HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Tomura, T. Articles by Ishizaka, K. Search for Related Content PubMed PubMed Citation Articles by Tomura, T. Articles by Ishizaka, K. Pubmed/NCBI databases Substance via MeSH

Immunosuppressive Activities of Recombinant Glycosylation -Inhibiting Factor Mutants

Takafumi Tomura^*, Hiroshi Watarai^*, Nakayuki Honma^*, Masahiro Sato^*, Akihiro Iwamatsu, Yoichi Kato, Ryota Kuroki, Tatsumi Nakano, Toshifumi Mikayama^* and Kimishige Ishizaka^1,

^* Pharmaceutical Research Laboratory, Kirin Brewery Co. Ltd., Takasaki, Gunma, Japan; Central Laboratories for Key Technology, Kirin Brewery Co. Ltd., Yokohama, Japan; and La Jolla Institute for Allergy and Immunology, San Diego, CA 92121

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
We have shown previously that glycosylation-inhibiting factor (GIF) in culture supernatants of suppressor T cell (Ts) hybridomas had bioactivity, while the same cells contained a substantial quantity of inactive GIF in cytosol. Mass-spectrometric analysis of GIF in the culture supernatant and cytosol of a Ts hybridoma provided direct evidence that GIF protein was posttranslationally modified in the Ts cells, and that the GIF bioactivity is associated with the posttranslationally modified species. Assuming that conformational changes induced by the posttranslational modifications are responsible for generation of bioactivity, we constructed cysteine mutants of human rGIF (rhGIF) in which cysteine at position 57, 60, or 81 was replaced with Ala, and the mutants were expressed in Escherichia coli. Replacement of Cys⁵⁷ or Cys⁶⁰ with Ala resulted in generation of bioactivity, while replacement of Cys⁸¹ with Ala failed to do so. It was also found that replacement of Cys⁵⁷ with Ala and carboxymethylation of a sulfhydryl group in Cys⁶⁰ synergistically increased the GIF bioactivity of the GIF derivatives. A mutated GIF protein, in which Cys⁵⁷ and Asn¹⁰⁶ in the rhGIF were replaced with Ala and Ser, respectively, had immunosuppressive effects on the IgE and IgG1 Ab responses of BDF₁ mice to DNP-OVA, while wild-type rhGIF did not. Evidence was obtained that the mutated GIF suppressed Ag priming of Th cells for the Ab responses and proliferative response.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
In the course of previous studies on the regulation of the IgE Ab response, we have described glycosylation-inhibiting factor (GIF),² a 13-kDa cytokine, which is involved in the selective formation of IgE-suppressive factors (1). GIF inhibited N-glycosylation of IgE binding factors (IgE-BF), and the unglycosylated IgE-BF, in turn, selectively suppressed the IgE synthesis. The major cell source of GIF is Ag-specific suppressor T (Ts) cells (2). It was also found that immunosuppressive effects of GIF were not restricted to the IgE isotype. Repeated injections of partially purified GIF into BDF₁ mice resulted in suppression of not only IgE Ab response, but also IgG Ab response to aluminum hydroxide gel (alum)-absorbed OVA (3). Subsequent experiments provided evidence that GIF is a subunit of Ag-specific suppressor T cell factors (TsF) (4, 5).

After molecular cloning of GIF cDNA, however, we realized that GIF mRNA was detectable in various cell types, including Th cells (6), and that many of the cell line cells secreted the 13-kDa peptide, which reacted with polyclonal Abs against Escherichia coli-derived rGIF. Nevertheless, only the peptide secreted by Ts cells demonstrated GIF bioactivity (7). It was also found that bioactivity of human rGIF (rhGIF) expressed in E. coli or BMT10 cells was barely detectable. However, transfection of a chimeric cDNA encoding a fusion protein consisting of the N-terminal pro-region of calcitonin precursor and hGIF into the same cells resulted in secretion of bioactive GIF. Evidence was obtained that the fusion protein was translocated into the endoplasmic reticulum and cleaved by endoproteinase at the Golgi apparatus to form the mature 13-kDa peptide (7). No difference was detected between the bioactive GIF and inactive GIF by SDS-PAGE. The results indicated that posttranslational modifications of the protein are involved in the generation of bioactivity, and suggested that heterogeneity of GIF in bioactivity is due to conformational transition of the same peptide.

This possibility was supported by our subsequent experiments, which indicated that E. coli-derived rhGIF could be converted to bioactive derivatives by chemical modifications of a single cysteine residue at protein position 60 with a sulfhydryl reagent such as iodoacetate or 5,5'-dithiobis (2-nitrobenzoic acid) (DTNB) (8). The GIF bioactivity of the DTNB derivative was comparable with that of the Ts-derived bioactive GIF. In view of evidence that an interaction of a cysteine residue in a protein with either heavy metal or lipid may affect conformation of the protein (9, 10), the possibility was considered that a point mutation at a cysteine residue in rhGIF may also generate bioactivity. The present experiment shows that substitution of alanine for Cys⁵⁷ or Cys⁶⁰, but not for Cys⁸¹, in rhGIF results in the generation of bioactivity. The results also indicate that substitution of Ser for Asn¹⁰⁶, together with the Cys⁵⁷Ala substitution, results in the generation of substantial immunosuppressive activity on the in vivo Ab responses.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Cell lines

The human Ts hybridoma 31E9 (6), and the murine T cell hybridoma 12H5 cells (11) were previously described. They were maintained in high glucose DMEM containing 10% FCS. Detailed ingredients for the medium were previously described (12).

Ags and Abs

Crystalline OVA was purchased from Sigma (St. Louis, MO). Urea-denatured (UD) OVA was prepared by the method previously described (13), with slight modifications. As indicated in a previous article (14), UD-OVA lacked the major B cell epitope in native OVA, and failed to react with Abs against native OVA.

2,4-Dinitrophenyl derivatives of OVA (DNP-OVA) were prepared by mixing 200 mg OVA with 100 mg 2,4-dinitrofluorobenzene (Kodak, Rochester, NY) for 18 h at room temperature. The number of DNP groups coupled to OVA was 3.2/molecule. To prepare biotin-coupled DNP derivatives of BSA (DNP-BSA), 200 mg BSA (Sigma) was mixed with 100 mg 2,4-dinitrofluorobenzene, and DNP-BSA thus prepared was biotinylated using Ab Labeling System kit (American Qualex, San Clemente, CA). A preparation of mouse IgE anti-DNP mAb H1 DNP--26 (15) was previously described (16). Mouse IgG1 anti-DNP mAb H1 DNP--109 (15) was kindly gifted from Dr. Fu-Tong Liu (La Jolla Institute for Allergy and Immunology, San Diego, CA). Anti-mouse IgE mAb 6HD5 (17) was kindly supplied by Dr. K. Okumura (Juntendo University, Tokyo, Japan). Alkaline phosphatase-coupled monoclonal anti-mouse IgG1, IgG2a, and IgG2b (Zymed, South San Francisco, CA), and purified mouse IgG2a and IgG2b (PharMingen, San Diego, CA) were purchased. Polyclonal anti-rhGIF was prepared by immunization of rabbits with 100 µg rhGIF emulsified in CFA. Eight booster injections of 100 µg rhGIF emulsified in IFA were given at 2–3-wk intervals, and anti-rhGIF antiserum was obtained 7 days after the last immunization. The IgG fraction of the antiserum was obtained using protein A affinity adsorbent (Prosep-A, Bioprocessing, Durham, U.K.), and 200 mg IgG was coupled to 20 ml N-hydroxysuccinimide -activated HiTrap (Pharmacia, Piscataway, NJ).

Expression of rhGIFs in E. coli

The rhGIF (wild-type rhGIF) was expressed in E. coli by the method previously described (6). For the expression of hGIF mutant, C57A, in which cysteine at position 57 in the hGIF sequence was replaced with alanine, PCR was performed using hGIF cDNA fragment inserted into pST811 vector (6) as template, and the following oligonucleotides as primers: 5'-AACCTTAAGAAAAACCAAGGAGGTAATAAATAATGCCGATGTTCATCGTAAACACCAACG-3' (primer 1) and 3'-CTCGGCCGGCGCGAGACGTCGGAC-5'. Amplified DNA were digested with AflII and PstI to obtain 5' end fragment of C57A cDNA. In a separate step, the pST811 plasmid containing hGIF cDNA was cleaved with PstI and BamHI, and 3' end DNA fragment of hGIF cDNA was recovered. The two DNA fragments were inserted into pST811 cleaved with AflII and BamHI. The plasmid obtained by the procedure was denominated pC57A-hGIF and expresses C57A.

Expression plasmid pC60A-hGIF was constructed in a similar manner. It expresses the hGIF mutant, C60A, in which the cysteine residue at position 60 of wild-type GIF was replaced with alanine. PCR was performed using primer 1 and the oligonucleotide primer, 3'-CGCGAGCGATCGGACGTGTCGTAG-5', and amplified 5' end DNA fragment was recovered by digestion with AflII and NheI. A 3' end DNA fragment was prepared by PCR using the primers 5'-GCGCTCGCTAGCCTGCACAGCATC-3' and 3'-CACCCGACCTTGTTGAGGTGGAAGCGGGATTATCCCTAGGCAA-5' (primer 2), and recovered by digestion with NheI and BamHI. The two DNA fragments were then inserted into vector pST811 that had been cleaved with AflII and BamHI.

Expression plasmid pC81A-hGIF expressing the mutant, C81A, in which the cysteine residue at position 81 was replaced with alanine, was constructed in a similar manner. PCR was performed using primer 1, and the primer having the sequence 3'-GGCCAGCAGGCCGGCTAGCAGCTT-5', and amplified DNA fragment was digested with AflII and NheI. A 3' end DNA fragment was prepared using primer 2 and the primer, 5'-AAGCTGCTAGCCGGCCTGCTGGCC-3', and digested with NheI and BamHI. The two DNA fragments were inserted into vector pST811 that had been cleaved with AflII and BamHI.

For the expression of C57A/N106S, in which Cys⁵⁷ and Asn¹⁰⁶ in rhGIF were replaced with alanine and serine, respectively, PCR was performed using pC57A-hGIF as template, and the oligonucleotide primers: 5'-GAGCCGGCCGCGCTCTGCAGCCTG-3' and 3'-CGCCGGTCGCACCCGACCTTGTTGAGGTGGAAGCGGGATTATCCCTAGGCCAA-5'. Amplified DNA fragments were digested with PstI and BamHI to obtain 3' end DNA fragment. 5'-end C57A fragment was obtained by digestion of pC57A-hGIF with AflII and PstI. These fragments were inserted into vector pST811 that had been cleaved with AflII and BamHI.

Each plasmid was transformed into competent E. coli RR1 cells, and the cells carrying the plasmid were cultured in M9 broth containing glucose (0.8%), casamino acid (0.4%), thiamine (10 µg/ml), and ampicillin (50 µg/ml). Cells were harvested 5 h after the addition of indoleacrylic acid.

Purification of GIFs

Culture supernatants of the 31E9 cells in serum-free high glucose DMEM were concentrated 40-fold, and the concentrated sample was circulated overnight at 4°C through a 20-ml column of anti-GIF-coupled HiTrap. After washing with 20 column volumes of PBS, proteins retained in the column were eluted with 0.1 M glycine-HCl buffer, pH 3. Cytosolic fraction of the 31E9 cells was obtained by the procedures previously described (7), and GIF in the cytosol was affinity purified by the same method. The purity of the GIFs from both culture supernatant and cytosol was higher than 95%, as verified by SDS-PAGE and silver staining (18).

Recombinant hGIF and mutant GIF proteins expressed in E. coli were partially purified from a soluble fraction of the cells by the method previously described (19). The partially purified rhGIF preparations were dialyzed against 20 mM sodium acetate buffer (pH 5.5) and applied to a CM-5PW column (Tosoh, Tokyo, Japan) equilibrated with the same buffer. The rhGIF was eluted with a gradient of 0–0.5 M NaCl to recover the major protein peak. The fractions containing GIF were determined by SDS-PAGE and the immunoblotting using anti-hGIF Abs by the same procedures described previously (6). Purity of the GIFs obtained by this procedures was higher than 99%, as determined by SDS-PAGE. To remove endotoxin in the purified GIF preparations, each preparation was dialyzed against PBS, and mixed with a 1/10 volume of PyroSep C (Daicel Chemical Industries, Tokyo, Japan) for 1–12 h. The concentration of endotoxin in the final preparations, determined by Limulus ES-II Single Test (Wako Pure Chemical Industries, Osaka, Japan), was less than 1 ng of LPS/mg of GIF. The concentration of rhGIF and its mutated protein was determined by absorbance at 280 nm, using molar extinction coefficients that were calculated from amino acid composition of each species. The amino acid composition was obtained by hydrolysis of the sample for 24 h at 120°C in twice-distilled 6 M HCl containing 0.1% phenol. The hydrolyzed sample was derivatized with 6-aminoquinolyl-N-hydroxysuccinimidyl carbonate (AccQ Fluor, Waters, Milford, MA) and analyzed using amino acid analysis system (AccQ Tag Amino Acid Analysis System, Waters, Milford, MA).

Carboxymethylation of GIF

Wild-type rhGIF and C57A were treated with iodoacetate (Sigma) by the method previously described (8). Purified C57A/N106S at 1 mg/ml in 50 mM Tris-HCl (pH 8.2) was incubated for 30 min at room temperature with 18 mM iodoacetate and applied to a Sephadex G-25 column to remove excess iodoacetate.

Multiangle laser light-scattering analysis

The concentration of GIF was adjusted to 1.2 mg/ml, and 200 µl of the sample was applied to a Shodex KW803 column (Showa Denko, Tokyo, Japan) equilibrated with 0.1 M NaCl in 50 mM phosphate buffer (pH 6.8). Proteins were eluted at a flow rate of 1 ml/min, and the absolute m.w. of GIF in the eluate was determined by using WYATT DAWN DSP-F light-scattering photometer (Wyatt, Santa Barbara, CA) and refractive index detector (RI SE-6; Showa Denko). The m.w. was calculated by the mass method using the ASTRA software program (Wyatt).

Reverse-phase chromatography and mass spectrometry

The HPLC system for reverse-phase chromatography was ABI micro liquid chromatograph (PE Applied Biosystems, Thronhil, Canada) with a C18 column (Brownlee, 0.5 x 150 mm; PE Applied Biosystems). Affinity-purified samples were applied to the column and eluted at a flow rate of 10 µl/min with a linear gradient from 24–48% acetonitrile containing 0.08% trifluoroacetic acid.

Mass spectra were obtained by using matrix-assisted laser desorption/ionization (MALDI)-time of flight (TOF) mass spectrometer (Voyager Elite DE STR, PerSeptive Biosystems, Framingham, MA) equipped with a pulsed nitrogen laser (337 nm) for desorption and ionization. Sample preparation for the analysis was made by mixing 1 µl analyte with 1 µl matrix solution on stainless steel plates and air drying at ambient temperature. The matrix solution was 10 mg/ml sinapinic acid (Sigma) in 50% aqueous acetonitrile. The laser fluency was kept just above the threshold for observing the peaks. The acceleration voltage was 20 kV. A delayed pulsed ion-extraction device was installed to increase mass resolution. The vacuum of the TOF chamber was typically 1.5 x 10^-7 Torr. The signals of positive ions were collected, digitized, and averaged by a digital oscilloscope (TDS 520A; Tectronix, Wilsonville, OR) controlled by a laboratory computer.

Measurement of circular dichroism (CD) spectrum

The CD spectra of the wild-type rhGIF and its mutants in 10 mM sodium acetate buffer, pH 5, were obtained at 20°C using Jusco J-720 CD spectrophotometer (Jusco, Tokyo, Japan) equipped with a temperature control unit. Approximately 80 µg/ml (OD280 nm = 0.1) protein solution was put into a cuvette with 2-mm light path. The CD signal was recorded by scanning the wave length from 200–260 nm.

Detection of GIF bioactivity

GIF activity was detected by its ability to switch the mouse T cell hybridoma 12H5 cells (11) from the formation of glycosylated IgE-BF to the formation of unglycosylated IgE-BF. Detailed procedures for the assay were previously described (20). Briefly, the 12H5 cells were cultured with mouse IgE (10 µg/ml) in the presence or absence of a test sample, and IgE-BF in culture filtrates was fractionated on a lentil lectin-Sepharose (Pharmacia) column. When the 12H5 cells were cultured in the absence of GIF, essentially all IgE-BF formed by the cells bound to the column and was recovered by elution with 0.2 M methyl -D-mannoside. When the 12H5 cells were cultured in the presence of a sufficient amount of GIF, majority of the IgE-BF formed by the cells was not retained in the column and was recovered in the effluent fraction. To titrate bioactivity of a sample, aliquots of the 12H5 cells were cultured with serial twofold dilutions of the sample, and the bioactivity was expressed by the minimum concentration of the GIF peptide required for switching the cells for the formation of unglycosylated IgE-BF.

Immunization and Ab measurement

Groups of four or six female mice (C57B/6 x DBA/2)F₁ (BDF₁) (Nippon Charles River Laboratories, Kanagawa, Japan) were immunized by an i.p. injection of 0.1 µg DNP-OVA absorbed to 1 mg alum. rGIF or its mutant was injected i.p. into the mice under the schedules indicated.

For the measurement of anti-DNP IgE Ab, microtiter plates (Nunc, Roskilde, Denmark) were coated overnight with 1 µg/ml rat anti-mouse IgE mAb 6HD5 (17). After blocking with SuperBlock (Pirece, Pockford, IL) and washing with Tris-buffered saline, pH 8, containing 0.05% Tween-20 (TBST), appropriate dilutions (1/50 to 1/500) of serum samples in 50% FCS in TBST or serial dilutions of anti-DNP IgE Ab, H1 DNP--26, as a standard, were added to the plate. After overnight incubation at 4°C, plates were washed and 2.5 µg/ml DNP-BSA-biotin was added to each well. The plates were incubated for 1 h at 37°C, and DNP-BSA-biotin bound to the wells was measured by using alkaline-phosphatase-streptavidin conjugate (Zymed) and an alkaline-phosphatase colormerizing kit for amplification of colorimetric signal (AMPAK; Dako, Cambrigeshire, U.K.). Anti-DNP IgG1 Ab was measured using anti-DNP IgG1 mAb, H1 DNP--109, as a standard. Microtiter plates were coated with 20 µg/ml DNP-BSA. After blocking and washing with TBST, appropriate dilutions of a mouse antiserum or serial dilutions of the standard Ab in 50% FCS/TBST were added to the plate. After overnight incubation at 4°C, plates were washed, and 1 µg/ml alkaline phosphatase-coupled anti-mouse IgG1 mAb was added to each well. The plates were incubated for 90 min at room temperature, and the binding of the Ab was measured by using phosphatase substrate system (Kirkegard & Perry Laboratories, Gaitherburg, MD). Anti-DNP IgG2a and IgG2b Abs were measured in a similar manner using alkaline phosphatase-coupled rat anti-mouse IgG2a (Zymed) and anti-mouse IgG2b (Zymed). Standard wells of a microtiter plate were coated with serial dilutions of purified mouse IgG2a or IgG2b (PharMingen).

T cell proliferation assay

BDF₁ mice were immunized with 0.1 µg DNP-OVA absorbed to 1 mg alum and treated by i.p. injections of C57A/N106S or wild-type rhGIF. Splenic T cells of DNP-OVA-primed, GIF-treated mice were purified using mouse T cell-enriched columns (R&D Systems, Mineapolis, MN), and CD4⁺ T cells were obtained from the splenic T cells by negative selection. The cells were treated with biotinylated anti-CD8 mAb (PharMingen) and BioMag Streptavidin (Perceptive Diagnostics, Cambrige, MA). Purified CD4⁺ cells (1 x 10⁵/well) were cultured in triplicate in 96-well plates together with the irradiated (2000 rad) T-depleted normal spleen cells (5 x 10⁵/well) in the presence of varying concentrations of OVA. After 2 days of culture, 1 µCi [³H]thymidine (Amersham, Buckinghamshire, U.K.) was added to each well, and the cells were harvested 24 h later. Incorporation of [³H]thymidine was determined by scintillation counting.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Molecular characteristics of bioactive GIF and inactive GIF from Ts cells

Experiments were conducted to obtain direct evidence for posttranslational modifications of GIF peptide in Ts cells. Human Ts hybridoma, 31E9 cells were cultured for 48 h, and GIF in both culture supernatant and cytosol was isolated using anti-GIF-coupled HiTrap. Both GIF preparations gave a single 13-kDa band in SDS-PAGE (reducing conditions) and silver staining. The N-terminal amino acid sequence of GIF peptide from the two sources was identical, and lacked the first methionine in the deduced sequence (6). However, the two preparations were quite different in their bioactivity. A total of 10 ng/ml of GIF from the culture supernatant was sufficient for the detection of GIF bioactivity, whereas even 1 µg/ml of cytosolic GIF failed to show the activity. E. coli-derived rhGIF, employed as a control, also failed to show the activity at the level of 1 µg/ml. Thus, we determined molecular characteristics of the three GIF preparations. Both the rhGIF and cytosolic GIF yielded a single protein peak in reverse-phase column chromatography (Fig. 1A). Mass-spectrometric analysis showed that the protein has a m.w. of 12,346, which is identical to the theoretical value calculated from the amino acid sequence. In contrast, GIF from the culture supernatant gave four protein peaks in reverse-phase chromatography (Fig. 1B). Mass-spectrometric analysis of each protein peak showed that peaks 1, 2, 3, and 4 represented the m.w. species of 12,467, 12,346, 12,551, and 12,429, respectively. The proportion of the 12,346 species in total GIF was approximately 40%. Considering that hGIF genome has a single functional gene (21), the 12,429, 12,467, and 12,551 m.w. species would represent posttranslationally modified GIF.

View larger version (15K): [in this window] [in a new window]   FIGURE 1. Reverse-phase column chromatography of purified GIF from 31E9 cells. Cytosolic GIF (A), GIF from culture supernatant (B), and that treated with 10 mM DTT (C) were fractionated. Proteins were eluted with a linear gradient from 24–48% acetonitrile (---) containing 0.08% trifluoroacetic acid. Ordinate indicates absorption at 210 nm. Mass-spectrometric analysis of each peak showed that peaks 1, 2, 3, and 4 represent 12,467, 12,346, 12,551, and 12,429 m.w. species, respectively, while a small peak after peak 4 did not represent a single protein. Analysis of peak 3' in C indicated that this peak did not contain either 12,551 or 12,429 species.

Bioactivity of mutated rGIF

We speculated that conformational changes of GIF protein induced by posttranslational modifications are responsible for the generation of bioactivity (7). If this is the case, one might expect that an appropriate mutation of rhGIF may result in the generation of bioactivity. Since cysteine residues are involved frequently in maintaining tertiary structure of proteins (22), we constructed three mutants of rhGIF, in which cysteine at position 57, 60, or 81 was replaced with alanine. Each of the mutated GIF, i.e., C57A, C60A, or C81A, was expressed in E. coli, and purified by HPLC on a CM-5PW column (see Materials and Methods). The NaCl concentrations for elution of the wild-type rhGIF, C57A, C60A, and C81 were 0.30 M, 0.30 M, 0.30 M, and 0.21 M, respectively. In SDS-PAGE analysis under reducing conditions, no difference was detected among the four preparations. However, determination of bioactivity of the purified mutants showed that the replacement of Cys⁵⁷ or Cys⁶⁰ with Ala resulted in the generation of bioactivity, while the replacement of Cys⁸¹ with Ala failed to do so (Table I).

View larger version (34K): [in this window] [in a new window]   FIGURE 2. Ribbon diagram of GIF monomer. The helices are indicated as 1 and 2. ß strands are indicated from ß1 to ß6. Four ß strands (ß1, ß2, ß4, and ß5) in a monomer and the other two ß strands of adjacent monomers (ß3' and ß6'') form a ß sheet. The side chains of Cys57, Cys60, and Asn106 are indicated in a ball-and-stick model. This diagram was prepared using the program MOLSCRIPT (24).

We wondered that attenuation of trimer formation of C57A/N106S might be due to some changes in conformational structure by the mutation. Indeed, significant difference was detected between the mutant and wild-type rhGIF in their CD spectrum. As shown in Fig. 3, the far UV CD spectra of wild-type rhGIF and C57A exhibited a broad negative signal at about 215 nm, while the spectrum of the C57A/N106S showed a large negative peak at about 210 nm. Because of the differences in CD spectrum, we expected that C57A/N106S may have higher GIF bioactivity than C57A. However, these two mutants were comparable in their ability to switch the 12H5 cells from the formation of glycosylated IgE-BF to the formation of unglycosylated IgE-BF (Table I).

View larger version (18K): [in this window] [in a new window]   FIGURE 3. CD spectrum analysis of rGIF. CD spectra of wild-type rhGIF (——), C57A (---), and C57A/N106S (—·—) were analyzed by the method described in Materials and Methods.

Immunosuppressive effects of rGIF derivatives

Previous experiments have shown that repeated injections of partially purified GIF from Ts hybridomas suppressed both IgE and IgG Ab responses of BDF₁ mice to OVA (3). Thus, we evaluated possible immunosuppressive effects of mutated rhGIF and wild-type rhGIF on the IgE Ab response. BDF₁ mice were immunized with 0.1 µg alum-absorbed DNP-OVA, and the animals received six injections of 20 µg GIF or three injections of 10 µg GIF. Under the experimental conditions employed, the wild-type rhGIF did not suppress the IgE anti-DNP Ab response, while C57A/N106S and C57A/N106S-CM markedly suppressed the Ab response (Table II). To determine possible dose-response relationship, DNP-OVA-immunized mice were treated with 10, 2, or 0.4 µg of C57A/N106S-CM per injection. In this series, the GIF derivative was injected five times from day -1 to day 7. It was found that 2 µg C57A/N106S-CM/injection was sufficient for suppressing the IgE Ab response (p < 0.05). Similar experiments with C57A/N106S indicated that 2 µg/injection of the GIF mutant appears to be less effective (p = 0.08 and 0.09 in two experiments) than the same dose of C57A/N106S-CM in suppressing the IgE Ab response. The series of the experiments also showed that 0.4 µg/injection of C57A/N106S or its carboxymethylated derivative neither enhanced nor suppressed the Ab response.

View larger version (15K): [in this window] [in a new window]   FIGURE 4. Suppression of Ab responses by C57A/N106S. BDF1 mice were immunized with 0.1 µg alum-absorbed DNP-OVA. Ten micrograms of C57A/N106S () or PBS () were injected on days -1, 1, and 3 (shown by arrows), and the anti-DNP IgE and IgG1 Abs in their sera were measured by ELISA.

View larger version (20K): [in this window] [in a new window]   FIGURE 5. Suppressive effects of GIF on T cell priming and secondary response. Four groups of BDF1 mice were primed with 0.1 µg alum-absorbed UD-OVA on day 0, and boosted with 0.1 µg DNP-OVA on day 14 (shown by an arrow). Each group received daily injections of 10 µg of C57A/N106S (•, ) or PBS (, ) from day -1 to day 3 (A) or from day 13 to day 17 (B). The period of the treatment was indicated under the figure (|—|). The anti-DNP IgE (, •) and IgG1 (, ) were determined by ELISA.

View larger version (20K): [in this window] [in a new window]   FIGURE 6. Suppression of Ag priming of CD4+ T cell for proliferative response by GIF. Two groups of BDF1 mice were immunized with 0.1 µg alum-absorbed DNP-OVA. One group received daily injections of 10 µg C57A/N106S from day -1 to day 3 (), and control group received PBS (). Splenic CD4+ T cells were obtained on day 14, and these cells were stimulated with varying dose of OVA in the presence of irradiated syngeneic spleen cells.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Chemical nature of the posttranslational modifications is unknown. However, we speculate that one of the two modifications described above is cysteinylation of a cysteine residue (25), because 120.5 Da correspond to the mass of a half cystine. The x-ray crystal structure of rhGIF indicated conformational flexibility in the two helices and adjacent regions (23) (cf Fig. 2), and suggested that incorporation of a substance or a chemical group in the groove between the two helices in the process of posttranslational modifications may cause conformational changes that are responsible for the generation of bioactivity (8, 23). Indeed, previous experiments have shown that either carboxymethylation of, or binding of 5-thio-2-nitrobenzoic acid group to Cys⁶⁰ in the rhGIF resulted in generation of bioactivity (8). Considering the fact that both iodoacetate and DTNB were accessible to the SH group in Cys⁶⁰, but not to the Cys⁵⁷ or Cys⁸¹ under physiologic conditions, we speculate that Cys⁶⁰ is involved in cysteinylation in the posttranslational modification in Ts cells.

It was also found in the present experiments that the bioactivity could be generated by replacement of Cys⁵⁷ or Cys⁶⁰ in the rhGIF with Ala, and that carboxymethylation of Cys⁶⁰ and the replacement of Cys⁵⁷ with Ala had synergistic effect for increasing the bioactivity of GIF derivatives (cf Table I). Generation of the same bioactivity by various modifications of inactive rhGIF supports our hypothesis that conformational changes in the GIF molecules are responsible for the generation of bioactivity. However, wild-type GIF, C57A, and C60A did not show a significant difference in the CD spectrum, indicating that the replacement of Cys⁵⁷ or Cys⁶⁰ with Ala did not cause a drastic change in the secondary structure of GIF molecules (cf Fig. 3). We anticipate that replacement of Cys⁵⁷ with Ala and the binding of a chemical group to Cys⁶⁰ may cause a common change in the conformational structure, and that such conformational change is responsible for the generation of bioactivity.

Previous experiments have shown that the GIF protein in the culture supernatants of Ts hybridomas and a stable transfectant of hGIF cDNA in murine Ts cells are 12–14 kDa, as determined by gel filtration and by SDS-PAGE (nonreducing conditions) (7, 26, 27). Inactive cytosolic GIF as well is a 13-kDa protein (7). In contrast, E. coli-derived rhGIF is a trimer of the 13-kDa protein, as determined by light scattering. Since the E. coli-derived rGIF is inactive, we wondered that generation of the bioactivity by the mutation of rhGIF may be related to attenuation of intermolecular association. Indeed, determination of molecular size by light scattering indicates that replacement of Cys⁵⁷ with Ala slightly enhanced dissociation of the trimer, and replacement of Asn¹⁰⁶ with Ser in C57A further attenuated the interaction among GIF monomers. However, the GIF bioactivity of C57A did not increase by the AsnSer substitution (Table I), indicating that dissociation of the trimer is not responsible for the generation of bioactivity.

More recently, the equilibrium constant (K_A) for trimer formation of rhGIF, i.e., (molar concentration of trimer)/(molar concentration of monomer)³, was determined to be 4.2 x 10¹⁰ M^-2 by measuring the relative concentration of monomers by ultracentrifugal analysis. If one applies the K_A to various concentrations of rhGIF, it became obvious that more than 99.9% of rhGIF molecules in a 1 µg/ml (or less concentration) solution must be in the form of monomer. As the bioassay of GIF activity is being conducted using 0.01–1 µg/ml of GIF samples (cf Table I), bioactivity of GIF derivatives appears to represent the activity of the monomers. It is reasonable that attenuation of intermolecular interaction by mutation of rhGIF did not affect their bioactivity.

The present experiments showed that some of the bioactive derivatives of rhGIF suppressed both the IgE and IgG1 anti-DNP Ab responses of BDF₁ mice to DNP-OVA (Fig. 4). Although the results are not shown, treatment with C57A/N106S or C57A/N106S-CM suppressed not only the anti-hapten Ab responses, but also anti-carrier Ab responses to DNP-OVA. Among the derivatives tested in the present study, C57A/N106S-CM had the highest GIF bioactivity determined by in vitro assay (cf Table I), and showed the highest immunosuppressive effect on in vivo Ab response. However, the immunosuppressive effect of the bioactive rhGIF derivatives does not necessarily parallel their activity to switch the 12H5 cells from the formation of glycosylated IgE-BF to the formation of unglycosylated IgE-BF. The C57A/N106S had much higher immunosuppressive effect than C57A (Table II), while the two mutated rhGIF proteins were comparable with respect to their in vitro GIF bioactivity (Table I). The reasons for the lack of parallelism between the immunosuppressive effect and the in vitro GIF bioactivity are unknown. Recent experiments have shown that bioactive GIF from Ts cells and the bioactive rhGIF derivatives bound to both Th1 and Th2 clones with high affinity, while inactive rhGIF lacked affinity for these cells, and that C57A/N106S had much higher affinity than C57A for the target cells (28). Evidence was obtained that replacement of Cys⁵⁷ with Ala and the replacement of Asn¹⁰⁶ with Ser synergistically increased the affinity of the rhGIF molecules for the target cells, and that C57A/N106S and Ts-derived bioactive GIF were comparable in the affinity for the target cells (28). One might speculate that the difference between C57A/N106S and C57A in their immunosuppressive effect may be related to the difference in their affinity for the target cells.

It is known that the nucleotide sequence of the coding region of hGIF cDNA was almost identical to the sequence of human macrophage migration-inhibitory factor (MIF) cDNA (6, 29). However, biologic activities of the product of the cDNA are controversial. In our previous experiments, neither the affinity-purified bioactive murine GIF nor rhGIF inhibited migration of human monocytes or mouse macrophages (6). Herriott et al. (30) also reported that E. coli-derived rhMIF/GIF, at up to 5 µg/ml, failed to inhibit migration of macrophages. In contrast, Bernhagen et al. (31) reported that E. coli-derived mouse rMIF, which is identical to mouse GIF in its amino acid sequence (6), inhibited migration of macrophages, and Calandra et al. (32) demonstrated that the same recombinant peptide induced activation of macrophage cell line cells, RAW 264.7, for the formation of TNF-. However, previous experiments conducted in our laboratories failed to reproduce their findings. Sugie et al. (28) have shown that both the wild-type rhGIF and bioactive GIF derivatives, such as C57A/N106S, failed to activate RAW 264.7 cells for the formation of TNF-, and that none of three macrophage and monocyte cell line cells expressed receptors for either bioactive GIF or inactive GIF. We speculate that macrophages may not be the target of the bioactive GIF derivatives for immunosuppression.

The present experiments showed that C57A/N106S could suppress Ag priming of Th cells. In the experiment shown in Fig. 5A, treatment with the mutated rhGIF at the time of priming with UD-OVA suppressed the anti-DNP Ab responses to a booster injection of DNP-OVA. Previous experiments have shown that UD-OVA could prime Th cells specific for OVA, but failed to induce the formation of Abs specific for native OVA or to prime B cells specific for the native Ag (14). Since 0.1 µg DNP-OVA without adjuvant cannot induce the primary Ab response in BDF₁ mice, it is obvious that OVA-specific Th cells, which had been primed by immunization with alum-absorbed UD-OVA, were involved in the anti-hapten Ab responses. Thus, suppression of the anti-hapten Ab responses of UD-OVA-primed mice to DNP-OVA must be due to the suppression of Ag priming of Th cells by C57A/N106S. However, treatment of the UD-OVA-primed mice with the same GIF derivative at the time of booster immunization with DNP-OVA was also effective in suppressing the anti-hapten Ab response (Fig. 5B). Sugie et al. (28) have shown that bioactive GIF and bioactive rhGIF derivatives, such as C57A/N106S, have high affinity for both Th1 and Th2 clones, but not for naive T cells. Their more recent experiments indicated that not only Ag-primed T cells, but also activated B cells express high affinity receptors for bioactive GIF, and that a bioactive rGIF derivative suppressed the formation of IgG1 and IgE by B cells stimulated with LPS and IL-4 (Sugie et al., unpublished results). One might speculate that C57A/N106S suppressed the Ab responses through binding to Ag-primed T cells and/or B cells. Previous experiments have provided evidence that Ag-specific TsF (GIF) is a posttranslationally formed conjugate of TCR with the 13-kDa GIF (4, 5). In Ts cells, GIF peptide is translocated into endoplasmic reticulum upon antigenic stimulation, forms a conjugate with TCR-chain, and facilitates the release of a soluble form TCR from the cells (33). Immunosuppressive activity of the 13-kDa bioactive rhGIF derivative, observed in the present experiment, revealed an additional role of GIF as a functional subunit of TsF.

	Acknowledgments

 
We thank Yuko Takeuchi, Naomi Kubota, and Nobuko Kubomura for technical assistance.

	Footnotes

 
¹ Address correspondence and reprint requests to Dr. Kimishige Ishizaka, Pharmaceutical Research Laboratory, Kirin Brewery Co. Ltd., 3 Miyahara, Takasaki, Gunma 371-1295, Japan.

² Abbreviations used in this paper: GIF, glycosylation-inhibiting factor; alum, aluminum hydroxide gel; CD, circular dichroism; DTNB, 5,5' dithiobis (2-nitrobenzoic acid); h, human; IgE-BF, IgE binding factor; MIF, macrophage migration-inhibitory factor; Ts, suppressor T; TsF, suppressor T cell factors; UD, urea denatured.

Received for publication December 1, 1997. Accepted for publication September 15, 1998.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Ishizaka, K.. 1984. Regulation of IgE synthesis. Annu. Rev. Immunol. 2:159.[Medline]
2. Jardieu, P., T. Uede, K. Ishizaka. 1984. IgE-binding factors from mouse T lymphocytes. III. Role of antigen-specific suppressor T cells in the formation of IgE-suppressive factor. J. Immunol. 133:3266.[Abstract]
3. Akasaki, M., P. Jardieu, K. Ishizaka. 1986. Immunosuppressive effects of glycosylation inhibiting factor on the IgE and IgG response. J. Immunol. 136:3172.[Abstract]
4. Nakano, T., Y. Ishii, K. Ishizaka. 1996. Biochemical characterization of antigen-specific glycosylation-inhibiting factor from antigen-specific suppressor T cells. I. Identification of a 55-kilodalton glycosylation-inhibiting factor peptide with TCR -chain determinant. J. Immunol. 156:1728.[Abstract]
5. Ishii, Y., T. Nakano, K. Ishizaka. 1996. Biochemical characterization of antigen-specific glycosylation-inhibiting factor from antigen-specific suppressor T cells. II. The 55-kDa glycosylation-inhibiting factor peptide is a derivative of TCR -chain and a subunit of antigen-specific glycosylation-inhibiting factor. J. Immunol. 156:1735.[Abstract]
6. Mikayama, T., T. Nakano, H. Gomi, Y. Nakagawa, Y. C. Liu, M. Sato, A. Iwamatsu, Y. Ishii, W. Y. Weiser, K. Ishizaka. 1993. Molecular cloning and functional expression of a cDNA encoding glycosylation-inhibiting factor. Proc. Natl. Acad. Sci. USA 90:10056.[Abstract/Free Full Text]
7. Liu, Y.-C., T. Nakano, C. Elly, K. Ishizaka. 1994. Requirement of posttranslational modifications for the generation of biologic activity of glycosylation-inhibiting factor. Proc. Natl. Acad. Sci. USA 91:11227.[Abstract/Free Full Text]
8. Nakano, T., H. Watarai, Y. C. Liu, Y. Oyama, T. Mikayama, K. Ishizaka. 1997. Conversion of inactive glycosylation inhibiting factor to bioactive derivatives by modification of a SH group. Proc. Natl. Acad. Sci. USA 94:202.[Abstract/Free Full Text]
9. Frankel, A. D., D. S. Bredt, C. O. Pabo. 1988. Tat protein from human immunodeficiency virus forms a metal-linked dimer. Science 240:70.[Abstract/Free Full Text]
10. Green, S., P. Chambon. 1987. Oestradiol induction of a glucocorticoid-responsive gene by a chimeric receptor. Nature 325:75.[Medline]
11. Iwata, M., M. Adachi, K. Ishizaka. 1988. Antigen-specific T cells that form IgE-potentiating factor, IgG-potentiating factor and antigen-specific glycosylation enhancing factor on antigen stimulation. J. Immunol. 140:2534.[Abstract]
12. Jardieu, P., M. Akasaki, K. Ishizaka. 1987. Carrier-specific suppression of antibody responses by antigen-specific glycosylation inhibiting factor. J. Immunol. 138:1494.[Abstract]
13. Shimonkevitz, R., J. Kappler, P. Marrach, H. Grey. 1981. Antigen recognition by H-2 restricted T cells. I. Cell-free antigen processing. J. Exp. Med. 158:303.[Abstract/Free Full Text]
14. Takatsu, K., K. Ishizaka. 1975. Reaginic antibody formation in the mouse. IV. Suppression of IgE and IgG antibody responses to ovalbumin following the administration of high dose urea-denatured antigen. Cell. Immunol. 20:276.[Medline]
15. Liu, F. T., J. W. Bohn, E. L. Ferry, H. Yamamoto, C. A. Molinaro, L. A. Sherman, N. R. Klinman, D. H. Katz. 1980. Monoclonal dinitrophenyl-specific murine IgE antibody: preparation, isolation, and characterization. J. Immunol. 124:2728.[Medline]
16. Vander-Mallie, R., T. Ishizaka, K. Ishizaka. 1982. Lymphocytes bearing Fc receptors for IgE VIII: affinity of mouse IgE for FcR on mouse B lymphocytes. J. Immunol. 128:2306.[Abstract]
17. Hirano, T., H. Miyajima, H. Kitagawa, N. Watanabe, M. Azuma, O. Taniguchi, H. Hashimoto, S. Hirose, H. Yagita, S. Furusawa, Z. Ovary, K. Okumura. 1988. Studies on murine IgE with monoclonal antibodies. I. Characterization of rat monoclonal anti-IgE antibodies and the use of these antibodies for determinations of serum IgE levels and for anaphylactic reactions. Int. Arch. Allergy Appl. Immunol. 85:47.[Medline]
18. Laemmli, U. K.. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680.[Medline]
19. Nakano, T., Y. C. Liu, T. Mikayama, H. Watarai, M. Taniguchi, K. Ishizaka. 1995. Association of the "major histocompatibility complex subregion" I-J determinant with bioactive glycosylation-inhibiting factor. Proc. Natl. Acad. Sci. USA 92:9196.[Abstract/Free Full Text]
20. Iwata, M., K. Ishizaka. 1988. Construction of antigen-specific suppressor T cell hybridomas from spleen cells of mice primed for the persistent IgE antibody formation. J. Immunol. 141:3270.[Abstract]
21. Paralkar, V., G. Wistow. 1994. Cloning the human gene for macrophage migration inhibitory factor (MIF). Genomics 19:48.[Medline]
22. Kim, P. S., R. L. Balswin. 1990. Intermediates in the folding reactions of small proteins. Annu. Rev. Biochem. 59:631.[Medline]
23. Kato, Y., T. Muto, T. Tomura, H. Tsumura, H. Watarai, T. Mikayama, K. Ishizaka, R. Kuroki. 1996. The crystal structure of human glycosylation inhibiting factor is a trimmer barrel with three 6-stranded ß-sheets. Proc. Natl. Acad. Sci. USA 93:3007.[Abstract/Free Full Text]
24. Kraulis, P. J.. 1991. MOLSCRIPT: a program to produce both detailed and schematic plots of protein structures. J. Appl. Crystallogr. 24:946.
25. Dörmann, P., T. Börchers, U. Korf, P. Højrup, P. Roepstorff, F. Spener. 1993. Amino acid exchange and covalent modification by cysteine and glutathione explain isoforms of fatty acid-binding protein occurring in bovine liver. J. Biol. Chem. 268:16286.[Abstract/Free Full Text]
26. Tagaya, Y., A. Mori, K. Ishizaka. 1991. Biochemical characterization of murine glycosylation-inhibiting factor. Proc. Natl. Acad. Sci. USA 88:9117.[Abstract/Free Full Text]
27. Thomas, P., H. Gomi, T. Takeuchi, C. Carini, Y. Tagaya, K. Ishizaka. 1992. Glycosylation-inhibiting factor from human T cell hybridomas constructed from peripheral blood lymphocytes of a bee venom-sensitive allergic patient. J. Immunol. 148:729.[Abstract]
28. Sugie, K., T. Nakano, T. Tomura, K. Takakura, T. Mikayama, K. Ishizaka. 1997. High-affinity binding of bioactive glycosylation-inhibiting factor to antigen-primed T cells and natural killer cells. Proc. Natl. Acad. Sci. USA 94:5278.[Abstract/Free Full Text]
29. Weiser, W. Y., P. A. Temple, J. S. Witek-Gianotti, H. G. Remold, S. C. Clark, J. R. David. 1989. Molecular cloning of a cDNA encoding a human macrophage migration inhibitory factor. Proc. Natl. Acad. Sci. USA 86:7522.[Abstract/Free Full Text]
30. Herriott, M. J., H. Jiang, C. A. Stewart, D. J. Fast, R. W. Leu. 1993. Mechanistic differences between migration inhibitory factor (MIF) and IFN- synergize with lipid A to mediate migration inhibition but only IFN- induces production of TNF- and nitric oxide. J. Immunol. 150:4524.[Abstract]
31. Bernhagen, J., R. A. Mitchell, T. Calandra, W. Voelter, A. Cerami, R. Bucala. 1994. Purification, bioactivity, and secondary structure analysis of mouse and human macrophage migration inhibitory factor (MIF). Biochemistry 33:14144.[Medline]
32. Calandra, T., J. Bernhagen, R. A. Mitchell, R. Bucala. 1994. The macrophage is an important and previously unrecognized source of macrophage migration inhibitory factor. J. Exp. Med. 179:1895.[Abstract/Free Full Text]
33. Ishii, Y., T. Nakano, K. Ishizaka. 1996. Cellular mechanisms for the formation of a soluble form derivative of T-cell receptor chain by suppressor T cells. Proc. Natl. Acad. Sci. USA 93:7207.[Abstract/Free Full Text]

This article has been cited by other articles:

				A. Kudrin, M. Scott, S. Martin, C.-w. Chung, R. Donn, A. McMaster, S. Ellison, D. Ray, K. Ray, and M. Binks Human Macrophage Migration Inhibitory Factor: A PROVEN IMMUNOMODULATORY CYTOKINE? J. Biol. Chem., October 6, 2006; 281(40): 29641 - 29651. [Abstract] [Full Text] [PDF]

				K. Sugie and J. Huang GIF Inhibits Th Effector Generation by Acting on Antigen-Presenting B Cells J. Immunol., April 1, 2001; 166(7): 4473 - 4480. [Abstract] [Full Text] [PDF]

				H. Watarai, R. Nozawa, A. Tokunaga, N. Yuyama, M. Tomas, A. Hinohara, K. Ishizaka, and Y. Ishii Posttranslational modification of the glycosylation inhibiting factor (GIF) gene product generates bioactive GIF PNAS, November 2, 2000; (2000) 230445397. [Abstract] [Full Text]

				K. Kasahara, T. Nakano, H. Takahashi, Y. Ishii, K. Ishizaka, and K. Imai Presence of the 55 kDa glycosylation inhibiting factor in human serum Int. Immunol., September 1, 2000; 12(9): 1303 - 1309. [Abstract] [Full Text] [PDF]

				T. Kobayashi, T. Miura, T. Haba, M. Sato, I. Serizawa, H. Nagai, and K. Ishizaka An Essential Role of Mast Cells in the Development of Airway Hyperresponsiveness in a Murine Asthma Model J. Immunol., April 1, 2000; 164(7): 3855 - 3861. [Abstract] [Full Text] [PDF]

				H. Watarai, R. Nozawa, A. Tokunaga, N. Yuyama, M. Tomas, A. Hinohara, K. Ishizaka, and Y. Ishii Posttranslational modification of the glycosylation inhibiting factor (GIF) gene product generates bioactive GIF PNAS, November 21, 2000; 97(24): 13251 - 13256. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Tomura, T. Articles by Ishizaka, K. Search for Related Content PubMed