Glycobiology

This Article Abstract FREE Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (9) Request Permissions Disclaimer Google Scholar Articles by Wester, L. Articles by Åkerström, B. Search for Related Content PubMed PubMed Citation Articles by Wester, L. Articles by Åkerström, B. Social Bookmarking          What's this?

Glycobiology, 2000, Vol. 10, No. 9 891-900
© 2000 Oxford University Press

Carbohydrate groups of ₁-microglobulin are important for secretion and tissue localization but not for immunological properties

Lena Wester, Jonas Fast, Tord Labuda², Tommy Cedervall, Karin Wingårdh³, Tor Olofsson⁴ and Bo Åkerström¹

Section for Molecular Signalling, Department of Cell and Molecular Biology, Lund University, Lund, Sweden, ²Section for Tumour Immunology, Department of Cell and Molecular Biology, Lund University, Lund, Sweden, ³Department of Radiation Physics, The Jubileum Institute, Lund University, Lund, Sweden, and ⁴Haematology Research Laboratory, Department of Laboratory Medicine, Lund University, Lund, Sweden

Received on February 7, 2000; accepted on February 29, 2000.

	Abstract

Top Abstract Introduction Results Discussion Materials and methods Acknowledgments References

 
The role of the carbohydrates for the structure and functions of the plasma and tissue protein ₁-microglobulin (₁m) was investigated by deletion of the sites for N-glycosylation by site-directed mutagenesis (N17,96Q). The mutated cDNA was expressed in a baculovirus-insect cell system resulting in a nonglycosylated protein. The biochemical properties of N17,96Q-₁m were compared to nonmutated ₁m, which carries two short non-sialylated N-linked oligosaccharides when expressed in the same system. Both proteins carried a yellow-brown chromophore and were heterogeneous in charge. Circular dichroism spectra and antibody binding indicated a similar overall structure. However, the secretion of N17,96Q-₁m was significantly reduced and 75% of the protein were found accumulated intracellularly. The in vitro immunological effects of recombinant nonmutated ₁m and N17,96Q-₁m were compared to the effects of ₁m isolated from plasma, which is sialylated and carries an additional O-linked oligosaccharide. All three ₁m variants bound to human peripheral lymphocytes and mouse T cell hybridomas to the same extent. They also inhibited the antigen-stimulated proliferation of peripheral lymphocytes and antigen-stimulated interleukin 2-secretion of T cell hybridomas in a similar manner. After injection of rats intravenously, the blood clearance of recombinant nonmutated and N17,96Q- ₁m was faster than that of plasma ₁m. Nonmutated ₁m was located primarily to the liver, most likely via binding to asialoglycoprotein receptors, and N17,96Q-₁m was located mainly to the kidneys. It is concluded that the carbohydrates of ₁m are important for the secretion and the in vivo turnover of the protein, but not for the structure or immunological properties.

Key words: ₁-microglobulin/carbohydrate deletion/immune suppression/secretion/tissue distribution

	Introduction

Top Abstract Introduction Results Discussion Materials and methods Acknowledgments References

 
₁-microglobulin (or protein HC) is a 26 kDa protein which has been isolated and characterized from plasma, liver and urine from several species including humans and plaice (Ekström et al., 1975; Tejler and Grubb, 1976; Lindqvist and Åkerström, 1999). The ₁m-polypeptide is glycosylated (Escribano et al., 1990) and covalently linked to yellow-brown chromophores of unknown structure (Berggård et al., 1999; Escribano et al., 1991). ₁m is a member of the lipocalin protein superfamily (Flower, 1996), a group of extracellular proteins, which are folded into an eight-stranded ß-barrel, inside which a hydrophobic ligand usually is fitted. Such a ligand for ₁m has not yet been identified.

The exact biological function of ₁m is not known, but it has a number of in vitro immunoregulatory properties. For example, it inhibits the antigen-induced proliferation of peripheral lymphocytes, IL-2 production of T-cells and migration and chemotaxis of granulocytes (Lögdberg and Åkerström, 1981; Méndez et al., 1986; Wester et al., 1998). Different antigen-models in several different species have been used, e.g., tuberculin (purified protein derivative; PPD), tetanus toxid, ovalbumin and collagen, suggesting that ₁m has a general immunosuppressive role. It was also demonstrated that ₁m binds to receptors on peripheral lymphocytes and T-cell hybridomas (Fernández-Luna et al., 1988; Åkerström and Lögdberg, 1984; Babiker-Mohamed et al., 1990a; Wester et al., 1998), but the connection between cell-surface binding and the immunosuppressive effects is not yet known. The antigen-induced proliferation of peripheral lymphocytes could be suppressed by a preparation of N-linked glycans from ₁m as well as by the whole protein (Åkerström and Lögdberg, 1984). The carbohydrates of ₁m were then suggested to exert the immunological effects.

₁m is co-synthesized in the liver together with bikunin (Kaumeyer et al., 1986), a proteinase inhibitor which is cross-linked to one or two out of four so called heavy chains forming the various members of the inter--inhibitor family (Salier et al., 1996). A precursor protein, ₁m-bikunin, is formed in the hepatocytes but cleaved in the Golgi apparatus before secretion of ₁m and bikunin separately (Bratt et al., 1993). The two proteins have no known functional or structural relation after the cleavage, so the reason for the co-synthesis of the two proteins is enigmatic, but since it has been found in all species so far investigated, it is apparently of some importance.

In plasma, ₁m is found in free form as well as covalently bound to other larger plasma proteins. Complexes with IgA, albumin, and prothrombin have been described in human plasma (Grubb et al., 1986; Berggård et al., 1997) and with fibronectin and ₁-inhibitor-3, an ₂-macroglobulin homologue, in rat plasma (Falkenberg, 1990; Falkenberg et al., 1994). Free ₁m and various high-molecular weight complexes are also present in extracellular matrix of most tissues, probably originating from plasma. Its localization in placenta in areas of contact between mother and fetus, especially at sites of injury, suggests that it is involved in local immunosuppression, protecting tissues from potentially dangerous inflammatory processes (Berggård et al., 1999).

In this work we have investigated the role of the carbohydrates of ₁m for expression, secretion, structure, biochemical properties, lymphocyte binding, immunosuppressive effects and in vivo turnover of the protein. Using the baculovirus-insect cell expression system, two recombinant forms of ₁m were obtained and one form of ₁m was isolated from human plasma (Figure 1). Plasma ₁m carries three oligosaccharides: one O-linked at pos. 5 (T5) and two N-linked at positions 17 and 96 (N17 and N96). Their exact structures have not been elucidated but were shown to contain sialic acid (Ekström et al., 1981; Escribano et al., 1990). Recombinant nonmutated ₁m, expressed and purified from baculovirus-infected insect cells, lacked the O-linked oligosaccharide on T5 and the N-linked oligosaccharides were smaller and lacked sialic acid (Wester et al., 1997). In this work, N17,96Q-₁m, a completely carbohydrate-free form of the protein, carrying glutamine instead of asparagine in positions 17 and 96, was made using site-directed mutagenesis and expression in an insect cell system. First, the efficiency of expression and secretion and the biochemical properties of the carbohydrate-free N17,96Q-₁m was examined and compared to nonmutated recombinant ₁m. Second, all three variants were tested with respect to lymphocyte binding and immunosuppressive effects. Third, the radiolabeled proteins were injected into rats intravenously and their distribution in blood and tissues were studied. These results suggest that the carbohydrates are important for the secretion from insect cells and protein turnover in vivo, but are less important for the structure and biochemical properties of the polypeptide chain. The results also show that the lymphocyte binding and the suppressive effects of ₁m on antigen-stimulation are most likely not dependent on the carbohydrates.

View larger version (15K): [in this window] [in a new window]   Fig. 1. Schematic representation of three variants of human 1m analyzed in this article. Plasma 1m is the wild-type form, which contains three glycosylation sites at T5, N17 and N96. A sialylated O-linked oligosaccharide is found at T5 (solid circles) and N-linked sialylated oligosaccharides are found at N17 and N96 (solid diamonds). Recombinant nonmutated 1m is expressed in insect cells infected by baculovirus carrying DNA coding for human 1m. No oligosaccharide is linked to T5 and the N-linked oligosaccharides are smaller than on plasma 1m and lacks sialic acid (open diamonds). N17,96Q-1m is expressed in insect cells infected by baculovirus carrying DNA in which the codons for N17 and N96 have been changed to codons for Q. This protein completely lacks O-linked and N-linked oligosaccharides.

 

	Results

Top Abstract Introduction Results Discussion Materials and methods Acknowledgments References

 
Expression and secretion
Hi-5 insect cells were infected with baculovirus coding for ₁m or N17,96Q-₁m using identical conditions. A moi of 10 which was found to be optimal for nonmutated recombinant ₁m (Wester et al., 1997) was also optimal for N17,96Q-₁m. The medium was investigated daily by RIA, SDS–PAGE and Western blotting. Nonmutated ₁m was found as two bands, 27 and 29 kDa, in agreement with previous observations (Wester et al., 1997), whereas N17,96Q-₁m was secreted as a 24 kDa band (Figure 2). As shown below, the 27 and 29 kDa bands represent ₁m with one and two N-glycans, respectively, and the 24 kDa band is ₁m without carbohydrate. For both forms, maximal concentration of ₁m secreted to the medium was reached after 4 days of incubation and prolonged times resulted in degradation. However, the concentration of N17,96Q-₁m in the medium was several-fold lower than for nonmutated ₁m. Lysates of the infected cells were analyzed on a daily basis by SDS–PAGE. High concentrations of nonglycosylated 24 kDa ₁m were seen intracellularly (Figure 2), suggesting that N17,96Q-₁m was translated but not secreted to the same extent as nonmutated ₁m. Based on protein staining approximately 25 % of the total amount of N17,96-₁m were secreted compared to almost 100 % of nonmutated ₁m. Western blotting showed that polyclonal anti-₁m antibodies bound much weaker to intracellular nonglycosylated ₁m than to intracellular glycosylated ₁m or both secreted forms of ₁m (not shown). This suggests misfolding of nonglycosylated ₁m intracellularly. The misfolding and decreased secretion could be caused by the NQ mutation or by the lack of carbohydrates. To investigate the second possibility, the expression of nonmutated ₁m was studied in the presence of the N-glycosylation inhibitor tunicamycin. As shown in Figure 2, large amounts of a 24 kDa band, corresponding to ₁m, were found intracellularly and small amounts in the medium. Thus, inhibition of N-glycosylation of ₁m gave similar results, as did mutation of the glycosylation sites, suggesting that the decreased secretion was caused by the lack of carbohydrates rather than the NQ exchange. The conclusion of these results is that glycosylation is important for an optimal folding and secretion of ₁m at least in insect cells.

View larger version (66K): [in this window] [in a new window]   Fig. 2. The amount of 1m inside the insect cells and the amount of 1m secreted to the media on different days after inoculation were analyzed by SDS–PAGE (T = 12%, C = 3.75%). Hi-5 insect cells, 1.5 x 106, were inoculated with 10 pfu/cell of nonmutated 1m-virus or N17,96Q-1m-virus and cultured in the presence or absence of 10 µg/ml of tunicamycin at 27°C. Each day between days 1 and 6 after inoculation, cells and media were harvested and separated by low-speed centrifugation. Protease inhibitors were added to the media and the cell-pellets were dissolved in 100 µl of PBS and lysed by incubation with 25 µl of lysis-buffer (50 mM Tris–HCl, pH 6.9, 25% glycerol, 10% SDS, 25% ß-mercaptoethanol, and 0.25% bromphenolblue) at 100°C for 5 min. 10 µl of cell media and 2.5 µl of the cell lysate from each day were analyzed by SDS–PAGE and Coomassie staining (intracellular panel) or Western blotting with anti-1m antibodies (extracellular panel).

 
Purification
N17,96Q-₁m was purified from the medium of large-scale 4 day cultivations of Hi-5 cells infected with baculovirus coding for N17,96Q-₁m. The medium was applied first to a monoclonal anti-₁m Affigel column and then to a Sephacryl S-200 column. The results of the different purification steps are shown in Figure 3A. Pure 24 kDa nonglycosylated ₁m was freeze-dried.

View larger version (61K): [in this window] [in a new window]   Fig. 3. (A) SDS–PAGE (T=12%, C=3.75 %) and protein staining of recombinant N17,96Q-1m at different steps of the purification. The cell medium was harvested, protease inhibitors added, cell debris removed by low-speed centrifugation and extracellular virus removed by ultracentrifugation. The supernatant (lane 1) was applied to an anti-1m affinity column. The eluate (lane 2) was concentrated and separated by Sephacryl S-200 gel chromatography. 1m containing fractions were pooled, dialyzed and lyophilized (lane 3). (B) SDS–PAGE (T = 12%, C = 3.75%) of carbohydrate digested 1m. Recombinant 1m (lanes 1–4) and N17,96Q-1m (lane 5–8) were sequentially digested by neuraminidase (lanes 2 and 6), O-glycosidase (lane 3 and 7) and N-glycosidase F (lanes 4 and 8). Approximately 1.5 µg of the digestion products together with the same amount of uncleaved 1m (lanes 1 and 5) were separated by SDS–PAGE (T = 12%, C = 3.75 %) and protein stained.

 
Carbohydrate content
The carbohydrate content of the recombinant nonmutated and N17,96Q-₁m was investigated by lectin blotting and glycosidase digestion. The lectins PNA and Con A were negative for N17,96Q-₁m. Digestion of N17,96Q-₁m with neuraminidase, O-glycosidase and N-glycosidase F gave no size-reduction, whereas nonmutated ₁m was size-reduced by digestion with N-glycosidase F but not neuraminidase or O-glycosidase (Figure 3B). Moreover, the size of N-glycosidase-digested nonmutated ₁m was the same as N17,96Q-₁m (Figure 3B, lane 4). The results are compatible with a lack of both O-linked and N-linked carbohydrates on N17,96Q-₁m but the presence of two N-linked oligosaccharides on nonmutated ₁m. The two N-linked oligosaccharides on insect cell ₁m were previously shown to be smaller than their counterparts on ₁m isolated from human plasma or urine, and also to lack sialic acid residues which are present on plasma and urinary ₁m (Wester et al., 1997).

Size, shape, charge, and color
Some biochemical and physicochemical properties of nonmutated and N17,96Q-₁m were analyzed and are displayed in Table I. The N-terminal sequences were identical. As described above, the molecular mass was lower for nonglycosylated, N17,96Q-₁m, 24,000 Da as compared to 29,000 Da for the glycosylated, nonmutated ₁m. The Stokes’ radius and frictional ratio were similar for the two proteins. The secondary structure of the proteins was studied by far UV CD-analysis. The shape of the CD spectra obtained (Figure 4C) were similar to previously reported spectra for urinary, recombinant and plasma ₁m (Gavilanes et al., 1984; Wester et al., 1997), indicating no major shift in the overall secondary structure of nonglycosylated ₁m.

View this table: [in this window] [in a new window]   Table I. Physicochemical properties of recombinant nonmutated 1m (N-glycosylated) and N17,96Q-mutated 1m (nonglycosylated)

 

View larger version (46K): [in this window] [in a new window]   Fig. 4. (A) Absorbance spectra of recombinant nonmutated 1m and N17,96Q-1m. Lyophilized 1m was dissolved in PBS at a concentration of 4.5 x 10–6 M, as determined by weighing. The absorbance was measured every third nanometer between 250 and 300 nm and every fifth nanometer between 300 and 450 nm. The background absorbance of PBS was subtracted before plotting the data. (B) Agarose gel electrophoresis of recombinant nonmutated 1m and N17,96Q 1m. Approximately 40 µg of native protein was applied to an agarose gel in each lane. Human serum was used as a standard. (C) CD spectra of recombinant nonmutated 1m and N17,96Q-1m in the far-UV region (200–260 nm) at 37°C. The spectra were obtained at 0.5 mg/ml in 20 mM sodium-phosphate pH 7.0.

 
Both forms of ₁m were negatively charged and strongly charge-heterogeneous, as determined by agarose electrophoresis (Figure 4B). N17,96Q-₁m was more negatively charged but the heterogeneity was not altered. This suggests that a non-carbohydrate component is responsible for the charge heterogeneity of the protein and is present in slightly higher amounts in N17,96Q-₁m. A yellow-brown, unidentified chromophore was present in both nonmutated and N17,96Q-₁m, resulting in characteristic absorbance spectra with only minor differences (Figure 4A). The absorbance coefficients at 280 and 330 nm were calculated and are shown in Table I. The ratio _330nm/_280nm indicate that N17,96Q-₁m is slightly more brown colored than nonmutated ₁m.

Antibody binding
Surface epitopes on ₁m were investigated by RIA, a competitive binding assay, and SPRIA, a direct-binding assay (Figure 5, A and B, respectively). The binding of polyclonal and monoclonal antibodies to recombinant nonmutated and N17,96Q- ₁m was compared to human plasma and urinary ₁m. All antibodies were prepared against human urinary ₁m, and the monoclonal antibodies have been demonstrated to recognize non-carbohydrate structures only (Babiker-Mohamed et al., 1991). As shown in Figure 5, similar results were obtained with the four ₁m variants in RIA and the three ₁m variants tested in SPRIA, suggesting that the surface epitopes are not structurally altered on N17,96Q-₁m.

View larger version (20K): [in this window] [in a new window]   Fig. 5. (A) RIA of urinary 1m (open square), plasma 1m (filled square), recombinant nonmutated 1m (open circle) and recombinant N17,96Q1m (filled circle). Dilution series in duplicates of the 1m variants were added to a fixed amount of 125I-labeled urinary 1m and antibodies against urinary 1m. The 1m-antibody complexes were precipitated and analyzed for radioactivity content. (B) Binding of mouse monoclonal and rabbit polyclonal anti-1m antibodies to dilution series of urinary 1m (open circles), recombinant nonmutated 1m (open squares) and N17,96Q-1m (solid circles) measured by SPRIA. Dilution series of the three 1m-forms were coated on microtiter plates and then incubated with either of the monoclonal anti-1m antibodies (MAb 1, 2, 3, 5, or 10) or the polyclonal anti-1m antibody (K:324). The plates were finally incubated with radiolabeled detection antibodies. The assay was made in duplicates.

 
Lymphocyte binding
The binding of recombinant nonmutated and N17,96Q- ₁m and plasma ₁m to the mouse T cell line HCQ.4 and to various peripheral human lymphocytes were analyzed by flow cytometry. The results are shown in Figure 6. All three ₁m-variants bound T cells (HCQ.4, CD4+, CD8+), B-cells (CD19+), and NK-cells (CD56+) to a similar degree. Minor differences between the three forms of ₁m could be seen, but these differences were small and varied to some degree between different experiments. In some cases, as for N17,96Q to CD19+ cells (Figure 6), a bimodal binding pattern was seen suggesting two subpopulations of CD19+ B-lymphocytes with different affinity for ₁m. Thus, the results indicate that the carbohydrate part of ₁m is of minor importance for the binding to lymphocytes.

View larger version (42K): [in this window] [in a new window]   Fig. 6. Flow cytometric analysis of the binding of 1m variants to lymphocytes. Cells were incubated with plasma 1m (dotted line), recombinant nonmutated 1m (unfilled peak and bold line) or recombinant N17,96Q-1m (shaded peak), followed by the monoclonal anti-1m antibody BN 11.3, and finally with a FITC-conjugated goat-anti-mouse antibody. To set the background, the cells were incubated with BSA in the first step (unfilled peak and thin line). A mouse T cell hybridoma, HCQ.4, or human PBL were analyzed. After the incubation with the FITC-conjugated antibody, the human PBL were washed carefully and incubated with phycoerythrin-conjugated mouse anti-human CD4, CD8, CD19, or CD56 antibodies. Plasma 1m was not used in the case of HCQ.4 cells. Three experiments were performed and the result of one of these is shown.

 
Inhibition of lymphocyte immune functions
It has been shown previously that ₁m inhibits the IL-2 secretion of antigen-stimulated mouse T cells and the proliferation of antigen (PPD)-stimulated human peripheral blood leukocytes (PBL). The effects of the three ₁m variants on these two assays were compared (Figure 7). The mouse T cell line HCQ.4 was stimulated with its antigen, a rat collagen peptide, and the resulting IL-2 secretion was inhibited by the different ₁m-variants but not by the control protein 1-acid glycoprotein (not shown). One representative experiment out of three is shown in Figure 7A. The results indicate that ₁m concentrations between 8 and 125 µg/ml inhibit the IL-2 production in a dose dependent manner, which is in agreement with previous observations (Wester et al., 1998). No significant difference in inhibition potential can be detected between N17,96Q- and recombinant nonmutated ₁m. Similar results were obtained with PPD-stimulated human PBL (Figure 7B). The proliferation was inhibited to the same extent and with equal dose-response by all three ₁m variants. These data indicate that the carbohydrates of ₁m are not important for the inhibition of immune response of lymphocytes.

View larger version (22K): [in this window] [in a new window]   Fig. 7. (A) Inhibition of IL-2 production by antigen-stimulated HCQ.4 cells. HCQ.4 cells were incubated with antigen, antigen presenting cells and increasing concentrations of recombinant nonmutated 1m (open squares) and N17,96Q-1m (solid squares). After 24 h the supernatants were analyzed by IL-2 ELISA. The inhibition studies were carried out in duplicate. The results from one of three experiments are shown as the mean values ± 2 SEM. (B). Inhibition of proliferation of PPD-stimulated human PBL. Human PBL were incubated with PPD in the presence of increasing concentrations of plasma 1m (solid circles), recombinant nonmutated 1m (open squares) and recombinant N17,96Q-1m (solid squares). After 92 h cells were pulsed with methyl-[3H]thymidine and 4 h later the cells were harvested and the incorporated [3H]thymidine was determined. The inhibition studies were carried out in triplicate. Maximum proliferation (+) was set by cells cultured in the presence of PPD and absence of 1m. Minimum proliferation (–) was set by cells cultured in the absence of PPD and 1m.

 
In vivo distribution of ¹²⁵I-₁m
Radiolabeled ₁m variants were injected into rats intravenously and the distribution studied by whole body counting. Besides plasma, recombinant and N17,96Q-₁m, neuraminidase-digested plasma ₁m (asialo ₁m) was included in these experiments to study the uptake by liver asialoglycoprotein receptors (Ashwell and Morell, 1974).The blood clearance of the injected proteins was studied from the selected regions of interest over the hearts. The T-values of the first clearance phases were obtained by regression analysis of the values. As shown in Figure 8A, recombinant nonmutated and neuraminidase digested plasma ₁m showed the fastest clearance from blood (T = 0.7 and 0.6 min, respectively), followed by N17,96Q-₁m (T = 2.1 min) and plasma ₁m (T = 3.1 min). Indications of a second, much slower, elimination phase were seen for all four ₁m variants. Recombinant nonmutated and neuraminidase-digested plasma ₁m also showed a much faster uptake in the liver than the other two variants, and N17,96Q-₁m showed the fastest uptake into the kidneys. After 45 min, recombinant nonmutated ₁m and neuraminidase-digested plasma ₁m were found primarily in the liver and plasma ₁m and N17,96-₁m in the kidneys (Figure 8B). The results are consistent with an uptake of recombinant nonmutated ₁m to the liver via asialoglycoprotein receptors (Ashwell and Morell, 1974) and a faster glomerular filtration in the kidneys of the nonglycosylated N17,96Q-₁m. The results thus show that the carbohydrates of ₁m are important for the in vivo turnover of the protein.

View larger version (29K): [in this window] [in a new window]   Fig. 8. In vivo turnover of 125I-plasma 1m, recombinant nonmutated 1m, N17,96Q-1m and neuraminidase-digested plasma 1m (asialo 1m). Radiolabeled 1m (4 MBq, 5 µg) was injected intravenously into a rat and the distribution registrated with a scintillation camera during 45 min. (A) Time activity curves showing the clearance from blood of the injected 125I-labelled proteins during 15 min after injection. (B) Percent of injected 125I-labeled plasma-, recombinant nonmutated-, N17,96Q- and asialo-1m uptake in blood, liver, and kidneys 45 min after injection. The rats were killed and blood samples taken, whole livers and kidneys dissected and analyzed by the scintillation camera. The staples represent the mean values of two experiments ± SEM.

 

	Discussion

Top Abstract Introduction Results Discussion Materials and methods Acknowledgments References

 
Glycosylation can be important for many properties of glycoproteins. It has been shown that N-glycosylation can influence protein folding, stability, antigenicity, intracellular transport and biological activity (Opdenakker et al., 1993). In this work we have investigated the role of the carbohydrates on a human plasma and tissue protein, ₁m. ₁m is a protein with immunoregulatory properties, which is synthesized in the liver and found in blood and all tissues. The results presented here suggest that the carbohydrates are important for the secretion and tissue localization of the protein, but not for the structure or immunosuppressive effects.

The secretion of carbohydrate-free N17,96Q-₁m from the recombinant insect cells was severely reduced as compared to nonmutated ₁m. Most of the mutated protein was found inside the cells. Nonmutated ₁m, which is normally secreted in high amounts, also accumulated intracellularly when it was expressed in the presence of tunicamycin. This indicates that the lack of carbohydrates rather than the mutation, an exchange of asparagine to glutamine, cause the reduced secretion. A defect secretion has been reported for other carbohydrate-deleted proteins expressed in the baculovirus-insect cell system, either as a result of point mutations (Sareneva et al., 1994) or culturing in the presence of tunicamycin (Chawla and Owen, 1995). It is generally believed that N-linked oligosaccharides are important for secretion of proteins (for a review, see Helenius et al., 1992). The mechanism behind this could be that the N-linked oligosaccharides are needed for a correct folding in the endoplasmatic reticulum and that the absence of the carbohydrates leads to misfolding and aggregation (Helenius et al., 1992). Another possibility is that a structure on the N-linked oligosaccharides is necessary for secretion. Our data support the former alternative since polyclonal antibodies showed a much weaker binding to intracellular ₁m, indicating misfolding of the protein. The fact that the carbohydrate-deleted protein was partly secreted argues against the N-glycans as a necessary signal for secretion. The secreted protein was recognized by polyclonal antibodies and presumably correctly folded. This can be explained by the activity of chaperone proteins, a system only capable of handling a part of the overexpressed protein.

Circular dichroism analysis of secreted recombinant nonmutated and N17,96Q-₁m indicated a similar overall structure. Human ₁m has previously been shown to consist of mainly ß-sheet structure with minor contributions of -helix (Ekström and Berggård, 1977; Gavilanes et al., 1984; Wester et al., 1997). The CD-spectrum analysis confirmed this also for the N17,96Q-mutated protein. Moreover, a panel of antibodies recognized the recombinant nonmutated and N17,96Q-mutated proteins to the same degree, indicating that the same epitopes are present on the surface of the two proteins. Both proteins were found to be heterogeneously charged and carrying yellow-brown chromophores. The structure and identity of the chromophores are still unknown, but they are most likely covalently bound to the protein core. There is a correlation between the degree of yellow-brown color and the net charge of ₁m (Calero et al., 1996; Wester et al., 1997), which is further supported in this work: N17,96Q-₁m displayed somewhat higher net charge and was more intensely colored. It was previously shown that the chromophore is not associated with the carbohydrate part of ₁m, since a complete removal of the carbohydrate by glycosidase treatment did not alter the optical properties (Åkerström et al., 1995). The results presented here show that the carbohydrates are not needed at all for the formation of the chromophore of ₁m. Finally, both recombinant forms of ₁m were significantly more colored than plasma ₁m (Wester et al., 1997), suggesting that the insect cell expression system somehow promotes the formation of the chromophore.

Based on the immunosuppressive properties of ₁m, several authors have suggested a regulatory role of ₁m on the immune system (Méndez et al., 1986; Åkerström, 1992; Santin and Cannas, 1999). Carbohydrates have been shown to be involved in many important processes in the immune system, as exemplified by selectins and orosomucoid (Lasky, 1992; Shiyan and Bovin, 1997). Glycopeptides isolated from human ₁m showed an inhibitory effect on the antigen-stimulated proliferation of human peripheral leukocytes (Åkerström and Lögdberg, 1984). The effect was also shown to be independent of the polypeptide backbone, and it was suggested that the N-linked oligosaccharides of ₁m are responsible for the immunosuppressive effect. The hypothesis was tested in this paper, on antigen-induced stimulation of mouse T cell hybridomas, a system in which plasma and recombinant ₁m previously have been shown to exert an inhibitory effect (Wester et al., 1998), and on PPD-induced stimulation of human peripheral lymphocytes, one of the most commonly used antigen model systems and identical to the one previously used to show the effect of isolated N-linked oligosaccharides of ₁m (Åkerström and Lögdberg, 1984). It was unexpectedly found that carbohydrate-free ₁m had the same immunosuppressive effects and bound to lymphocytes to the same degree as fully glycosylated ₁m. We can therefore conclude that the oligosaccharides are not responsible for the immunosuppressive effects or binding to lymphocytes. A possible explanation for the immunosuppressive effects of the glycopeptide preparation is that a co-purified low molecular weight substance exerts the effect. From the results in this paper it must be considered unlikely that a contaminating substance, i.e. not related to ₁m, causes the effect, since ₁m isolated from two completely different sources like human plasma and serum-free cell culture medium exert the same effect. Instead, the hypothetical lipocalin ligand could be a possible candidate for such a substance.

₁m has been shown to bind to receptors on PBL and T cell hybridomas from mouse, and to the human histiocytic cell-line U937 (Fernández-Luna et al., 1988; Babiker-Mohamed et al., 1990a; Wester et al., 1998). We have shown here that the protein also binds to human peripheral T cells, B cells and NK cells. The binding is weak, similar to the previous results. The binding is consistent with immunohistochemical findings, which have shown a staining of lymphocytes, monocytes and macrophages with antibodies against ₁m ((Berggård et al., 1999; Bouic et al., 1984). Thus, it can be concluded that ₁m binds weakly to a wide variety of blood cells. This is logical considering that the protein has been shown to regulate many different functions of several blood cell populations (Lögdberg and Åkerström, 1981; Lögdberg et al., 1986; Méndez et al., 1986; Babiker-Mohamed et al., 1990a,b; Wester et al., 1998).

₁m is synthesized almost exclusively by the liver (see, for instance, Berggård et al., 1998). It is secreted to the blood and, due to its relatively small size, filtrated and degraded in the kidneys. However, it is also found in the interstitial tissue of all examined organs except the brain (Bouic et al., 1984; Ødum and Nielsen, 1994; Berggård et al., 1998). The transcription of the ₁m-gene was increased during inflammation but the levels of ₁m were unchanged in blood and urine (Falkenberg et al., 1997). It has therefore been speculated that, after synthesis in the liver and secretion to blood, ₁m is transported to tissues where its immunosuppressive properties help protect these tissues from unwanted immune and inflammatory reactions (Ødum and Nielsen, 1994; Falkenberg et al., 1997; Berggård et al., 1999). This view is supported by the results in this work. In our rat model, the protein was cleared from blood with a half-life of 3.1 min, but more than half of the protein was found in other compartments than blood, liver, and kidneys after 45 min. The results also show that the distribution of ₁m is dependent on the carbohydrate part of the molecule. Carbohydrate-free ₁m was metabolized by the kidneys faster than wild-type ₁m, and the sialic acid apparently protected the protein from uptake, and possibly degradation (Ashwell and Morell, 1974) by the liver.

The results in this paper have implications for future applications of ₁m in immunotherapy. We have shown that immunologically functional ₁m can be expressed in large amounts. Furthermore, the structure of the carbohydrates is probably important for the delivery of ₁m to adequate body compartments. The carbohydrates apparently prolong the half-life of the protein in the blood and sialic acid prevents it from uptake in the liver. In line with this, it should be possible to use other expression systems to obtain a functional ₁m in which the structure of the carbohydrate part and the number of oligosaccharide units have been optimized for drug delivery purposes.

	Materials and methods

Top Abstract Introduction Results Discussion Materials and methods Acknowledgments References

 
Mutation of ₁-microglobulin
cDNA coding for human ₁m and its signal peptide was mutated to eliminate the sites for N-glycosylation. The amino acid substitution Asn to Gln was made in position 17 and 96 (N17,96Q) using a PCR-based primer extension mutagenesis method. Four oligomeres, Gln 17 sense, Gln 17 antisense, Gln 96 sense, and Gln 96 antisense, containing the mutations and the two oligomeres, HA5 and HA3, flanking the ₁m-part of the ₁m-bikunin cDNA were synthesized by BioMolecular Service Center, Lund University. The flanking oligomeres also contained restriction sites for XbaI and BamHI. The N17 mutation was produced in the first step, using cDNA coding for human ₁m as a template. Pfu DNA polymerase (Stratagene, La Jolla, CA) was used in all PCR reactions. The resulting fragments were extracted from an agarose-gel and then denatured in a 95°C-degree waterbath and slowly annealed together as this water cooled to room temperature. The fragments were then extended with Klenow polymerase at 37°C for 1 h to one new blunt-ended fragment. The fragment was amplified with PCR, subcloned into the pCRscript plasmid (pCR-Script SK(+) Cloning kit; Stratagene) and sequenced. The double mutation, N17,96Q was produced in the second step using N17Q mutated ₁m as template. The double-mutated fragment was ligated into the pCRscript plasmid and the pCRscript-₁m construct was cleaved by XbaI and BamHI (Gibco BRL, Life Technologies Inc., Gaithersburg, MD) and ligated into pVL1392 (Invitrogen, San Diego, CA). Bsu36I-digested BacPAK 6 baculovirus DNA (Clontech, Palo Alto, CA) and pVL1392 containing the ₁m DNA insert were cotransfected into Spodoptera frugiperda (Sf9) insect cells (Invitrogen) using insectin (Invitrogen) according to the manufacturers description. Individual virus clones were isolated, analyzed for ₁m production, and amplified to large virus stocks containing approximately 10⁸ pfu/ml. Trichoplusia ni (Hi-5) cells cultured in serum-free Ex-cell 401 media (JRH Biosciences, Lenexa, KS), were infected for protein expression, using a moi of 10, and the optimal time of harvest was determined. The procedures followed mainly the description of Bratt and Åkerström (1995) and King and Posse (1992).

Analysis of ₁-microglobulin secretion
Nonmutated ₁m and N17,96Q-mutated ₁m were expressed in the presence or absence of tunicamycin. Insect cells, 1.5 x 10⁶, were grown in 6-well plates (Corning Costar, Bodenheim, Germany) and inoculated with 10 pfu/cell of baculovirus. The cells were left for incubation in 1.5 ml medium at 27°C. In some wells tunicamycin (Calbiochem, La Jolla, CA) was added to the medium at a final concentration of 10 µg/ml. Each day between day 1 and 6 after inoculation, infected cells cultured in presence or absence of tunicamycin, were harvested. Cells and media were separated by low-speed centrifugation. Protease inhibitors were added to the media (leupeptin 10 µg/ml, antipain 5 µg/ml, and pepstatin 1 µg/ml). The cell-pellets were suspended in 100 µl of PBS and lysed by incubation with 25 µl of lysis-buffer (50 mM Tris–HCl, pH 6.9, 25% glycerol, 10% SDS, 25% ß-mercaptoethanol and 0.25% bromphenolblue) at 100°C for 5 min. 10 µl of cell media and 2.5 µl of the cell lysates were analyzed by SDS–PAGE and Western blotting.

Purification of ¹-microglobulin
All ₁m-variants were purified by anti-₁m affinity chromatography and gel chromatography. Nonmutated and N17,96Q-mutated recombinant ₁m were purified as described by Wester et al. (1997). Plasma and urinary ₁m were isolated as described by Wester et al. (1997) and Åkerström et al. (1995), respectively.

Radiolabeling
Proteins were labeled with ¹²⁵I using the chloramine T method (Greenwood et al., 1963). Labeled proteins were separated from free iodide by gel-filtration on Sephadex G-25 columns (Pharmacia). The specific radioactivity achieved was around 1 MBq/µg.

Electrophoresis
SDS–PAGE under reducing conditions was done according to the procedure described by Laemmli (1970) and detailed previously (Wester et al., 1997). Agarose electrophoresis was done according to (Johansson, 1972) with a 0.8% agarose gel (SeaKem ME, FMC BioProducts, Rockland, ME) as outlined (Wester et al., 1997).

Immunochemical methods
Proteins were separated by SDS–PAGE and transferred to polyvinylidene difluoride (PVDF) membranes (Immobilon, Millipore, Bedford, MA) as described previously (Madisudaira, 1987). ₁m was then detected by Western blotting as described previously (Wester et al., 1997). Briefly, the membranes were blocked and incubated with polyclonal rabbit anti-human ₁m antibodies, K: 324 (prepared in our laboratory) and finally with ¹²⁵I-goat anti-rabbit IgG (prepared in our laboratory) (0.5 x 10⁶ c.p.m./ml).

Solid-phase radioimmunoassay was performed following mainly the description of Nilson et al. (1986). Briefly, serial dilutions of the proteins were coated on microtiter plates overnight, at room temperature. The plates were then incubated with either one of five monoclonal mouse antibodies (BN 11.1, BN 11.2, BN 11.3, BN 11.5 or BN 11.10), described in Babiker-Mohamed et al. (1991) at a concentration of 10 µg/ml, or with polyclonal rabbit anti human-₁m antiserum, K:324, (diluted 500 times). They were then incubated with either radiolabeled rabbit anti-mouse Ig (DAKO A/S, Denmark) or goat anti-rabbit Ig (5 kBq/ml). Finally, the plates were washed, cut and counted in a -counter (Packard, Meriden, CT). Background level was set by excluding the primary antibody step.

Radioimmunoassay (RIA) was done by mixing fixed amounts of ¹²⁵I-labeled ₁m and polyclonal goat anti-human urinary-₁m antibodies (prepared in our laboratory) with samples of unknown ₁m concentration (Åkerström, 1985). The mixture was left overnight at 4°C, and then precipitated by 10% polyethylene glycol 6000. The pellets were analyzed for radioactivity in a -counter.

Stokes’s radius and frictional ratio
Stokes’s radius, r_s, and frictional ratio (f/f₀) were determined by gel chromatography (Laurent and Killander, 1964) on a Sephacryl S-200 column as described previously (Bratt and Åkerström, 1995) Bovine serum albumin (BSA) (Boehringer-Mannheim, GmbH, Mannheim, Germany) was used as a standard with known Stokes’s radius.

Carbohydrate analysis
₁m variants were digested sequentially with neuraminidase, O-Glycosidase and N-Glycosidase F as described previously (Wester et al., 1997). Agarose-insolubilized neuraminidase (from Clostridium perfringens) was purchased from Sigma Chemical Co. O-Glycosidase (from Diplococcus pneumoniae), and N-Glycosidase F (from Flavobacterium meningosepticum) were purchased from Boehringer-Mannheim, Mannheim, Germany. The binding of the lectins Concanavalin A (ConA) and peanut agglutinin (PNA) was tested by incubation of proteins, separated by SDS–PAGE and transferred to PVDF-membranes, with digoxigenin-labeled lectins (DIG Glycan Differentation Kit, Boehringer-Mannheim) using the protocol supplied with the kit.

Circular dichroism studies
CD spectra were recorded at 37°C using a JASCO J-720 spectropolarimeter with a thermostated cell holder. Ten scans were performed on each protein sample using a scan speed of 20 nm/min, a sampling interval of 1 nm, and 4 s response time. Measurements were performed on protein samples dissolved in 20 mM sodium-phosphate buffer, pH 7.0 in 0.1 cm quartz cuvettes. The contribution from the buffer was subtracted and the spectra were moderately noise reduced. The protein concentration of the samples was determined by absorbance at 280 nm.

Flow cytometric analysis
The binding of ₁m to different cells was analyzed by flow cytometry. The mouse CD4+ T hybridoma cell line, HCQ.4, which is specific for rat collagen II, was cultivated as described (Michaëlsson et al., 1994). Human peripheral lymphocytes were prepared from blood from normal healthy donors by centrifugation on lymphoprep (Nycomed Amersham plc) at 800 x g for 20 min and washed once by PBS at 400 x g for 10 min. The contaminating red blood cells were lysed by treating the cells with ortholyse (Ortho-mune Lysing reagent, Ortho Diagnostic Systems Inc., Johnson & Johnson, Raritan, NJ) for 7 min. Cells were centrifuged and resuspended in PBS + 1 mg/ml BSA. Flow cytometric analysis was made by incubating the cells with 1 mg/ml plasma ₁m, recombinant nonmutated or N17,96Q-₁m or BSA (control), in the PBS-BSA buffer. After 15 min , the cells were washed and incubated for 10 min with 10 µg/ml monoclonal mouse anti-₁m antibodies (BN 11.3), washed and incubated 10 min with FITC-conjugated goat anti-mouse immunoglobulin (GAM-FITC, DAKO A/S, Denmark). PBL were then washed carefully to remove excess of GAM-FITC and finally incubated with phycoerythrin-conjugated mouse monoclonal anti-human CD8, CD19 (DAKO) or CD4, CD56 (BD immunocytometry systems, San Jose, CA) diluted 1:10 in PBS+1 mg/ml BSA. The cells were washed, fixed with 1% paraformaldehyde and analyzed on a FACScan flow cytometer (Becton Dickinson AB, Sweden). At least 10,000 cells were registered and a lymphocyte region combined with a region including the CD4, CD8, CD19, and CD56 positive cells, respectively, was analyzed to evaluate the binding of ₁m to each lymphocyte subset.

Assay for inhibition of IL-2 production
The ₁m inhibition of IL-2 production of the HCQ.4 T cell line stimulated by antigen (a rat collagen II peptide) in the presence of mouse (B10Q x DBA/1)F₁ splenocytes was studied as described (Wester et al., 1998). Briefly, serial dilutions of ₁m variants were mixed with mouse T hybridoma cells, mouse splenocytes, and antigen. After 24 h at 37°C, the supernatants were analyzed for IL-2 content with an ELISA Development system for mouse IL-2 (Duoset, Genzyme Diagnostics, USA).

Assay for inhibition of proliferation
The procedure followed mainly the description of Åkerström and Lögdberg (1984). I principle, human PBL were incubated with an antigen-mixture from tubercle bacilli, PPD (purified protein derivative, also called tuberculin), which is used for testing of immunity against tuberculosis, and the resulting cell proliferation was inhibited by various concentrations of added ₁m. PBL were isolated from buffy coats obtained from normal healthy donors on Ficoll-Paque (Amersham-Pharmacia biotech, Uppsala, Sweden). The cells were then washed once in RPMI 1640 medium (Gibco, Paisley, UK) supplemented with 10 mM HEPES, 4 mM L-glutamine, 1 mM pyruvat, 0.1% NaHCO₃, and 10% fetal calf serum and then resuspended in this medium. The cells were cultured in flat-bottomed 96-well microtiter plates (Corning Costar) at 37°C in a CO₂- incubator. Each well contained 2 x 10⁵ cells in 0.2 ml medium. PPD, 1 mg/ml in PBS (Statens seruminstitut, Copenhagen, Denmark) was added to 50 µg/ml, and plasma ₁m, recombinant nonmutated or N17,96Q-₁m were added to final concentrations of 250, 125, 62.5, 31.3, 15.6, 7.81, 3.90, and 0 µg/ml. The cultivation time was 4 days. Four hours before termination the cells were pulsed with 0.5 µCi methyl-[³H]thymidine and then harvested onto filter paper and the thymidine incorporation was measured in a ß-scintillation counter. The inhibition studies were carried out in triplicate. The maximal proliferation was set by the mean value of triplicate of cells cultured in the presence of PPD and absence of ₁m and a minimal proliferation was set by the mean value of triplicate of cells cultured in the absence of PPD and ₁m.

Biokinetic studies
The biokinetic distribution of ¹²⁵I-labeled plasma ₁m, neuraminidase-digested plasma ₁m, recombinant nonmutated ₁m and N17,96Q-₁m was performed in female Sprague-Dawley rats. The animals were anaesthetized first firmly with ether and then intraperitoneally with chloral hydrate (5%) using a volume of 0.6 ml/100 g body weight. Labeled protein containing an activity of 4.25 ± 0.27MBq ¹²⁵I in a volume of 0.6 ml was injected intravenously in the right femoral vein. Dynamic studies were carried out during 45 min using a scintillation camera (Maxi Camera I, General Electric), equipped with a Low Energy High Resolution Parallel collimator. Immediately after the end of the study the animals were killed and dissected. Blood sample and whole organs (heart, kidneys, liver, spleen, lungs, muscle, stomach, intestine, appendix, and thymus) were removed and taken to analysis for static image collection. In order to generate time-curves for the blood clearance and the activity uptake in organs, regions of interest were selected over the heart, liver, kidneys and bladder in the dynamic images. The T-values for the blood clearance were obtained by regression analysis (least-square curve fitting to exponential polynoms) of the values over the heart regions. To estimate the activity in the dissected organs a scintillation camera correction factor for ¹²⁵I was used.

	Acknowledgments

Top Abstract Introduction Results Discussion Materials and methods Acknowledgments References

 
The authors which to thank Meirav Holmdahl, Rickard Holmdahl, Lotta Jansson, and Maria Silow for valuable discussions and Lisa Palm, Kerstin Boll, and Maria Allhorn for excellent technical assistance. This work was supported by grants from the Swedish Medical Research Council (project no. 7144), King Gustav V’s 80-year Foundation, the Medical Faculty at Lund University, the Swedish Society for Medical Research, the Royal Physiographic Society in Lund, the Swedish Rheumatism Society, and the Foundations of Crafoord, Greta and Johan Kock, and Alfred Österlund.

	Footnotes

 
¹ To whom correspondence should be addressed at: Section for Molecular Signalling, Department of Cell and Molecular Biology, P.O. Box 94, University of Lund, S-221 00 Lund, Sweden

	References

Top Abstract Introduction Results Discussion Materials and methods Acknowledgments References

 
Åkerström,B. (1985) Immunological analysis of a1-microglobulin in different mammalian and chicken serum. 1-microglobulin is 5–8 kDa larger in primates. J. Biol. Chem., 260, 4839–4844.

Åkerström,B. (1992) Role of -1-microglobulin in immune response and inflammation. Folia Histochem. Cytobiol., 30, 183–186.[Medline]

Åkerström,B. and Lögdberg,L. (1984) 1-microglobulin glycopeptides inhibit antigen-specific stimulation of human peripheral blood leukocytes. Scand. J. Immunol., 20, 559–563.[Web of Science][Medline]

Åkerström,B., Bratt,T. and Enghild,J.J. (1995) Formation of the a1-microglobulin chromophore in mammalian and insect cells: a novel post-translational mechanism? FEBS Lett., 362, 50–54.[Web of Science][Medline]

Ashwell,G. and Morell,A.G. (1974) The role of surface carbohydrates in the hepatic recognition and transport of circulating glycoproteins. Adv. Enzymol. Relat. Areas Mol. Biol., 41, 99–128.[Web of Science][Medline]

Babiker-Mohamed,H.F.,M. Olsson,M.L. Winquist,O. Nilson,B.H.K. Lögdberg,L. and Åkerström,B. (1991) Characterization of monoclonal anti-1-microglobulin antibodies: binding strength, binding sites and inhibition of 1-microglobulin-induced lymphocyte stimulation. Scand. J. Immunol., 34, 655–666.[Web of Science][Medline]

Babiker-Mohamed,H., Åkerström,B. and Lögdberg,L. (1990b) Mitogenic effects of 1-microglobulin on mouse lymphocytes. Evidence of T and B cell co-operation, B-cell proliferation and a low affinity receptor on mononuclear cells. Scand. J. Immunol., 32, 37–44.[Web of Science][Medline]

Babiker-Mohamed,H., Olsson,M.L., Boketoft,Å., Lögdberg,L. and Åkerström,B. (1990a) Alpha-1-microglobulin is mitogenic to human peripheral blood lymphocytes. Regulation by both enhancing and suppressive serum factors. Immunobiology, 180, 221–234.

Berggård,T., Enghild,J.J., Badve,S.,Salafia,C.M., Lögdberg,L. and Åkerström,B. (1999) Histologic distribution and biochemical properties of -1-microglobulin in human placenta. Am. J. Reprod. Immunol., 41, 52–60.

Berggård,T., Oury,T.D., Thøgersen,I.B., Åkerström,B. and Enghild,J.J. (1998) -1-microglobulin is found both in blood and in most tissues. J. Histochem. Cytochem., 46, 887–893.[Abstract/Free Full Text]

Berggård,T., Thelin,N., Falkenberg,C., Enghild,J.J. and Åkerstrom,B. (1997) Prothrombin, albumin and immunoglobulin A form covalent complexes with 1-microglobulin in human plasma. Eur. J. Biochem., 245, 676–683.[Web of Science][Medline]

Bouic,P., Vincent,C. and Revillard,J.-P. (1984) Localization of -1-microglobulin (HC protein) in normal human tissues: an immunohistochemical study using monoclonal antibodies. Histochem. J., 16, 1311–1324.[Web of Science][Medline]

Bratt,T. and Åkerström,B. (1995) Expression of rat 1-microglobulin-bikunin in baculovirus-transformed insect cells. Protein Expr. Purif., 6, 431–438.[Web of Science][Medline]

Bratt,T., Olsson,H., Sjöberg,E.M., Jergil,B. and Åkerström,B. (1993) Cleavage of the 1-microglobulin-bikunin precursor is localized to the Golgi apparatus of rat liver cells. Biochim. Biophys. Acta, 1157, 147–154.[Medline]

Calero,M.E.J., Fernando,S., Grubb,A., Brew,K. and Méndez,E. (1996) Spectroscopic characterization by photodiode array detection of human urinary and amniotic protein HC subpopulations fractionated by anion-exchange and size-exclusion high-performance liquid chromatography. J. Chromatogr. A, 719, 149–157.[Web of Science][Medline]

Chawla,D. and Owen,J.S. (1995) Secretion of active human lecithin-cholesterol acyltransferase by insect cells infected with a recombinant baculovirus. Biochem. J., 309, 249–253.

Ekström,B. and Berggård,I. (1977) Human 1-microglobulin: purification procedure, chemical and physicochemical properties. J. Biol. Chem., 252, 8048–8057.

Ekström,B., Lundblad,A., Svensson,S. (1981) Structural studies on the carbohydrate portion of human -1-microglobulin. Eur. J. Biochem., 114, 663–666.[Web of Science][Medline]

Ekström,B., Peterson,P.A., Berggård,I. (1975) A urinary and plasma -1-glycoprotein of low molecular weight: isolation and some properties. Biochem. Biophys. Res. Commun., 65, 1427–1433.[Web of Science][Medline]

Escribano,J., Grubb,A., Calero,M. and Méndez,E. (1991) The protein HC chromophore is linked to the cysteine residue at position 34 of the polypeptide chain by a reduction-resistant bond and causes the charge heterogeneity of protein HC. J. Biol. Chem., 266, 15758–15763.[Abstract/Free Full Text]

Escribano,J., Lopex-Otin,C., Hjerpe,A., Grubb,A. and Méndez,E. (1990) Location and characterization of the three carbohydrate prosthetic groups of human protein HC. FEBS Lett., 266, 167–170.[Web of Science][Medline]

Falkenberg,C.G.A., Åkerström,B. (1990) Isolation of rat serum -1-microglobulin. Identification of a complex with -1-inhibitor-3. A rat -2-macroglobulin homologue. J. Biol. Chem., 265, 16150–16157.[Abstract/Free Full Text]

Falkenberg,C., Blom,A., Fries,E., Ekström,G., Särnstrand,B., Salier,J.-P. and Åkerström,B. (1997) -1-microglobulin and bikunin in rats with collagen II-induced arthritis: Plasma levels and liver mRNA content. Scand. J. Immunol., 46, 122–128.[Web of Science][Medline]

Falkenberg,C., Enghild,J.J., Thøgersen,I., Salvesen,G. and Åkerström,B. (1994) Isolation and characterization of fibronectin--1-microglobulin complex in rat plasma. Biochem. J., 301, 745–751.

Fernández-Luna,J.L., Leyva-Cobian,F. and Mollinedo,F. (1988) Identification of the protein HC receptor. FEBS Lett., 236, 471–474.[Web of Science][Medline]

Flower,D.R. (1996) The lipocalin protein family: structure and function. Biochem. J., 318, 1–14.

Gavilanes,J.G., Lopez-Otin,C., Gavilanes,F. and Méndez,E. (1984) Conformational studies of the human complex-forming glycoprotein, heterogeneous in charge: protein HC. Biochemistry, 23, 1234–1238.[Medline]

Greenwood,F.C., Hunter,W.M. and Glover,J.S. (1963) The preparation of ¹³¹I-labelled human growth hormone. Biochem. J., 89, 114–123.[Web of Science][Medline]

Grubb,A., Méndez,E., Fernández-Luna,J.L., Lopez,C., Mihaesco,E. and Vaerman,J.-P. (1986) The molecular organization of the protein HC-IgA complex (HC-IgA). J. Biol. Chem., 261, 14313–14320.[Abstract/Free Full Text]

Helenius,A., Marquardt,T. and Braakman,I. (1992) The endoplasmic reticulum as a protein-folding compartment. Trends Cell Biol., 2, 227–231.

Johansson,B.G. (1972) Agarose gel electrophoresis. Scand. J. Clin. Lab. Invest., 29, 7–19.

Kaumeyer,J.F., Polazzi,J.O. and Kotick,M.P. (1986) The mRNA for a proteinase inhibitor related to the HI-30 domain of inter--trypsin inhibitor also encodes 1-microglobulin (protein HC). Nucleic Acids Res., 14, 7839–7850.

King,L.A. and Posse,R.D. (1992) The baculovirus expression system: a laboratory guide. Chapman and Hall, London.

Laemmli,U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 227, 680–685.

Lasky,L.A. (1992) Selectins: interpreters of cell-specific carbohydrate information during inflammation. Science, 258, 964–969.[Abstract/Free Full Text]

Laurent,T.C. and Killander,J. (1964) A theory of gel filtration and its experimental verification. J. Chromatogr., 14, 317–330.

Lindqvist,A. and Åkerström,B. (1999) Isolation of plaice (Pleuronectes platessa) -1-microglobulin: conservation of structure and chromophore. Biochim. Biophys. Acta 1430, 222–233.[Medline]

Lögdberg,L. and Åkerström,B. (1981) Immunosupressive properties of 1-microglobulin. Scand. J. Immunol., 13, 383–390.

Lögdberg,L., Åkerström,B. and Shevach,E.M. (1986) Alpha-1-microglobulin is mitogenic for guinea pig lymphocytes. Scand. J. Immunol., 24, 575–581.

Madisudaira,P. (1987) Sequence from picomole quantities of proteins electroblotted onto polyvinylidene difluoride membranes. J. Biol. Chem., 262, 10035–10038.[Abstract/Free Full Text]

Méndez,E., Fernández-Luna,J.L., Grubb,A. and Leyva-Cobián,F. (1986) Human protein HC and its IgA complex are inhibitors of neutrophil chemotaxis. Proc. Natl. Acad. Sci. USA, 88, 1472–1475.

Michaëlsson,E., Malmström,V., Reis,S., Engström,A., Burkhardt,H. and Holmdahl,R. (1994) T cell recognition of carbohydrates on type II collagen. J. Exp. Med., 180, 745–749.[Abstract/Free Full Text]

Nilson,B., Björck,L. and Åkerström,B. (1986) Detection and purification of rat and goat immunoglobulin G antibodies using protein G-based solid-phase radioimmunoassays. J. Immunol. Methods, 91, 275–281.

Ødum,L. and Nielsen,H.W. (1994) Human protein HC ( 1-microglobulin) and inter--trypsin inhibitor in connective tissue. Histochem. J., 26, 799–803.[Web of Science][Medline]

Opdenakker,G., Rudd,P.M., Ponting,C.P. and Dwek,R.A. (1993) Concepts and principles of glycobiology. FASEB J., 7, 1330–1337.[Abstract]

Salier,J.-P., Rouet,P., Raguenez,G. and Daveau,M. (1996) Then inter--inhibitor family: from structure to regulation. Biochem. J., 315, 1–9.

Santin,M. and Cannas,M. (1999) Collagen-bound 1-microglobulin in normal and healed tissues and its effect on immunocompetent cells. Scand. J. Immunol., 50, 289–295.[Web of Science][Medline]

Sareneva,T., Pirhonen,J., Cantell,K., Kalkkinen,N. and Julkunen,I. (1994) Role of N-glycosylation in the synthesis, dimerization and secretion of human interferon-gamma. Biochem. J., 303, 831–840.

Shiyan,S.D. and Bovin,N.V. (1997) Carbohydrate composition and immunomodulatory activity of different glycoforms of -1-acid glycoprotein. Glycoconj. J., 14, 631–638.

Tejler,L. and Grubb,A.O. (1976) A complex-forming glycoprotein heterogeneous in charge and present in human plasma, urine and cerebrospinal fluid. Biochim. Biophys. Acta, 439, 82–94.[Medline]

Wester,L., Johansson,M.U. and Åkerström,B. (1997) Physicochemical and biochemical characterization of human 1- microglobulin expressed in baculovirus-infected insect cells. Protein Expr. Purif., 11, 95–103.[Web of Science][Medline]

Wester,L., Michaëlsson,E., Holmdahl,R., Olofsson,T. and Åkerström,B. (1998) Receptor for 1-microglobulin on T lymphocytes: inhibition of antigen-induced interleukin-2 production [in process citation]. Scand. J. Immunol., 48, 1–7.[Web of Science][Medline]

CiteULike    Connotea    Del.icio.us    What's this?

This article has been cited by other articles:

				R. Jayachandran, M. S. Shaila, and A. A. Karande Analysis of the Role of Oligosaccharides in the Apoptotic Activity of Glycodelin A J. Biol. Chem., March 5, 2004; 279(10): 8585 - 8591. [Abstract] [Full Text] [PDF]

				M. Allhorn, T. Berggard, J. Nordberg, M. L. Olsson, and B. Akerstrom Processing of the lipocalin alpha 1-microglobulin by hemoglobin induces heme-binding and heme-degradation properties Blood, March 15, 2002; 99(6): 1894 - 1901. [Abstract] [Full Text] [PDF]

				E. Bulow, W. M. Nauseef, M. Goedken, S. McCormick, J. Calafat, U. Gullberg, and I. Olsson Sorting for storage in myeloid cells of nonmyeloid proteins and chimeras with the propeptide of myeloperoxidase precursor J. Leukoc. Biol., February 1, 2002; 71(2): 279 - 288. [Abstract] [Full Text] [PDF]

This Article Abstract FREE Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (9) Request Permissions Disclaimer Google Scholar Articles by Wester, L. Articles by Åkerström, B. Search for Related Content PubMed PubMed Citation Articles by Wester, L. Articles by Åkerström, B. Social Bookmarking          What's this?

Online ISSN 1460-2423 - Print ISSN 0959-6658

Copyright © 2009 Oxford University Press

Oxford Journals Oxford University Press

• Site Map
• Privacy Policy
• Frequently Asked Questions

Other Oxford University Press sites: Oxford University Press Oxford Journals China Oxford Journals Japan American National Biography Booksellers' Information Service Children's Fiction and Poetry Children's Reference Corporate & Special Sales Dictionaries Dictionary of National Biography Digital Reference English Language Teaching Higher Education Textbooks Humanities International Education Unit Law Medicine Music Online Products Oxford English Dictionary Reference Rights and Permissions Science School Books Social Sciences Very Short Introductions World's Classics