-old nanny goat, supplied by Laboratory Animal Center of Northwest A&FUniversity (Shaanxi, China). All samples were immediately frozen with liquid nitrogen andkept at −80 °C until they were used. Total RNA was extracted using RNAiso Plus (Takara,Japan) and quantified by determining its absorbance at OD260/OD280. First-strand cDNAwassynthesized from 2 μg of the isolated RNA using PrimeScript™ II 1st strand cDNA SynthesisKit (TaKaRa). Based on the mRNA sequences of Ovis aries GMII (Genebank ID:XM_004009111) and Bos taurus GMII (Genebank ID: NM_001205886), a pair of primers(F 5′-ATGAAGTTGAGCCGGCAGTT-3′ and R 5′-GTCCTCACAGTAGCCAGAAGC-3′)were designed to target the chGMII cDNA. The amplification was performed usingTransStart FastPfu DNA Polymerase (TransGen, Beijing). The PCR cycling conditions wereas follows: 95 °C for 1 min, 35 cycles of amplification (95 °C for 20 s, 56 °C for 20 s, and72 °C for 110 s), and 72 °C for 5 min. The target PCR products were purified using UniversalDNA Purification Kit (TianGen, Beijing) and cloned into pEASY-Blunt Cloning Vector(TransGen) using E.coli DH5α as a host bacterium. The recombinant chGMII-Bluntplasmid was sequenced by Genscript Biotech Co., Ltd.

Sequence Analysis

The open reading frame (ORF) of chGMII was analyzed by ORF Finder; the deducedamino acid sequence was determined with the Expert Protein Analysis System; themolecular weight and theoretical isoelectric point was analyzed using ProtParam tool; theconserved domains were searched in the NCBI Conserved Domains program; the trans-membrane domain was predicted by the program TMHMM Server v. 2.0; the signalpeptide sequence was predicted by SignalP 4.1 Server [20]. The protein secondary structureand N-linked glycosylation site prediction was accomplished with PredictProtein. Thededuced amino acid sequence of chGMII was used as Blastp input to explore theevolutionary relationship.

Real-Time qPCR

Quantitative PCR (qPCR) is a method for rapid and reliable quantification of mRNA tran-scription. Total RNA from various tissues was extracted and reverse transcribed as describedabove. A pair of specific primers, F 5′-GATGGTGGAGTTTGGAAGCA-3′ and R 5′-CAACCTGGGTCATTATGGGAG-3′, was used to amplify a segment of chGMII 110 bp. Aconstitutive expression gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was usedfor normalization, a fragment of which was amplified by two primers F 5′-GAGAAGGCTGGGGCTCACTT-3′ and R 5′-GTTCACGCCCATCACAAACA-3′. qPCRwas performed on Bio-Rad iQ5 System with SYBR Premix Ex Taq™ kit according to themanual (Takara). The relative expression level of chGMII in different tissues were calculatedusing the 2−△△Ct method and subject to a one-way analysis of variance (one-way ANOVA)using SPSS 13.0 software [21]. The results were presented as mean ± standard error, anddifferences were considered significant at P < 0.05.

Appl Biochem Biotechnol (2015) 177:1241–1251 1243

Construction of Expression Vector

With recombinant chGMII-Blunt plasmid as template, the coding regions of chGMII genewere obtained with a pair of primers, F 5′-TCCCCGCGGATGAAGTTGAGCCGGCAGTT-3′and R 5′-ATAAGAATGCGGCCGCGGTGGATTCGGAATGCGCTG-3′. The SacII and NotIrestriction sites (underlined) were respectively designed into the forward and reverse primers,permitting the directional cloning of the amplified DNA in frame with the α-factor leadersequence in the pPICZαA expression vector. A His6-tag sequence of pPICZαAwas fused withthe open reading frame of the chGMII gene at the C-terminal to facilitate the detection andpurification of the recombinant protein. The amplified fragment and pPICZαAwere digestedby SacII/NotI (Takara) at 37 °C for 3 h. Afterwards, linearized insert and vector were ligatedtogether to construct the recombinant expression plasmid pPICZαA-chGMII. In order to checkthe correct sequence, this plasmid was introduced into E. coli DH5α and sequenced.

Transformation and Expression of chGMII in P. pastoris X-33

pPICZαA-chGMII plasmid was linearized by Pme I and then introduced into the Pichia hoststrain, X-33, with an electroporation method described by the manufacturer (Invitrogen easyselect Pichia Expression Kit). pPICZαA plasmid without an insert was conducted as negativecontrol. The transformants were coated on YPDS agar medium containing Zeocin™(200 μg/ml) for screening [22, 23]. After being verified by colony PCR, positive transformantwas streak-inoculated in another YPDS (300 μg/ml of Zeocin™) for growing, a single colonyfrom which was inoculated in 15-ml BMGY medium at 30 °C in a shaking incubator(220 rpm) for about 10 h. The cells (OD600 = 4.0–6.0) were harvested by centrifugation at5000g for 5 min and resuspended in conical flask with 10 ml of BMMY to induce expressionat 28 °C for 5 days. Absolute methanol was added to a final concentration of 1 % every 24 h tomaintain the induction.

SDS-PAGE, Purification, and Western-Blotting Analysis

After methanol induction, the culture medium was segregated from yeast cells by centrifuga-tion at 4000g for 10 min. All of the 10-ml culture medium was concentrated by lyophilizationusing a freeze dryer and freezing bath [24] and reconstituted in 100-μl equilibration buffer(300 mM NaCl, 50 mM NaH2PO4, 10 mM Trisbase, 10 mM imidazole, pH 8.0). Yeast cellswere resuspended in the same equilibration buffer (1 g/ml) for ultrasonic decomposition(400 W, 3 s, 3 s for 40 min) and then centrifugated at 8000g for 5 min. Concentrated culturemedium and supernatant of cell lysate were collected for SDS-PAGE. Target protein waspurified by ProteinIso Ni-NTA Resin (TransGen) and identified using western blot with Anti-His Mouse Monoclonal Antibody and Anti-Mouse IgG HRP Conjugate (TransGen).

Enzyme Assay

Enzyme activity was measured by the chromogenic substrate p-nitrophenyl α-D-mannopyranoside (pNP-α-Man, Sigma-Aldrich). The mixture, containing 30-μl sample,50-μl MES buffer (80 mM, pH 5.75), and 20-μl pNP-α-Man (4 mM), was incubated in 96-well microtiter plates for 90 min at 37 °C [19]. Substrate was replaced by isopycnic H2O in thecontrol group. Reaction was stopped by adding 100 μl of 500 mM sodium carbonate and the

1244 Appl Biochem Biotechnol (2015) 177:1241–1251

absorbance measured at 405 nm using a plate reader. All reactions and controls were carriedout in triplicate and the results averaged. One unit of enzyme activity was defined as theamount of enzyme releasing 1 nmol of p-nitrophenol per second at 37 °C [25].

Effects of pH, Temperature, and Metal Ions on the Activity of chGMII

For optimum pH, chGMII were incubated with pNP-α-Man at 37 °C for 60 min in the 40 mMMES buffers with various pH levels (3.0–9.5). The effects of temperature were also examined atvarious temperatures (22–75 °C) followed by standard assay method described in the earliersection. The enzymatic activities were calculated as percent (%) relative to the highest sampleobserved in the assay. The effects of metal ions on the activity of chGMII were investigated usingZn2+, Cu2+, Mg2+, Fe2+, Ca2+, Mn2+, Fe3+, Co2+, Al3+, Cr3+, and EDTA at concentration of 2 mM,respectively. The relative activities were calculated as percent (%) relative to the negative control.

Homology Modeling and Docking Analysis

3D structure of chGMII was constructed by homology modeling using a SWISS-MODELRepository and then assessed by PROCHECK [26]. In order to explore the potential active siteof chGMII, molecular docking was performed to investigate the binding mode of compoundsSW and GnMan5Gn to chGMII using AutoDock Vina 1.1.2 [27]. AutoDock Tools 1.5.6package (http://mgltools.scripps.edu) was employed to generate the docking input files.Ligand structure was prepared for docking by merging nonpolar hydrogen atoms anddefining rotatable bonds. The search grid of chGMII was identified as center_x 30.487,center_y 66.731, and center_z 8.954 with dimensions size_x 15, size_y 15, and size_z 15 fordMNJ and SW, while for GnMan5Gn, the dimensions size_x, y, z were 20.25, 18.75, and 18.75, respectively. In order to increase the docking accuracy, the value of exhaustiveness was setto 20. For Vina docking, the default parameters were used as described in the AutoDock Vinamanual unless otherwise specified. The top-ranked pose as judged by the Vina docking scorewas subject to visually analyze using PyMOL software (http://www.pymol.org/).

Results

Cloning and Sequence Analysis of chGMII cDNA

The sequence analysis indicated that the fragment contained an ORF of 3432 bp encoding apolypeptide of 1144 amino acid residues. The gene sequence has been deposited in GenBankdatabase under accession number KF717587. The deduced molecular mass and pI of chGMIIwas 130.5 kDa and 8.04, respectively. There are two conserved domains (GH38_Man2A1domain and α-mannosidase middle domain) in the amino acid sequence of chGMII.GH38_Man2A1 domain was a subfamily represented by GMII, which confirmed chGMII isa kind of GMII. No signal peptide and only one transmembrane domain (7–26 aa) weredetected in the amino acid sequence. In the result of Blastp, the protein sequence of chGMII ishighly similar to that of B. taurus GMII (NP_001192815, 96 %), Myotis davidii GMII(ELK37318, 83 %), Homo sapiens GMII (NP_002363, 82 %), Macaca fascicularis GMII(EHH54438, 82 %), etc. Most importantly, the identity between H. sapiens and C. hircus is82 %, while that between H. sapiens and D. melanogaster is only 42 %.

Appl Biochem Biotechnol (2015) 177:1241–1251 1245

Expression Analysis of chGMII mRNA in Various Tissues

The results of gene expression profiling reveal that chGMII gene is generally expressed invarious tissues but at different levels. The greatest chGMII gene mRNA abundance is found inthe spleen, followed by the ovary, liver, lung, and intestine. In contrast, chGMII transcript wasless abundant in the kidney and heart (Fig. 1a).

Expression and Purification of chGMII

To investigate enzyme activity, the recombinant chGMII was over-expressed in P. pastoris X-33 and purified by ProteinIso Ni-NTA Resin. The expression result was analyzed by SDS-PAGE. The protein with the target molecular mass of 130 kDa was more abundant in celllysate than in culture medium, while it was not detected in the negative control(Supplementary file 1). The results manifested that chGMII had been expressed but not secreted.Purified from supernatant of the cell lysate, the recombinant chGMII yielded a single bandwith molecular weight of 130 kDa observed in SDS-PAGE (Fig. 1b). In the result of westernblotting, the bands of unpurified and purified chGMII appeared clearly in the same position of130 kDa, which confirmed the expression of recombinant chGMII as well as the highspecificity between His-chGMII and anti-His antibody, while the same band was not displayedin negative control expressing pPICZαA vector (Fig. 1c).

Enzymatic Characteristics of chGMII

The purified chGMII was analyzed for enzyme activity using pNP-α-Man as the substrate. Theabsorbances of the sample and control group were measured to be 0.5005 and 0.03, respec-tively. Calculated from the standard curve of p-nitrophenol (y = 0.0039x + 0.0270,R2 = 0.9932), the concentrations p-nitrophenol that they release were respectively 121 and0.77 μM. The enzyme activity of chGMII was computative 149.38 U. The recombinantchGMII showed optimum activity at pH 5.4 and temperature around 42 °C for substratepNP-α-Man (Fig. 2a, b). The metal ion screen displayed that chGMII was activated by Fe2+,Zn2+, Ca2+, and Mn2+ and strongly inhibited by Co2+, Cu2+, and EDTA; in comparison, Mg2+,Al3+, or Cr3+ at concentrations up to 2 mM had no effect on the activity (Fig. 2c).

Fig. 1 Expression characteristics of chGMII. a Relative expression level of chGMII in different tissues. Theliver was regarded as the control, the error bars represent the mean ± SE values (n = 3). Significant differencesare indicated with asterisks: *P<0.1; **P<0.05; ***P<0.01. b SDS-PAGE analysis of recombinant chGMIIexpressed in Pichia X-33. Lanes M, protein standard; 1, unpurified supernatant of cell lysate; 2, purified chGMII.c Western blotting analysis of recombinant chGMII. Lanes 1, unpurified supernatant of cell lysate; 2, purifiedchGMII with molecular weight of 130 kDa (arrow) ; 3, negative control expressing pPICZαA vector

1246 Appl Biochem Biotechnol (2015) 177:1241–1251

Structure of chGMII

3D structure of chGMII was generated with SWISS-MODEL Repository (Fig. 3). An α-mannosidase complex (Protein Data Bank (PDB): 2alw) was searched and screened from PDBdatabase as the template of chGMII, which possesses a higher resolution (2.0 Å) and higheridentity with chGMII (48.07 %) than other templates. Evaluated by PROCHECK, the qualityof the modeled structure was revealed with a ramachandran plot. The modeled structure ofchGMII had 99.4 % of all its residues in conformationally permitted region, and only 0.6 %residues were in the disallowed region. The results implied the quality of the model wasfavorable. The predicted secondary structure of chGMII, 23.5 % of α-helix, 22.1 % of β-strand, and 54.4 % of loop, was embodied in the 3D structure. A large groove centered arounda zinc ion contains the active site of chGMII.

Potential Active Site of chGMII

To further explore the active site of chGMII, GnMan5Gn and SW were respectively dockedwith the enzyme. As shown in Fig. 4a, b, GnMan5Gn binds in a large groove surrounded bythe aromatic side Trp505, Tyr809, Phe292, Trp181, His570, His176, Phe428, Tyr381, andTyr353. The zinc ion, located in a small cavity of the groove, is chelated by residues Asp178,His176, Asp290, and His570, forming a pentavalent geometry (T5-square-based pyramidal)(Fig. 4c). The α1,6-linked mannose (M5) is tightly bound in the cavity, 4-hydroxyl and 5-hydroxymethyl of the glycosyl coordinate with zinc ion by two hydrogen bonds, while the

Fig. 2 Effects of pH, temperature, and metal ion on the activity of chGMII. a The optimal pH was assayed withpNP-α-Man at 37 °C for 60 min in the 40 mM MES buffers with various pH levels (3.0–9.5). b The optimumtemperature was assayed in 40 mM MES buffer (pH 5.75) at various temperatures (22–75 °C). c The effects ofmetal ions on the activity of recombinant chGMII were investigated using Zn2+, Cu2+, Mg2+, Fe2+, Ca2+, Mn2+,Fe3+, Co2+, Al3+, Cr3+, and EDTA at a concentration of 2 mM, and the control group was added with isopycnicH2O; the error bars represent the mean ± SE values of three independent experiments

Fig. 3 3D structure of chGMII.The active site zinc ion is showedin gray, and α-helix, β-strand, andloop are respectively colored inred, yellow, and green

Appl Biochem Biotechnol (2015) 177:1241–1251 1247

hydroxyls at positions 3 and 4 hydrogen-bond to Tyr355. Additional hydrophobic bonds areobserved with Asp290, His176, and His570, stabilizing the M5 bound in the cavity. The α1,3-linked mannose (M4) occupies a fairly open site; with no stacking contribution to binding,only three loose hydrogen bonds bound it with Lys192 and Tyr381. The required N-acetylglucosamine (G3) is stably bound in another tight pocket. It forms strong stackinginteractions with Tyr381 and multiply coordinates with two residues Arg375 and Gly382 byfive hydrogen bonds. In comparison, the other N-acetylglucosamine (G2) is not located in anypocket, just forms two hydrogen bonds with residue Arg314 and Tyr355. SW occupies thecavity identical to that of M5 and binds to it in similar manners (Fig. 4d). The binding of theinhibitor involve a large contribution of hydrophobic interactions involving aromatic residuesTrp505, Tyr809, Phe292, Trp181, His570, His176, and Tyr353. Besides, SW is stabilized bythe hydrogen bonds with zinc ion, Tyr809, Trp181, and Tyr355.

Discussion

Previous work has confirmed that GMII is involved in the cancer process. The unusualcomplex carbohydrate structures on the cell surface caused by GMII are associated withdisease progression, metastasis, and poor clinical outcome in breast, colon, and skin cancers.

Fig. 4 Molecular docking analysis of chGMII. aA surface model of chGMII involving its substrate GnMan5Gn.Glycosyls M5, M4, G3, and G2 were located in different pockets. b The molecular docking conformations ofchGMII-GnMan5Gn. c The T5-square-based pyramidal in chGMII-GnMan5Gn. d The molecular dockingconformations of chGMII-SW. Amino acid residues involved in catalytic process are revealed in different colors;a single zinc ion is presented in gray, hydrogen bonds are depicted as yellow dashed lines, and the bond lengthsare labeled near to them

1248 Appl Biochem Biotechnol (2015) 177:1241–1251

Indeed, inhibition of GMII resulted in reduced tumor growth and metastasis. SW, a potentinhibitor of GMII, has therapeutic value because it inhibits metastasis and improves clinicaloutcome in cancers; however, its co-inhibition of LM leads to lysosomal storage disease.Understanding of the structure of GMII and the atomic basis of its catalytic mechanism and itsinhibition would be a powerful approach to novel glycosylation inhibitors as anticancer agents.However, only the structure and mechanism of dGMII was reported, and little research hasbeen conducted concerning mammalian or human GMII to date. The evolutionary relationshiprevealed human shared higher homology with C. hircus than with D. melanogaster. It thuscould make more sense for human and animal anticancer therapy to research the chGMII ratherthan dGMII. However, mammalian GMII have been proved difficult to isolate and purify inquantities suitable for structural studies [4]. Protein could be obtained by numerous eukaryoticexpression systems, such as cells and yeast. GMII have been successfully expressed in COScells, CHOP cell, Sf9 cell, and Drosophila cells. Nevertheless, few reports were relevant to itsexpression in yeast.

Yeast expression system enjoy the properties of posttranslational modification and externalsecretion, making the expression products more identical to original protein in structure andfunction as well as more convenient to be purified. As shown in the results, chGMII wassuccessfully expressed in P. pastoris X-33 but was not secreted as expected. With a secretionsignal α-factor in the expression construct, the recombinant protein is generally secreted intothe media. It is difficult to pinpoint a definite reason for this phenomenon, but it could be dueto the presence of latent proteinase which proteolytically cleaves the kex2 cleavage site of theα-factor secretion signal sequence after the expression of chGMII in Pichia cells [23]. Besides,if the kex2 cleavage site sequence had mutations, it might not be recognized and hydrolyzedby kex2 protease located at Pichia cells’ membrane. In addition, without signal peptide in theamino acid sequence, chGMII was not able to secrete by itself; the transmembrane domain (7–26 aa) make the secretion more difficult.

Despite the distant evolutionary relationship, chGMII exhibits many conserved featuressimilar to dGMII. For example, both chGMII and dGMII are Zn2+-dependent enzymes.chGMII displayed optimum activity at pH 5.4 and 42 °C, which is in agreement withdGMII and other GMIIs; it was strongly inhibited by Cu2+, which has also been previouslyverified to inhibit GMII of insect, Drosophila, and mouse [13, 19, 28–30]. Additionally, thesimilarity between chGMII and dGMII is not only in biological function but also in structure.In the modeled 3D structure of chGMII, a large α, β domain contains the active site, centeredaround a tightly associated zinc ion. Such a domain is highly conserved in the crystal structureof dGMII. The T5-square-based pyramidal geometry, involving a zinc ion and residuesAsp178, His176, Asp290, and His570, corresponds to that of dGMII (with a similar pyramidalgeometry involving residues Asp92, His90, Asp204, and His471) [4]. We can deduce thepotential active site of chGMII according to that of dGMII.

Comparing the two docking models, the aromatic-rich binding pocket surrounded by thearomatic side Trp505, Tyr809, Trp181, His570, His176, and Tyr355 is conjectured to be theposition of the hydrolysis reaction. Both M5 and SW coordinate with the zinc ion, Tyr355, andAsp290, revealing the active center is the small cavity containing Tyr355 and T5-square-basedpyramidal geometry. Neither the pocket occupied byM4 nor G3 linkedwith SWbut the necessityof them was clear when considering their multiple hydrogen bonds with glycosyls. According tothe active site of dGMII, the two pockets occupied byM4 andG3 are speculated to be the holdingsite and anchor site of chGMII, respectively. G3 have been reported to assist in binding andorienting the substrate for the hydrolysis reaction as well as increasing the local concentration of

Appl Biochem Biotechnol (2015) 177:1241–1251 1249

the linkages to be cleaved by GMII [31]. Consequently, research on G3 and the key residues ofthe anchor site will open up a path for specific combination and inhibition of chGMII.

Conclusions

In this study, we cloned chGMII cDNA and expressed it using a P. pastoris expression system.With the research on its structural and biological properties, we discovered that there are manysimilarities between chGMII and dGMII. Comparing with dGMII, we speculated someresidues of chGMII might participate in catalytic process. According to the assumption, furtherstudy of the enzyme is being carried out to determine the precise roles of the catalytic residues,to seek a new way to specifically inhibit GMII specifically, laying a molecular foundation fortherapy against cancer in humans and mammals.

Acknowledgments This work was supported by Shaanxi Province Program for Science and TechnologyResearch Development Plan (2014k01-17-02) of China and the central university basic scientific researchoperation cost special fund subsidizes (ZD2012009).

References

1. Zhong, W., Kuntz, D. A., Ember, B., Singh, H., Moremen, K. W., Rose, D. R., & Boons, G. J. (2008).Probing the substrate specificity of Golgi alpha-mannosidase II by use of synthetic oligosaccharides and acatalytic nucleophile mutant. Journal of the American Chemical Society, 130, 8975–8983.

2. Feizi, T. (1985). Demonstration by monoclonal antibodies that carbohydrate structures of glycoproteins andglycolipids are onco-developmental antigens. Nature, 314, 53–57.

3. M. Misago, Y. F. Liao, S. Kudo, S. Eto, M. G. Mattei, K. W. Moremen & M. N. Fukuda (1995). Molecularcloning and expression of cDNAs encoding human alpha-mannosidase II and a previously unrecognizedalpha-mannosidase IIx isozyme. Proc. Natl. Acad. Sci. U. S. A, 92, 11766–11770.

4. van den Elsen, J. M., Kuntz, D. A., & Rose, D. R. (2001). Structure of Golgi alpha-mannosidase II: a targetfor inhibition of growth and metastasis of cancer cells. The EMBO Journal, 20, 3008–3017.

5. Dennis, J. W., & Laferte, S. (1985). Recognition of asparagine-linked oligosaccharides on murine tumor cellsby natural killer cells. Cancer Research, 45, 6034–6040.

6. Kiyohara, T., Dennis, J. W., & Roder, J. C. (1987). Double restriction in NK cell recognition is linked totransmethylation and can be triggered by asparagine-linked oligosaccharides on tumor cells. CellularImmunology, 106, 223–233.

7. Rose, D. R. (2012). Structure, mechanism and inhibition of Golgi alpha-mannosidase II. Current Opinion inStructural Biology, 22, 558–562.

8. Thompson, A. J., Williams, R. J., Hakki, Z., Alonzi, D. S., Wennekes, T., Gloster, T. M., Songsrirote, K.,Thomas-Oates, J. E., Wrodnigg, T. M., Spreitz, J., Stutz, A. E., Butters, T. D., Williams, S. J., &Davies, G. J. (2012). Structural and mechanistic insight into N-glycan processing by endo-alpha-mannosidase. Proceedings of the National Academy of Sciences of the United States of America,109, 781–786.

9. Polakova, M., Sestak, S., Lattova, E., Petrus, L., Mucha, J., Tvaroska, I., & Kona, J. (2011). Alpha-D-mannose derivatives as models designed for selective inhibition of Golgi alpha-mannosidase II. EuropeanJournal of Medicinal Chemistry, 46, 944–952.

10. Kuntz, D. A., Nakayama, S., Shea, K., Hori, H., Uto, Y., Nagasawa, H., & Rose, D. R. (2010). Structuralinvestigation of the binding of 5-substituted swainsonine analogues to Golgi alpha-mannosidase II.Chembiochem, 11, 673–680.

11. Kumar, N. S., Kuntz, D. A., Wen, X., Pinto, B. M., & Rose, D. R. (2008). Binding of sulfonium-ionanalogues of di-epi-swainsonine and 8-epi-lentiginosine to Drosophila Golgi alpha-mannosidase II: the roleof water in inhibitor binding. Proteins, 71, 1484–1496.

12. Moremen, K. W. (1989). Isolation of a rat liver Golgi mannosidase II clone by mixed oligonucleotide-primedamplification of cDNA. Proceedings of the National Academy of Sciences of the United States of America,86, 5276–5280.

1250 Appl Biochem Biotechnol (2015) 177:1241–1251

13. Moremen, K. W., & Robbins, P. W. (1991). Isolation, characterization, and expression of cDNAs encodingmurine alpha-mannosidase II, a Golgi enzyme that controls conversion of high mannose to complex N-glycans. The Journal of Cell Biology, 115, 1521–1534.

14. Misago, M., Liao, Y. F., Kudo, S., Eto, S., Mattei, M. G., Moremen, K. W., & Fukuda, M. N. (1995).Molecular cloning and expression of cDNAs encoding human alpha-mannosidase II and a previouslyunrecognized alpha-mannosidase IIx isozyme. Proceedings of the National Academy of Sciences of theUnited States of America, 92, 11766–11770.

15. Strasser, R., Schoberer, J., Jin, C., Glossl, J., Mach, L., & Steinkellner, H. (2006). Molecular cloning andcharacterization of Arabidopsis thaliana Golgi alpha-mannosidase II, a key enzyme in the formation ofcomplex N-glycans in plants. The Plant Journal, 45, 789–803.

16. Foster, J. M., Yudkin, B., Lockyer, A. E., & Roberts, D. B. (1995). Cloning and sequence analysis of GmII, aDrosophila melanogaster homologue of the cDNA encoding murine Golgi alpha-mannosidase II. Gene, 154,183–186.

17. Jarvis, D. L., Bohlmeyer, D. A., Liao, Y. F., Lomax, K. K., Merkle, R. K., Weinkauf, C., & Moremen, K. W.(1997). Isolation and characterization of a class II alpha-mannosidase cDNA from lepidopteran insect cells.Glycobiology, 7, 113–127.

18. Paschinger, K., Hackl, M., Gutternigg, M., Kretschmer-Lubich, D., Stemmer, U., Jantsch, V., Lochnit, G., &Wilson, I. B. (2006). A deletion in the golgi alpha-mannosidase II gene of Caenorhabditis elegans results inunexpected non-wild-type N-glycan structures. The Journal of Biological Chemistry, 281, 28265–28277.

19. Rabouille, C., Kuntz, D. A., Lockyer, A., Watson, R., Signorelli, T., Rose, D. R., van den Heuvel, M., &Roberts, D. B. (1999). The Drosophila GMII gene encodes a Golgi alpha-mannosidase II. Journal of CellScience, 112(Pt 19), 3319–3330.

20. Qiu, X., Li, D., Cui, J., Liu, Y., & Wang, X. (2014). Molecular cloning, characterization and expressionanalysis of melanotransferrin from the sea cucumber Apostichopus japonicus. Molecular Biology Reports,41, 3781–3791.

21. Feng, X., Yu, X., Pang, M., Liu, H., & Tong, J. (2015). Molecular characterization and expression of threepreprosomatostatin genes and their association with growth in common carp (Cyprinus carpio). ComparativeBiochemistry and Physiology. Part B, Biochemistry & Molecular Biology, 182, 37–46.

22. Fazel, R., Zarei, N., Ghaemi, N., Namvaran, M. M., Enayati, S., Mirabzadeh Ardakani, E., Azizi, M., &Khalaj, V. (2014). Cloning and expression of Aspergillus flavus urate oxidase in Pichia pastoris.Springerplus, 3, 395.

23. Hossain, M. A., Nakano, R., Nakamura, K., Hossain, M. T., & Kimura, Y. (2010). Molecular characterizationof plant acidic alpha-mannosidase, a member of glycosylhydrolase family 38, involved in the turnover of N-glycans during tomato fruit ripening. Journal of Biochemistry, 148, 603–616.

24. Joshi, S., & Satyanarayana, T. (2014). Optimization of heterologous expression of the phytase (PPHY) ofPichia anomala in P. pastoris and its applicability in fractionating allergenic glycinin from soy protein.Journal of Industrial Microbiology & Biotechnology, 41, 977–987.

25. Xiangya, K., Jiangye, Z., Ying, W., Jianfei, L., & Qinfan, L. (2014). Molecular characterization of Caprahircus lysosomal alpha-mannosidase and potential mutant site for the therapy of locoweed poisoning. ActaBiochimica Polonica, 61, 77–84.

26. Kiefer, F., Arnold, K., Kunzli, M., Bordoli, L., & Schwede, T. (2009). The SWISS-MODEL repository andassociated resources. Nucleic Acids Research, 37, D387–D392.

27. Trott, O., & Olson, A. J. (2010). AutoDock Vina: improving the speed and accuracy of docking with a newscoring function, efficient optimization, and multithreading. Journal of Computational Chemistry, 31, 455–461.

28. Altmann, F., & Marz, L. (1995). Processing of asparagine-linked oligosaccharides in insect cells: evidencefor alpha-mannosidase II. Glycoconjugate Journal, 12, 150–155.

29. Kaushal, G. P., Szumilo, T., Pastuszak, I., & Elbein, A. D. (1990). Purification to homogeneity and propertiesof mannosidase II from mung bean seedlings. Biochemistry, 29, 2168–2176.

30. Ren, J., Castellino, F. J., & Bretthauer, R. K. (1997). Purification and properties of alpha-mannosidase IIfrom Golgi-like membranes of baculovirus-infected Spodoptera frugiperda (IPLB-SF-21AE) cells. TheBiochemical Journal, 324(Pt 3), 951–956.

31. Shah, N., Kuntz, D. A., & Rose, D. R. (2008). Golgi alpha-mannosidase II cleaves two sugars sequentially inthe same catalytic site. Proceedings of the National Academy of Sciences of the United States of America,105, 9570–9575.

Appl Biochem Biotechnol (2015) 177:1241–1251 1251

本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。

学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，

提供一站式文献检索和下载服务”的24 小时在线不限IP

图书馆。

图书馆致力于便利、促进学习与科研，提供最强文献下载服务。

图书馆导航：

图书馆首页 文献云下载 图书馆入口 外文数据库大全 疑难文献辅助工具