Skip to main content
NCBI home page
Applied and Environmental Microbiology logoLink to Applied and Environmental Microbiology
. 2007 Nov 26;74(2):446–453. doi: 10.1128/AEM.02106-07

Efficient Antibody Production upon Suppression of O Mannosylation in the Yeast Ogataea minuta▿

Kousuke Kuroda 1,8, Kazuo Kobayashi 1, Yoshinori Kitagawa 1, Taishiro Nakagawa 1, Haruhiko Tsumura 2, Toshihiro Komeda 3, Daisuke Shinmi 4, Eiji Mori 4, Kazuhiro Motoki 4, Kazumi Fuju 4, Teruyuki Sakai 4, Koichi Nonaka 5, Takeshi Suzuki 5, Kimihisa Ichikawa 6, Yasunori Chiba 7, Yoshifumi Jigami 7,8,*
PMCID: PMC2223252  PMID: 18039826

Abstract

When antibodies were expressed in the methylotrophic yeast Ogataea minuta, we found that abnormal O mannosylation occurred in the secreted antibody. Yeast-specific O mannosylation is initiated by the addition of mannose at serine (Ser) or threonine (Thr) residues in the endoplasmic reticulum via protein O mannosyltransferase (Pmt) activity. To suppress the addition of O-linked sugar chains on antibodies, we examined the possibility of inhibiting Pmt activity by the addition of a Pmt inhibitor during cultivation. The Pmt inhibitor was found to partially suppress the O mannosylation on the antibodies. Surprisingly, the suppression of O mannosylation was associated with an increased amount of assembled antibody (H2L2) and enhanced the antigen-binding activity of the secreted antibody. In this study, we demonstrated the expression of human antibody in O. minuta and elucidated the relationship between O mannosylation and antibody production in yeast.

Antibodies for pharmaceutical applications have recently received a great deal of attention due to their specific antigen-binding activities, antibody-dependent cellular cytotoxicity, and other useful characteristics. Indeed, sales of antibodies for pharmaceuticals have grown rapidly over the past decade and are expected to exceed 30 billion U.S. dollars by 2010 (1). A large volume of antibodies is required for pharmaceuticals, as antibodies are primarily marketed for chronic conditions. Antibodies possess relatively low potency, which results in the need for accumulating doses over time (10). Although therapeutic antibodies have great potential value, they have also been viewed with some skepticism. Antibodies are expensive to produce, in part because they are commonly manufactured by using batch/fed-batch cultures of mammalian cells. The increasing demand to reduce the cost of antibody production has promoted research into the development of an antibody expression system that is more practical than expression with mammalian cells.

Development of transgenic plants and animals as alternative hosts is therefore a promising field of study. Monoclonal antibodies have, thus far, been successfully produced from a number of sources, including plants, the milk of transgenic goats, the eggs of transgenic chickens, etc. (4, 12, 32). Moreover, antibodies derived from certain newly developed transgenic systems share physical characteristics that are similar to those of antibodies from mammalian cells such as Chinese hamster ovary (CHO) cells while exhibiting higher antibody-dependent cellular cytotoxicity activity due to the absence of fucose residues in N-linked sugar chains (6, 45). These alternative transgenic expression systems could reduce the cost of large-scale antibody production. However, the prolonged construction of transgenics remains a major disadvantage in terms of market demands.

The production of antibodies and antibody fragments has been studied by using various microorganism expression systems, including Escherichia coli, fungus, and yeast (7, 13, 37, 41). Since microorganisms are known to possess certain advantages as hosts (e.g., rapid growth rate, high heterologous protein productivity, and inexpensive cultivation costs), they have great potential to overcome some of the current hindrances to antibody production. With these advantages in mind, we focused on the construction of an antibody production system using a methylotrophic yeast, Ogataea minuta. For the production of pharmaceutical-grade proteins by yeast, it is necessary to convert the yeast-specific, N-linked, high-mannan-type sugar chain, which is modified after protein translation, into a human-compatible type sugar chain, because high-mannan sugar chains can exhibit antigenicity against humans. Therefore, we previously introduced a novel approach to humanizing the yeast-specific N-linked sugar chain (5), and some researchers have shown the usefulness of a yeast-based system for generating a human-compatible N-linked sugar chain, in a report on antibody production in yeast (25).

On the other hand, yeast-specific O mannosylation also has to be considered in the context of pharmaceutical manufacture. In yeast, protein O mannosylation is initiated in the endoplasmic reticulum (ER) by the addition of mannose to Ser or Thr residues, and this reaction occurs via protein O mannosyltransferase (Pmt) (38). O glycosylation in mammalian cells is initiated generally by the addition of N-acetylgalactosamine via polypeptide N-acetylgalactosaminyltransferase activity in the Golgi apparatus or else by the addition of a mannose via members of the Pomt family, which are homologues of yeast Pmt, in the ER (27, 34). The structure of O-linked sugar chains in yeast differs from that of mammalian cells (18, 43). Moreover, since yeasts possess a mechanism with which to increase the solubility of aberrant proteins by O mannosylation in order to suppress aggregation in the ER, heterologous proteins, which are not O glycosylated in mammalian cells, may suffer from O mannosylation in yeast (19, 29).

Previously, we demonstrated the usefulness of protease-deficient strains of O. minuta for the production of antibodies; in this strain, the O. minuta PEP4 (OmPEP4) and the OmPRB1 genes, which code for vacuolar protease, and the OmYPS1 gene, which codes for an aspartic protease, were disrupted (23). Additionally, we recently found that abnormal O mannosylation occurred in antibodies secreted from O. minuta. To the best of our knowledge, there have been no reports of O mannosylation occurring in antibodies produced by mammalian cells. For the production of pharmaceuticals, it is generally considered that yeast-specific O-linked sugar chains have to be removed from antibodies, because some mannoses added by Pmt enzymes might possess immunogenicity against humans. To suppress the addition of O-linked sugar chains to antibodies, we examined the suppression of Pmt activity by the addition of a Pmt inhibitor during cultivation.

In the present study, we report the relationship between O mannosylation and antibody production in yeast and discuss the possibility that yeast could be used to produce antibodies identical to those from mammalian cells via the suppression of O mannosylation.

MATERIALS AND METHODS

Yeast strain and media.

Cells of O. minuta YK6 (Δoch1 Δyps1Δura3Δade1), the host strain used in this study, were constructed as described in our previous study (23). The yeast cells were cultured at 27°C in the following media: YPG medium (containing the following components per liter, 10 g of yeast extract, 20 g of peptone, 20 ml of glycerol), and 2× BYPG medium (containing the following components per liter, 40 g of peptone, 20 g of yeast extract, 13.4 g of yeast nitrogen base [YNB] without amino acids, 100 ml of 0.1 M phosphate buffer [pH 6.0], 40 ml of glycerol, 0.2 g of uracil, 0.2 g of adenine, and 0.4 mg of biotin). Transformants were selected on SG agar plates (containing the following components per liter, 6.7 g of YNB without amino acids, 10 ml of glycerol, and 20 g of agar).

DNA methods.

E. coli DH5α cells were used for the subcloning of the plasmids. The plasmids were prepared using a QIAprep Spin miniprep kit (Qiagen, Hilden) from E. coli DH5α cells. DNA fragments were recovered from agarose gel using a QIAquick gel extraction kit (Qiagen). The DNA fragments amplified by PCR were subjected to DNA sequence analysis using a DNA sequencer (model 3700; Applied Biosystems, Foster, CA). The O. minuta genome was prepared using GENtorukun (Takara Bio, Shiga, Japan).

Plasmid preparation.

The plasmid pOMEU1, which contains the O. minuta orotidine-5′-phosphate decarboxylase (OmURA3) gene and sL/pOMGPA1 and sH/pOMGPU1ΔSp, which carry the human anti-TRAIL-R antibody light or heavy chain gene fused with a Saccharomyces cerevisiae invertase secretion signal under the control of the OmTDH1 promoter, were prepared as described in our previous studies (23, 24). The NdeI-EcoRI-digested plasmid pUC19 (Takara Bio) was subjected to Klenow treatment and was self-ligated. The plasmid obtained was designated pUC19ΔNd-E. The HindIII-KpnI-digested antibody heavy chain expression cassette from sH/pOMGPU1ΔSp was inserted into the HindIII-KpnI-digested pUC19ΔNd-E plasmid. The resultant plasmid was named pUC19ΔNd-E/H. DNA fragments of the O. minuta PGK promoter, the terminator (submitted to the DDBJ/GenBank/EMBL under the accession number AB363404), and the hygromycin B (HYG)-resistant gene were amplified by PCR using a genomic DNA template or pGARH (35) as a template of the HYG-resistant gene. The antibody heavy chain expression plasmid sH/pOMGPH1 was obtained by inserting PCR fragments by means of an In-Fusion PCR cloning kit (BD Biosciences, San Jose, CA) into the KpnI-digested pUC19ΔNd-E/H plasmid.

Construction of the strain expressing antibody genes.

The NotI-digested pOMEU1 plasmid was introduced into O. minuta YK6 cells (Δoch1Δyps1Δura3Δade1) to complement the OmURA3 marker gene by means of the electric pulse method described previously (24). This strain, which was obtained by selection on SG-plus-Ura plates (SG agar plates with 0.77 g/liter of −Ura dropout supplement [BD Biosciences]), was designated O. minuta YK6U (Δoch1Δyps1Δade1). To construct the antibody expression strain, the NotI-digested antibody light chain expression vector sL/pOMGPA1 and the Sse8387I-digested heavy chain expression vector sH/pOMGPH1 were cotransformed into O. minuta YK6U cells. The transformant was screened on SG-plus-ADE plates (SG agar plates with 0.6 g/liter of −Ade/−His/−Leu/−Trp dropout supplement [BD Biosciences]) supplemented with 50 μg/ml HYG, and the introduction of the light chain and the heavy chain genes was confirmed by PCR using a genomic DNA template. The obtained strain, in which both the heavy and the light chain genes were integrated into the genome, was designated O. minuta YK6U-HL (Δoch1 Δyps1).

Cultivation conditions.

A rhodanine-3-acetic acid derivative, the Pmt inhibitor used in this study, was designated R3AD and was prepared according to the method described in a report by Orchard et al. (31). The R3AD compound was dissolved in a 10 mM concentration of dimethyl sulfoxide (DMSO). The 2× BYPG3RD medium was prepared by the addition of 10 μl of R3AD to 50 ml of 2× BYPG medium. O. minuta YK6U-HL cells were precultured in 100 ml of YPG medium for 24 h at 27°C. Cells were harvested by centrifugation and grown at an initial optical density at 600 nm of 10 in 50 ml 2× BYPG3RD medium or in 50 ml 2× BYPG medium containing 10 μl of DMSO, which was the same amount as that used in the case of 2× BYPG3RD. The cell density at OD600 nm was measured at 24-h intervals, and 1 μl of R3AD in DMSO was added to the 2× BYPG3RD medium for every 1-unit increase of cell density at OD600 (e.g., if the detected absorbance at OD600 was 30 in 50 ml of medium, then 20 μl of R3AD was added). As a control, 1 μl of DMSO without R3AD was also added to 2× BYPG medium consecutively for the 1-unit increase of cell density at OD600. These cultures were maintained for 72 h at 27°C. The cells were then harvested by centrifugation, and the supernatants were analyzed as described below.

Western blotting analysis.

Twenty microliters of the supernatant or 3 μl of the cell suspension in phosphate-buffered saline (PBS) buffer (Gibco, Grand Island, NY) at a concentration of OD600 50 was subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) under the reducing condition in the presence of dithiothreitol (DTT). In order to detect deglycosylated antibodies, N-glycanase treatment was carried out according to the manufacturer's protocol. Samples, including antibodies, were boiled at 100°C for 5 min in the presence of 0.5% SDS and 50 mM 2-mercaptoethanol and were then incubated with 1.25% Nonidet P-40 and N-glycosidase F (Roche, Indianapolis, IN) at 37°C overnight. Samples equivalent to 4.5 μl of supernatant were also subjected to SDS-PAGE under the reducing condition. One microliter of the supernatant or 3 μl of the cell suspension described above was subjected to SDS-PAGE under the nonreducing conditions in the absence of DTT. The polyvinylidene difluoride membranes on which the proteins were electroblotted were blocked by using Block Ace (Dainippon Sumitomo Pharma, Osaka, Japan), and the samples were incubated with horseradish peroxidase (HRP)-conjugated goat polyclonal anti-human Fc antibody (1:5,000 dilution; Sigma-Aldrich, St. Louis, MO) and alkaline phosphatase (AP)-conjugated goat anti-human kappa antibody (1:5,000 dilution; Sigma-Aldrich). The antibodies were detected using a Supersignal West Dura kit (Pierce, Rockford, IL) for HRP and an alkaline phosphatase conjugate substrate kit (Bio-Rad Laboratories, Hercules, CA) for AP. The control, human anti-TRAIL-R antibody (hMAb) was produced at Kirin (Takasaki, Japan) by using CHO DG44 cells.

Quantification of the secreted antibody by EnVision.

The antibody productivity in supernatants cultivated with or without R3AD was measured using EnVision, based on fluorescent resonance energy transfer (FRET) technology. Supernatants were diluted 20 times in 20 mM Tris-HCl buffer (pH 7.2) including 10% Block Ace. Two microliters of diluted samples was incubated with 12 μl of 20 mM Tris-HCl buffer (pH 7.2) including 0.97 μg/ml LANCE europium (Eu)-W1024-labeled mouse anti-human immunoglobulin G (Fc-specific, monoclonal antibody) (PerkinElmer, Boston, MA), 8.3 μg/ml of biotin-conjugated mouse anti-human immunoglobulin G (kappa-specific, monoclonal antibody) (BD Biosciences), 16.7 μg/ml Surelight allophycocyanin (APC) streptavidin (PerkinElmer), and 10% Block Ace on 96-well half-area plates for 1 h at 25°C with light shielding. The fluorescence of each well was detected using EnVision (PerkinElmer), and the concentration of the assembled antibody in the supernatant was calculated by the ratio of absorbance at 665 nm/615 nm. In order to calculate the antibody concentration in the supernatants, the hMAb (Kirin, Takasaki, Japan) produced by CHO DG44 cells was used as the standard to generate the calibration curve. These procedures were performed according to the manufacturer's protocol.

FACS analysis.

The antigen-binding activity of the antibodies was measured according to a method described by Mori et al. (28). Antibodies from O. minuta YK6U-HL were prepared by means of protein A column chromatography as described below. L929/DR5ΔDD cells, which display the antigen TRAIL-Receptor on the cell surface, were cultured in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum and streptomycin. Cells (5 × 105 cells/sample) were washed with staining medium (PBS buffer containing 1% fetal bovine serum and 0.1% NaN3) and incubated with yeast-derived or CHO cell-derived antibodies in various concentrations at 4°C for 2 h. The cells were washed with staining medium and labeled with R-phycoerythrin-conjugated rabbit polyclonal anti-human kappa antibody (1:40 dilution; DAKO, Glostrup) in staining medium at 4°C for 1 h. The labeled cells were washed, resuspended with staining medium, and analyzed on a FACSCalibur model fluorescence-activated cell sorter (BD Biosciences). Histograms were generated using the fluorescence signal intensity of 10,000 counts, and the geometric mean fluorescence, which referred to values normalized to the log scale, was calculated using CellQuest software with the FACScan system (BD Biosciences).

Antibody purification.

The culture supernatants were purified by protein A column chromatography (Hi-Trap protein A HP 5 ml/CV; GE Healthcare Bio-Sciences, Piscataway, NJ). The columns were equilibrated using 10 mM PBS buffer (pH 6.0). Each supernatant was applied to a column at a flow rate of 5 ml/min, and each column was washed with 10 mM PBS buffer (pH 6.0). The antibodies present in the column were eluted with 20 mM sodium citrate buffer (pH 3.2). The antibodies eluted by protein A purification were subsequently subjected to gel permeation column chromatography (Superdex 200 HR 10/30, 23.562 ml/CV; GE Healthcare Bio-Sciences) equilibrated with PBS buffer (pH 7.35 to 7.65, using 9.5 mM PBS tablets [Takara Bio]). The samples were applied to the gels at a flow rate of 0.5 ml/min, and a chromatogram was detected by the integration of the UV absorbance signal at OD280. All separations were carried out on an Akta high-performance liquid chromatography system (GE Healthcare Bio-Sciences).

RESULTS

Effect of R3AD on O mannosylation in antibody production.

O. minuta YK6U-HL cells (Δoch1 Δyps1), in which light and heavy antibody chains were expressed, were cultivated in 2× BYPG medium at 24°C for 72 h, and the supernatant was subjected to Western blotting (WB) analysis under the reducing condition. The WB analysis revealed that although heavy and light chains of the antibody were secreted from O. minuta YK6U-HL cells, the smear signals were detected at a higher molecular weight than that of the corresponding intact heavy and light chains (Fig. 1A and B, lanes 3 and 4). Incidentally, the anti-TRAIL-R antibody used in this study has only one N-glycosylation site in the CH2 region. The smear signals in the heavy chain, which had not completely disappeared after N-glycanase treatment, were thought to indicate that O mannosylation had occurred in the antibodies. Alternatively, since the molecular weight shift of the antibody heavy chain to the lower molecular weight was detected after the N-glycanase treatment in WB analysis (Fig. 1A, lanes 4 and 6), it was confirmed that the secreted antibody had an N-linked sugar chain. Next, the Pmt inhibitor R3AD was added to the culture medium in order to verify its efficacy for inhibiting cell growth and O mannosylation. The inhibitor used in this study was a rhodanine-3-acetic acid derivative designated 1c by Orchard et al. (31). The inhibitor 1c was found to inhibit Candida albicans Pmt1p with a 50% inhibitory concentration at a dose of 2.3 μM, and this inhibitor was shown to be effective at inducing morphological changes in C. albicans associated with a loss of Pmt activity in vitro. (31). The addition of R3AD had no effect on cell growth (data not shown), and a decrease in smear signals at the higher molecular weight was observed (Fig. 1A and B, lanes 5 and 6). Although the remaining smear signals for both antibody chains indicated that O mannosylation had not been completely suppressed, O mannosylation in the antibody was inhibited by the addition of R3AD. On the other hand, the signal of the heavy chain detected in the antibody obtained from O. minuta YK6U-HL cell cultures with R3AD appeared to be increased. This may not indicate the enhanced secretability of the single heavy chain but the increment of the assembled antibody, because the heavy chain is thought to be secreted after it has assembled with the light chain in the ER (2).

FIG. 1.

FIG. 1.

Open in a new tab

Detection of the heavy and light chains of secreted antibody. Supernatants prepared with and without R3AD from O. minuta YK6U-HL (Δoch1 Δyps1) cell cultures and the purified antibody from CHO cell cultures were treated with N-glycanase, and the samples were subjected to WB analysis under the reducing conditions in the presence of DTT. HRP-conjugated anti-human Fc antibody (A) and AP-conjugated anti-human kappa antibody (B) were used for antibody detection. Lanes 1, control antibody from CHO cells; 2, control antibody from CHO cells with N-glycanase treatment; 3, supernatant from O. minuta YK6U-HL cells; 4, supernatant from O. minuta YK6U-HL cells with N-glycanase treatment; 5, supernatant from O. minuta YK6U-HL cells with the addition of R3AD; 6, supernatant from O. minuta YK6U-HL cells with the addition of R3AD and N-glycanase treatment.

Suppression of O mannosylation results in an increase in secreted antibody.

Supernatants prepared with and without R3AD were subjected to WB analysis under nonreducing conditions. The amount of secreted assembled antibody (H2L2) was clearly increased in the antibody obtained from O. minuta YK6U-HL cell cultures with R3AD (Fig. 2A and B, lanes 2 and 3). This result indicates that the suppression of O mannosylation enhances the secretability of the assembled antibody. Moreover, the increase of the secreted assembled antibody was quantified by EnVision analysis, which detects assembled antibodies in supernatants by measuring the interaction between the Eu-labeled anti-human Fc and biotin-labeled anti-human kappa antibody, using FRET technology (Fig. 3A). The titer of the secreted antibody reached 60 μg/ml, i.e., threefold higher than that of the secreted antibody produced without the addition of R3AD (Fig. 3B).

FIG. 2.

FIG. 2.

Open in a new tab

Detection of the secreted assembled antibodies. The supernatants prepared with and without R3AD from O. minuta YK6U-HL (Δoch1 Δyps1) cell cultures and the purified antibody from CHO cells were subjected to WB analysis under nonreducing conditions in the absence of DTT. HRP-conjugated anti-human Fc antibody (A) and AP-conjugated anti-human kappa antibody (B) were used for antibody detection. Lanes 1, control antibody from CHO cells; 2, supernatant from O. minuta YK6U-HL cells; 3, supernatant from O. minuta YK6U-HL cells with R3AD added.

FIG. 3.

FIG. 3.

Open in a new tab

Quantification of secreted assembled antibodies by EnVision. Schematic diagram of EnVision (A) shows the interaction of Eu-labeled anti-human Fc antibody with the sample antibody under excitation from 360 nm to 615 nm. Fluorescence at 665 nm, which is the emission wavelength of allophycocyanin (APC), was generated by the energy transfer of the wavelength at 615 nm to APC conjugated with streptavidin, which bound to the biotin-labeled anti-human kappa antibody associated with the target antibody. The wavelength generated at 665 nm was measured to quantify the antibody concentration in the supernatant. Supernatants from O. minuta YK6U-HL (Δoch1 Δyps1) cells were incubated with anti-Fc and anti-kappa antibodies, and the samples were analyzed by EnVision (B). Data are averages from quadruplicate experiments, and error bars are mean maximum and minimum values.

Increase in the productivity of intracellular antibodies by the suppression of O mannosylation.

The intracellular antibodies in O. minuta YK6U-HL cells were detected by WB analysis. Production of both the light and heavy chains of the antibody was enhanced by the suppression of O mannosylation (Fig. 4A and B, lanes 2 and 3). Moreover, the addition of R3AD facilitated the efficient formation of the assembled antibody (H2L2), while only a small amount of assembled antibody (H2L2) was detected in samples prepared without O mannosylation inhibition (Fig. 4C, lanes 2 and 3). The O-mannosylated light and heavy chains of the antibody could be sorted into the protein degradation pathway in the cell, as a result of their inefficient assembly. Alternatively, the secretory capacity of the yeast is not likely to be sufficient to secrete whole assembled antibody formed in the cell with the suppression of O mannosylation.

FIG. 4.

FIG. 4.

Open in a new tab

Detection of intracellular antibodies. O. minuta YK6U-HL (Δoch1 Δyps1) cells resuspended in PBS buffer and purified antibody from CHO cells were subjected to WB analysis under the reducing (A and B) or nonreducing (C) conditions. Antibodies were detected by HRP-conjugated anti-human Fc antibody (A and C) and AP-conjugated anti-human kappa antibody (B). Lanes 1, control antibody from CHO cells; 2, resuspension of O. minuta YK6U-HL cells; 3, resuspension of O. minuta YK6U-HL cells prepared with R3AD.

Effect of O mannosylation on the affinity of the secreted antibody.

Antibodies from O. minuta YK6U-HL cells prepared with and without R3AD were purified by means of protein A column chromatography, and the purified products were subjected to FACS analysis to verify their abilities to bind to the antigen, which was displayed on the surface of the L929/DR5ΔDD cells. Antibodies produced by yeast or CHO cells, which interact with antigen on the cell surface, were detected using R-phycoerythrin-conjugated rabbit polyclonal anti-human kappa antibody. As a result, the suppression of O mannosylation by R3AD enhanced the antibody binding activity (Fig. 5). However, inhibited binding activity was observed for the antibodies from O. minuta YK6U-HL cell cultures with R3AD, compared to that of antibodies from CHO DG44 cells. This is possibly caused by the incomplete suppression of O mannosylation by R3AD or by the O mannosylation due to the remaining activity of some Pmt proteins, which are not suppressed by R3AD.

FIG. 5.

FIG. 5.

Open in a new tab

FACS binding analysis of the secreted antibodies. The binding activities of the antibodies were evaluated by flow cytometry. L929/DR5ΔDD cells were incubated with antibodies produced by CHO cells (closed triangles), and O. minuta YK6U-HL cells were treated without R3AD (closed circles) or with R3AD (closed squares). The geometric means of the fluorescence intensities were calculated by CellQuest.

Properties of antibodies produced with and without R3AD.

The antibodies purified by protein A column chromatography were subjected to gel permeation column chromatography. The major peak was detected in the high-molecular-weight fraction of the antibody prepared from cultures without R3AD (Fig. 6A). In the case of the antibody prepared from cultures with R3AD, the main peak was detected in the same fraction as that of the antibody derived from CHO cells, although in the former cultures, the polymer fraction was present (Fig. 6B). These results suggest that O mannosylation causes antibody aggregation; however, it is not clear that the aggregation of the antibody detected in the gel permeation column chromatography occurred in the cell. The suppression of O mannosylation enables the production of antibodies at a level similar to that of antibodies from mammalian cells. Although the polymer fraction detected in the antibody prepared from cultures with R3AD could have been due to the residual O-linked mannoses, the usefulness of the R3AD was clearly demonstrated by these results.

FIG. 6.

FIG. 6.

Open in a new tab

Analysis of secreted antibodies by gel permeation column chromatography. After the antibodies were purified by protein A column chromatography, gel permeation column chromatography was carried out in order to analyze the quality of the antibodies obtained from O. minuta YK6U-HL cell cultures without R3AD (A) and with R3AD (B). The desired peak fraction, in which the monomer antibody was detected, is shown in the chromatogram of the antibody from CHO cells (C).

DISCUSSION

The O mannosylation that occurs in the ER increases the solubility of aberrant proteins to prevent aggregation in yeast cells (29). In this study, we showed that the yeast secreted O-mannosylated antibodies. The antibody is likely to be secreted after the assembly of the light and heavy antibody chains in the ER because the interface between the chains is highly hydrophobic (2, 9). Since the hydrophobic interface causes aggregation in the ER, the antibody may suffer from O mannosylation, which is initially helpful to suppress the protein aggregation in yeast cells. In addition, inefficient secretion of antibody has been observed with yeast cells. It is conceivable that the inhibition of antibody assembly in the ER occurred by the addition of mannoses to the interface of the antibody chains via O mannosylation. Moreover, an increase in intracellular antibody was associated with the suppression of O mannosylation. This result indicated that antibodies with O-linked mannoses cannot assemble efficiently, resulting in their aggregation in the ER, and were then degraded by delivery to vacuoles (23, 44). Secreted antibodies with O mannosylation exhibited low antigen-binding activity and were of higher molecular weight than expected, whereas antibodies in which O mannosylation had been suppressed by R3AD showed increased antigen-binding activity and formation of the oligomeric species (H2L2). In yeast, O mannosylation generally stabilizes the endogenous secretory protein (26, 42). However, in antibody secretion, O mannosylation could work not as a stabilizer but as a destabilizer by inhibiting the assembly of both chains. These results indicate that with the suppression of O mannosylation, yeast can be used to produce antibodies similar to those produced by mammalian cells.

In mass spectrometric analysis of the secreted antibody, O-linked sugar chains on light chain in the regions of amino acid residues 1 to 18 and 25 to 45 and on heavy chain in the region of amino acid residues 149 to 211 were detected for the antibody produced in O. minuta YK6U-HL cells without R3AD. On the other hand, the O-linked sugar chains on the light chain in the region of amino acid residues 1 to 18 disappeared in the antibody produced with R3AD; however, other O-linked sugar chains still existed. Additionally, any O mannosylation could not be detected in the Fc region. Since mass peaks corresponding to the fragments of the VH1 to CH1 region, as opposed to those of the Fc region, were not clearly detected, the VH-to-CH1 region could be highly O mannosylated, forming high-molecular-weight species. The detailed analyses of the O. minuta O-linked sugar chains and the O-mannosylated Ser and/or Thr residues in the antibody are in progress.

It is possible that other Pmt proteins, not suppressed by R3AD, could be responsible for the remaining O mannosylation. Generally, in yeast, Pmt proteins have been categorized as belonging to the Pmt1p, Pmt2p, and Pmt4p subfamilies (17, 33, 39). The Pmt1p subfamily (i.e., Pmt1p and 5p) interacts with the Pmt2p subfamily (i.e., Pmt2p and 3p) to form active heteromeric complexes, while the Pmt4p subfamily forms a homomeric complex (17). Pmt1-2p subfamily complexes have different substrate specificities from that of the Pmt4p subfamily complex (21). Although R3AD was developed as a C. albicans Pmt1p inhibitor and the rhodanine inhibitors were shown to be relatively specific for C. albicans Pmt1p (3), it is not understood whether the inhibitors have an effect on the other Pmt proteins. Some compounds have been developed as C. albicans Pmt1p inhibitors (31). These compounds, designated as 1d and 4a, strongly inhibit the proliferation of C. albicans in the presence of G418, while 1d, 4a, and R3AD are almost equally effective at the inhibition of C. albicans Pmt1p. It is likely that 1d and 4a are more effective than R3AD at suppressing cell wall biosynthesis. Since some Pmt proteins are responsible for cell wall integrity (16), it is possible that these inhibitors inactivate not only C. albicans Pmt1p but also other Pmt proteins that are not inhibited by R3AD. Since triple- or double-knockout PMT gene strains lead to severe growth defects (16), it is difficult to disrupt the entire PMT genes, a factor which might be related to O mannosylation of the antibody. Therefore, the use of these Pmt inhibitors, which can control the inactivation of the other Pmt proteins (with the exception of Pmt1p) by the additive volume, is thought to be useful in antibody production by yeast.

Some reports have addressed the production of antibody by yeast, although in those studies, productivity was either low or not mentioned (15, 20, 25, 30, 44). For the production of antibodies in yeast, it will be necessary to increase productivity. To attain this goal, the introduction of chaperones may improve the assembly of the antibody in the ER, resulting in efficient secretion. However, introduction of the chaperones was not sufficient to yield high amounts of full-length antibody in yeast, while the overexpression of chaperones such as disulfide isomerase have been reported to contribute partially to the enhancement of the secretability of the antibody fragment (8, 14, 36). In contrast, an increase in secretion was confirmed with several kinds of different antibodies by the addition of R3AD to O. minuta cultures (data not shown). Therefore, the suppression of O mannosylation, as shown in this study, is expected to be useful for increasing antibody productivity in yeast; indeed, this is the first report to demonstrate a successful enhancement of secreted full-length antibody in yeast.

The overexpression of the antibody fragment referred to as 4-4-20 scFv induced an unfolded protein response (UPR) (22). Additionally the Pmt inhibitor or the pmt1 mutant stimulates the UPR pathway in C. albicans (3). Since the UPR system allows for an increase in the capacity for protein folding, secretion, and degradation by the up-regulation of various genes, to prevent the accumulation of unfolded proteins (40), up-regulation of the UPR pathway could be beneficial for the production of some secretory proteins. However, an increase in the secretion of human insulin-like growth factor was not detected in the pmt-deficient strain in which the UPR may be up-regulated, while improvement in the quality was confirmed (11). Therefore the UPR may not always contribute to the enhancement of the productivity of secretory proteins. On the other hand, the overexpression of whole-antibody genes may also induce the UPR and result in an enhancement of O mannosylation in the antibody because up-regulation of PMT genes occurs in the UPR (40). Therefore, the inactivation of Pmt proteins could improve the quality and secretability of the overexpressed and overproduced antibody.

Although the improved secretion of the protein by suppression of O mannosylation might not be a general phenomenon, the suppression of O mannosylation could be beneficial for the production of proteins forming either homomeric or heteromeric complexes through their hydrophobic interaction in yeast. Further study is needed to characterize the Pmt proteins in O. minuta and to inactivate the Pmt proteins associated with O mannosylation in the antibody by the construction of the PMT knockout strains and the use of various inhibitors.

Acknowledgments

We thank N. Kawashima, A. Ohnishi, Y. Yamamoto, U. Tezuka, K. Sato, and A. Ohshima for valuable assistance with this research. We also thank M. Tsukahara for helpful suggestions.

This work was supported by the New Energy and Industrial Technology Development Organization (NEDO) as part of the Research and Development Projects of the Industrial Science and Technology Frontiers Program, Japan.

Footnotes

▿

Published ahead of print on 26 November 2007.

REFERENCES

  • 1.Baker, M. 2005. Upping the ante on antibodies. Nat. Biotechnol. 23:1065-1072. [DOI] [PubMed] [Google Scholar]
  • 2.Bole, D. G., L. M. Hendershot, and J. F. Kearney. 1986. Posttranslational association of immunoglobulin heavy chain binding protein with nascent heavy chains in nonsecreting and secreting hybridomas. J. Cell Biol. 102:1558-1566. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Cantero, P. D., C. Lengsfeld, S. K. Prill, M. Subanović, E. Román, J. Pla, and J. F. Ernst. 2007. Transcriptional and physiological adaptation to defective protein-O-mannosylation in Candida albicans. Mol. Microbiol. 64:1115-1128. [DOI] [PubMed] [Google Scholar]
  • 4.Chadd, H. E., and S. M. Chamow. 2001. Therapeutic antibody expression technology. Curr. Opin. Biotechnol. 12:188-194. [DOI] [PubMed] [Google Scholar]
  • 5.Chiba, Y., M. Suzuki, S. Yoshida, A. Yoshida, H. Ikenaga, M. Takeuchi, Y. Jigami, and E. Ichishima. 1998. Production of human compatible high mannose-type (Man5GlcNAc2) sugar chains in Saccharomyces cerevisiae. J. Biol. Chem. 273:26298-26304. [DOI] [PubMed] [Google Scholar]
  • 6.Cox, K. M., J. D. Sterling, J. T. Regan, J. R. Gasdaska, K. K. Frantz, C. G. Peele, A. Black, D. Passmore, C. Moldovan-Loomis, M. Srinivasan, S. Cuison, P. M. Cardarelli, and L. F. Dickey. 2006. Glycan optimization of a human monoclonal antibody in the aquatic plant Lemna minor. Nat. Biotechnol. 24:1591-1597. [DOI] [PubMed] [Google Scholar]
  • 7.Damasceno, L. M., I. Pla, H. J. Chang, L. Cohen, G. Ritter, L. J. Old, and C. A. Batt. 2004. An optimized fermentation process for high-level production of a single-chain Fv antibody fragment in Pichia pastoris. Protein Expr. Purif. 37:18-26. [DOI] [PubMed] [Google Scholar]
  • 8.Damasceno, L. M., K. A. Anderson, G. Ritter, J. M. Cregg, L. J. Old, and C. A. Batt. 2007. Cooverexpression of chaperones for enhanced secretion of a single-chain antibody fragment in Pichia pastoris. Appl. Microbiol. Biotechnol. 74:381-389. [DOI] [PubMed] [Google Scholar]
  • 9.Elkabetz, Y., Y. Argon, and S. Bar-Nun. 2005. Cysteines in CH1 underlie retention of unassembled Ig heavy chains. J. Biol. Chem. 280:14402-14412. [DOI] [PubMed] [Google Scholar]
  • 10.Farid, S. S. 2007. Process economics of industrial monoclonal antibody manufacture. J. Chromatogr. B 848:8-18. [DOI] [PubMed] [Google Scholar]
  • 11.Finck, M., N. Bergmann, B. Jansson, and J. F. Ernst. 1996. Defective threonine-linked glycosylation of human insulin-like growth factor in mutants of the yeast Saccharomyces cerevisiae. Glycobiology 6:313-320. [DOI] [PubMed] [Google Scholar]
  • 12.Fischer, R., R. M. Twyman, and S. Schillberg. 2003. Production of antibodies in plants and their use for global health. Vaccine 21:820-825. [DOI] [PubMed] [Google Scholar]
  • 13.Freyre, F. M., J. E. Vázquez, M. Ayala, L. Canaán-Haden, H. Bell, I. Rodríguez, A. González, A. Cintado, and J. V. Gavilondo. 2000. Very high expression of an anti-carcinoembryonic antigen single chain Fv antibody fragment in the yeast Pichia pastoris. J. Biotechnol. 76:157-163. [DOI] [PubMed] [Google Scholar]
  • 14.Gasser, B., M. Maurer, J. Gach, R. Kunert, and D. Mattanovich. 2006. Engineering of Pichia pastoris for improved production of antibody fragments. Biotechnol. Bioeng. 94:353-361. [DOI] [PubMed] [Google Scholar]
  • 15.Gasser, B., and D. Mattanovich. 2007. Antibody production with yeasts and filamentous fungi: on the road to large scale? Biotechnol. Lett. 29:201-212. [DOI] [PubMed] [Google Scholar]
  • 16.Gentzsch, M., and W. Tanner. 1996. The PMT gene family: protein O-glycosylation in Saccharomyces cerevisiae is vital. EMBO J. 15:5752-5759. [PMC free article] [PubMed] [Google Scholar]
  • 17.Girrbach, V., and S. Strahl. 2003. Members of the evolutionarily conserved PMT family of protein O-mannosyltransferases form distinct protein complexes among themselves. J. Biol. Chem. 278:12554-12562. [DOI] [PubMed] [Google Scholar]
  • 18.Goto, M. 2007. Protein O-glycosylation in fungi: diverse structures and multiple functions. Biosci. Biotechnol. Biochem. 71:1415-1427. [DOI] [PubMed] [Google Scholar]
  • 19.Harty, C., S. Strahl, and K. Römisch. 2001. O-mannosylation protects mutant alpha-factor precursor from endoplasmic reticulum-associated degradation. Mol. Biol. Cell 12:1093-1101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Horwitz, A. H., C. P. Chang, M. Better, K. E. Hellstrom, and R. R. Robinson. 1988. Secretion of functional antibody and Fab fragment from yeast cells. Proc. Natl. Acad. Sci. USA 85:8678-8682. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Hutzler, J., M. Schmid, T. Bernard, B. Henrissat, and S. Strahl. 2007. Membrane association is a determinant for substrate recognition by PMT4 protein O-mannosyltransferases. Proc. Natl. Acad. Sci. USA 104:7827-7832. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Kauffman, K. J., E. M. Pridgen, F. J. Doyle III, P. S. Dhurjati, and A. S. Robinson. 2002. Decreased protein expression and intermittent recoveries in BiP levels result from cellular stress during heterologous protein expression in Saccharomyces cerevisiae. Biotechnol. Prog. 18:942-950. [DOI] [PubMed] [Google Scholar]
  • 23.Kuroda, K., Y. Kitagawa, K. Kobayashi, H. Tsumura, T. Komeda, E. Mori, K. Motoki, S. Kataoka, Y. Chiba, and Y. Jigami. 2007. Antibody expression in protease-deficient strains of methylotrophic yeast Ogataea minuta. FEMS Yeast Res. 17:1307-1316. [DOI] [PubMed] [Google Scholar]
  • 24.Kuroda, K., K. Kobayashi, H. Tsumura, T. Komeda, Y. Chiba, and Y. Jigami. 2006. Production of Man5GlcNAc2-type sugar chain by the methylotrophic yeast Ogataea minuta. FEMS Yeast Res. 6:1052-1062. [DOI] [PubMed] [Google Scholar]
  • 25.Li, H., N. Sethuraman, T. A. Stadheim, D. Zha, B. Prinz, N. Ballew, P. Bobrowicz, B. K. Choi, W. J. Cook, M. Cukan, N. R. Houston-Cummings, R. Davidson, B. Gong, S. R. Hamilton, J. P. Hoopes, Y. Jiang, N. Kim, R. Mansfield, J. H. Nett, S. Rios, R. Strawbridge, S. Wildt, and T. U. Gerngross. 2006. Optimization of humanized IgGs in glycoengineered Pichia pastoris. Nat. Biotechnol. 24:210-215. [DOI] [PubMed] [Google Scholar]
  • 26.Lommel, M., M. Bagnat, and S. Strahl. 2004. Aberrant processing of the WSC family and Mid2p cell surface sensors results in cell death of Saccharomyces cerevisiae O-mannosylation mutants. Mol. Cell. Biol. 24:46-57. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Manya, H., A. Chiba, A. Yoshida, X. Wang, Y. Chiba, Y. Jigami, R. U. Margolis, and T. Endo. 2004. Demonstration of mammalian protein O-mannosyltransferase activity: coexpression of POMT1 and POMT2 required for enzymatic activity. Proc. Natl. Acad. Sci. USA 101:500-505. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Mori, E., M. Thomas, K. Motoki, K. Nakazawa, T. Tahara, K. Tomizuka, I. Ishida, and S. Kataoka. 2004. Human normal hepatocytes are susceptible to apoptosis signal mediated by both TRAIL-R1 and TRAIL-R2. Cell Death Differ. 11:203-207. [DOI] [PubMed] [Google Scholar]
  • 29.Nakatsukasa, K., S. Okada, K. Umebayashi, R. Fukuda, S. Nishikawa, and T. Endo. 2004. Roles of O-mannosylation of aberrant proteins in reduction of the load for endoplasmic reticulum chaperones in yeast. J. Biol. Chem. 279:49762-49772. [DOI] [PubMed] [Google Scholar]
  • 30.Ogunjimi, A. A., J. M. Chandler, C. M. Gooding, A. Recinos III, and P. V. Choudary. 1999. High-level secretory expression of immunologically active intact antibody from the yeast Pichia pastoris. Biotechnol. Lett. 21:561-567. [Google Scholar]
  • 31.Orchard, M. G., J. C. Neuss, C. M. Galley, A. Carr, D. W. Porter, P. Smith, D. I. Scopes, D. Haydon, K. Vousden, C. R. Stubberfield, K. Young, and M. Page. 2004. Rhodanine-3-acetic acid derivatives as inhibitors of fungal protein mannosyl transferase 1 (PMT1). Bioorg. Med. Chem. Lett. 14:3975-3978. [DOI] [PubMed] [Google Scholar]
  • 32.Pollock, D. P., J. P. Kutzko, E. Birck-Wilson, J. L. Williams, Y. Echelard, and H. M. Meade. 1999. Transgenic milk as a method for the production of recombinant antibodies. J. Immunol. Methods 231:147-157. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Prill, S. K., B. Klinkert, C. Timpel, C. A. Gale, K. Schröppel, and J. F. Ernst. 2005. PMT family of Candida albicans: five protein mannosyltransferase isoforms affect growth, morphogenesis and antifungal resistance. Mol. Microbiol. 55:546-560. [DOI] [PubMed] [Google Scholar]
  • 34.Roth, J., Y. Wang, A. E. Eckhardt, and R. L. Hill. 1994. Subcellular localization of the UDP-N-acetyl-D-galactosamine: polypeptide N-acetylgalactosaminyltransferase-mediated O-glycosylation reaction in the submaxillary gland. Proc. Natl. Acad. Sci. USA 91:8935-8939. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Shimada, H., K. Kondo, P. D. Fraser, Y. Miura, T. Saito, and N. Misawa. 1998. Increased carotenoid production by the food yeast Candida utilis through metabolic engineering of the isoprenoid pathway. Appl. Environ. Microbiol. 64:2676-2680. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Shusta, E. V., R. T. Raines, A. Plückthun, and K. D. Wittrup. 1998. Increasing the secretory capacity of Saccharomyces cerevisiae for production of single-chain antibody fragments. Nat. Biotechnol. 16:773-777. [DOI] [PubMed] [Google Scholar]
  • 37.Simmons, L. C., D. Reilly, L. Klimowski, T. S. Raju, G. Meng, P. Sims, K. Hong, R. L. Shields, L. A. Damico, P. Rancatore, and D. G. Yansura. 2002. Expression of full-length immunoglobulins in Escherichia coli: rapid and efficient production of aglycosylated antibodies. J. Immunol. Methods 263:133-147. [DOI] [PubMed] [Google Scholar]
  • 38.Strahl-Bolsinger, S., M. Gentzsch, and W. Tanner. 1999. Protein O-mannosylation. Biochim. Biophys. Acta 1426:297-307. [DOI] [PubMed] [Google Scholar]
  • 39.Tanaka, N., Y. Fujita, S. Suzuki, M. Morishita, Y. Giga-Hama, C. Shimoda, and K. Takegawa. 2005. Characterization of O-mannosyltransferase family in Schizosaccharomyces pombe. Biochem. Biophys. Res. Commun. 330:813-820. [DOI] [PubMed] [Google Scholar]
  • 40.Travers, K. J., C. K. Patil, L. Wodicka, D. T. Lockhart, J. S. Weissman, and P. Walter. 2000. Functional and genomic analyses reveal an essential coordination between the unfolded protein response and ER-associated degradation. Cell 101:249-258. [DOI] [PubMed] [Google Scholar]
  • 41.Ward, M., C. Lin, D. C. Victoria, B. P. Fox, J. A. Fox, D. L. Wong, H. J. Meerman, J. P. Pucci, R. B. Fong, M. H. Heng, N. Tsurushita, C. Gieswein, M. Park, and H. Wang. 2004. Characterization of humanized antibodies secreted by Aspergillus niger. Appl. Environ. Microbiol. 70:2567-2576. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Weber, Y., S. K. Prill, and J. F. Ernst. 2004. Pmt-mediated O mannosylation stabilizes an essential component of the secretory apparatus, Sec20p, in Candida albicans. Eukaryot. Cell 3:1164-1168. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Willer, T., M. C. Valero, W. Tanner, J. Cruces, and S. Strahl. 2003. O-mannosyl glycans: from yeast to novel associations with human disease. Curr. Opin. Struct. Biol. 13:621-630. [DOI] [PubMed] [Google Scholar]
  • 44.Wood, C. R., M. A. Boss, J. H. Kenten, J. E. Calvert, N. A. Roberts, and J. S. Emtage. 1985. The synthesis and in vivo assembly of functional antibodies in yeast. Nature 314:446-449. [DOI] [PubMed] [Google Scholar]
  • 45.Zhu, L., M. C. van de Lavoir, J. Albanese, D. O. Beenhouwer, P. M. Cardarelli, S. Cuison, D. F. Deng, S. Deshpande, J. H. Diamond, L. Green, E. L. Halk, B. S. Heyer, R. M. Kay, A. Kerchner, P. A. Leighton, C. M. Mather, S. L. Morrison, Z. L. Nikolov, D. B. Passmore, A. Pradas-Monne, B. T. Preston, V. S. Rangan, M. Shi, M. Srinivasan, S. G. White, P. Winters-Digiacinto, S. Wong, W. Zhou, and R. J. Etches. 2005. Production of human monoclonal antibody in eggs of chimeric chickens. Nat. Biotechnol. 23:1159-1169. [DOI] [PubMed] [Google Scholar]

Articles from Applied and Environmental Microbiology are provided here courtesy of American Society for Microbiology (ASM)

RESOURCES