Galectin-1 Promotes Immunoglobulin Production during Plasma Cell Differentiation¹

Galectins are a family of lectins having characteristic amino acid sequences for carbohydrate recognition and an affinity for β-galactosides. Galectins play important roles in regulating immune cell homeostasis, in host-pathogen interactions, and in tumorigenesis (1, 2, 3). Thus far, 15 mammalian galectins have been identified and sequenced (3, 4). Galectins are localized to the cytoplasm, nucleus, and the extracellular environment (5). Secretion of galectins occurs via a nonclassical secretion pathway that requires association of galectins with glycosylated counter-receptors (6, 7).

Galectin-1 modulates cells of the immune system in a number of ways. Galectin-1 produced by thymic epithelial cells causes apoptosis in human thymocytes (8). In peripheral blood, galectin-1 causes apoptosis of activated T cells by cooperating with TCR engagement, but it supports the survival of naive T cells (9, 10, 11). Galectin-1 has also been proposed to shift the T cell polarization reaction from Th1 to Th2 by triggering apoptosis in Th1 cells (12, 13). Galectin-1 also promotes surface exposure of phosphatidylserine (PS)³ without accompanying apoptosis in human T cell lines (14). In addition to its role in regulating many aspects of T cell function, galectin-1 appears to have a role in stromal cells. Galectin-1 is expressed in bone marrow stromal cells (15, 16), and secreted galectin-1 anchors to integrins to interact with pre-BCR and acts as a survival signal for pre-B cells during development (15, 17). In late-stage B cell activation and maturation, soluble galectin-1 produced by activated B cells resulting from Trypanosoma cruzi infection in mice causes T cell apoptosis and affects IFN-γ production (18). Intracellular galectin-1 may associate with the B cell-specific Oct-1-associated coactivator, OCA-B, to negatively regulate BCR signaling (19). Whether there is a role for galectin-1 in plasma cell differentiation, however, remains elusive.

Plasma cell differentiation is regulated by a master regulator, termed B lymphocyte-induced maturation protein-1 (Blimp-1) (20, 21). B cell-specific deletion of Prdm1, the gene encoding Blimp-1, in mice showed that in response to either thymus-dependent or thymus-independent Ags, short-lived plasma cells, postgerminal center plasma cells, and plasma cells in a memory response are absent (22). Blimp-1 is a transcriptional repressor (23). Our previous microarray study revealed that galectin-1 is up-regulated upon ectopic expression of Blimp-1 in mature human B cell lines (24). Herein, we investigated the importance of Blimp-1-dependent induction of galectin-1 during plasma cell differentiation. We found that the effect of galectin-1 on promoting Ig production appears to be mediated through an extracellular receptor(s) and to depend on the binding of β-galactosides before terminal differentiation of B cells. Our results may broaden current knowledge of lectins in modulating B cell function.

BCL-1 cells were maintained in RPMI 1640 (Invitrogen) containing 10% FBS (Invitrogen) and 50 μM 2-ME (Invitrogen); CESS and MOLT-4 cells were grown in RPMI 1640 containing 10% FBS. All the abovementioned cell lines were grown in medium containing penicillin/streptomycin (100 U/ml; Invitrogen). IL-2 (20 ng/ml; eBioscience) and IL-5 (20 ng/ml; eBioscience) were used to stimulate BCL-1 cells. Splenic B cells were purified using B220 microbeads (Miltenyi Biotec) from 12–16-wk-old C57BL/6 mice (purchased from BioLASCO Taiwan), Prdm1^f/fCD19^Cre+/+, or littermate control Prdm1^f/fCD19^+/+ mice, as described previously (22). Purified splenic B cells (purity >95%) were cultured as described (25) and were stimulated with LPS (2 μg/ml; Sigma-Aldrich), IL-2 (20 ng/ml) + IL-5 (20 ng/ml), IL-4 (10 ng/ml; PeproTech) + CD40L (1 ng/ml; PeproTech) or LPS + anti-IgM (5 μg/ml; Sigma-Aldrich). Recombinant galectin-1, with endotoxin level <1.0 EU/μg (determined by Limulus amebocyte lysate method), was purchased from R&D Systems. Human PBMC were isolated from healthy donors and grown as described (25). CD19⁺ human B cells, at a density of 2 × 10⁶ cells/ml, were stimulated with IL-21 (500 ng/ml; BioSource) + anti-CD40 (1 μg/ml; R&D Systems) + IL-2 (100 U/ml; PeproTech) + anti-IgM (5 μg/ml; Jackson ImmunoResearch Laboratories). 293T and 3T3 cells were maintained in DMEM supplemented with 10% FBS and penicillin/streptomycin (100 U/ml). 293F cells were maintained in FreeStyle 293 expression medium (Invitrogen). Lactose and sucrose were purchased from Sigma-Aldrich. Sugar compounds, N-acetyllactosamine (LacNAc), and Lac-di-NAc (provided by the Consortium of Functional Glycomics) were dissolved in distilled water.

pGC-yellow fluorescent protein (YFP), pGC-HABlimp-YFP, pFUGW, and pFUW-BlimpGFP (a gift from Dr. Kathryn Calame, Columbia University, New York, NY) were described previously (26, 27). Full-length human galectin-1 was amplified by RT-PCR from H929 multiple myeloma cells. The amplified product was cloned into a pCMV2-FLAG vector (Sigma-Aldrich) at BamHI/NotI sites to generate N-terminally FLAG-tagged galectin-1. Mutated human galectin-1, W69G, was generated by mutagenesis using the following primers with the mutated bases underlined: 5′-ACGGCGGGGCCGGGGGGACCGAG-3′ and 5′-CTCGGTCCCCCCGGCCCCGCCGT-3′. The FLAG-tagged galectin-1 and FLAG-tagged W69G PCR products were then blunt-end digested and subcloned into vector pGC-YFP at the NotI site to generate pGC-NGal-1-YFP and pGC-W69G-YFP, respectively. To generate a galectin-1-hIgGFc^m (Gal-1-Fc) or W69G-hIgGFc^m (W69G-Fc) fusion gene, primers 5′-ATTAGGCCCAGCCGGCCGCTTGTGGTCTGGTCGCCAGCAACC-3′ and 5′-GGATGCGGCCGCGCGTCAAAGGCCACACATTTGATCTTG-3′ were used to PCR amplify from pCMV2-FLAG galectin-1 or pCMV2-FLAG-W69G vector. The amplified product was ligated in-frame into the pSecTagFc^m vector (a kind gift from Dr. H.-H. Lin, Chang-Gung University, Taiwan) at the Sfi/NotI sites.

Total RNA isolation, cDNA synthesis, and subsequent QPCR analysis in an ABI Prism 7000 sequence detection system (Applied Biosystems) or RT-QPCR analysis was according to published protocols (25). The TaqMan primer sets used here were: Prdm1 (assay ID: Mm 00476128_ml), L32 (assay ID: Mm 00777741_sH), Lgals1 (assay ID: Mm 00777741_sH), peptidylprolyl isomerase A (PPIA) (assay ID: Hs99999904_ml), and LGALS1 (assay ID: Hs 00169327_ml), all purchased from Applied Biosystems. The primers used for SYBR Green detection were: GAPDH, 5′-TTAGCACCCCTGGCCAAGG-3′ and 5′-CTTACTCCTTGGAGGCCAT-3′; μs (secreted form of μ-chain transcript), 5′-TCTGCCTTCACCACAGAAG-3′ and 5′-TAGCATGGTCAATAGCAGG-3′; μm (membrane form of μ-chain transcript), 5′-AGGACAGCAGAGACCAAGAGAT-3′ and 5′-GCCAGACATTGCTTCAGAT-3′; and total Ig μ-chain mRNA, 5′-ATCTGCATGTGCCCATTC-3′ and 5′-TTAGGATGTCTGTGGAGG-3′.

Cells were lysed in 2× SDS loading buffer as reported (22). The procedures for performing immunoblotting essentially followed published protocols (25, 28). Abs specific to Blimp-1 or α-tubulin were as described (29). The anti-galectin-1 Ab was purchased from Santa Cruz Biotechnology (1/500 dilution was used). The immunoreactive proteins were detected by an enhanced chemiluminescence system (Amersham Biosciences) according to the manufacturer’s protocol. Chemiluminescent signals were captured using a CCD camera (Fujifilm LAS-3000).

Gal-1-Fc or W69G-Fc fusion proteins were overexpressed using the FreeStyle 293 expression system as reported (30). Briefly, 30 μg pSecTaggalectin-1-hIgGFc^m or pSecTagW69G-hIgGFc^m was used to transfect 3 × 10⁷ 293F cells with 293fectin. At 48 and 72 h posttransfection, culture supernatants were collected for further purification. The cultured medium was passed through a protein A column (Amersham Biosciences), and the bound proteins were then eluted with 0.1 M glycine buffer (pH 3.0). The purity of the recombinant protein was analyzed by 8% SDS-PAGE and visualized by Coomassie blue staining.

Cell surface marker staining was performed as described (28). Fluorescence intensity was analyzed by FACSCanto (BD Biosciences) and FCS Express 3.0 software. The Abs used in this study (all from BD Pharmingen) were: PE-conjugated anti-mouse CD138/syndecan-1 (clone 281-2), allophycocyanin-conjugated anti-mouse CD45R/B220 (clone RA3-6B2), FITC-conjugated anti-human IgD (clone IA6-2), and APC-conjugated anti-human CD38 (clone HB7). Apoptotic and dead cells stained by annexin V and 7-aminoactinomycin D (7-AAD), respectively, were determined as described (29). To detect the binding of galectin-1, 10 μg of Gal-1-Fc or W69G-Fc was added to the binding buffer (PBS plus 2% FBS in 100 μl) with 1 × 10⁵ cells for 60 min, followed by the addition of FITC-conjugated anti-human IgG (Fc-specific; Sigma-Aldrich) for 30 min. To detect the blocking of Gal-1-Fc binding by sugar compounds, various concentrations of sugar compounds were incubated together with 10 μg Gal-1-Fc before the addition of secondary Ab. For the B cell proliferation assay, mouse splenic B cells (2 × 10⁷ cells/ml in PBS) were stained with 1 μM CFSE (Sigma-Aldrich) for 8 min at 25°C. Cells were then washed twice with complete medium and plated as previously described (22).

Cell supernatants were harvested for ELISA to determine the amount of IgM or IgG as described (25). For galectin-1 ELISA, supernatants were serially diluted in PBS containing 1% BSA into 96-well plates coated with goat anti-mouse galectin-1 (100 μg/ml) (R&D Systems). Captured galectin-1 was further incubated with biotinylated anti-mouse galectin-1 (50 μg/ml) (R&D Systems) and streptavidin-HRP conjugate (100 μg/ml) (R&D Systems). ELISPOT analysis for detecting IgM-secreting cells isolated by YFP-positive sorting essentially followed a reported procedure (31). Photomicrographs of the ELISPOT assay were read and analyzed by the AID EliSpot Reader System (AID Autoimmun Diagnostika).

The preparation of retroviral vectors and transduction of virus followed a described protocol (24, 32). We transduced BCL-1 and CESS cell lines at a multiplicity of infection of 2–5 and primary splenic B cells at a multiplicity of infection of 10–15. For cells transduced with retroviral vector-expressing YFP or lentiviral vector expressing GFP, YFP, or GFP, positive cells were sorted on FACSAria (BD Biosciences) for additional experiments.

The carbohydrate-binding profile of galectin-1 was determined by the Core H, Consortium for Functional Glycomics, using a printed glycan microarray. Briefly, Gal-1-Fc (200 μg/ml) in binding buffer (1% BSA, 150 mM NaCl, 2 mM CaCl₂, 2 mM MgCl₂, 0.05% (w/v) Tween 20, and 20 mM Tris-HCl (pH 7.4)) was applied onto covalent printed glycan array (version 2.1) slides and incubated for 1 h at room temperature. A secondary incubation was conducted using Alexa Fluor 488-conjugated goat anti-human IgG (10 μg/ml in binding buffer). Slides were scanned, and the average signal intensity (shown as “Avg” in Table I) was calculated. The common features of glycans with stronger binding are depicted in Fig. 6A. The average signal intensity detected from all glycans was calculated and set as the baseline.

Data were analyzed statistically using the Student’s t test. p values of <0.5 (two-sided tests) were considered statistically significant.

A previous microarray study of changes in gene expression upon introduction of Blimp-1 into human B cell lines demonstrated that galectin-1 expression is induced under these circumstances (24). We therefore sought to examine a possible causal relationship between Blimp-1 and galectin-1 expression. The lentiviral vector expressing either GFP alone or GFP-tagged Blimp-1 (27) was used to transduce CESS cells, an EBV-transformed human lymphoblastoid B cell line. GFP⁺ CESS cells ectopically expressing Blimp-1 (GFP-Blimp-1) showed a 10-fold increase in IgG production compared with GFP⁺ CESS cells (GFP) (Fig. 1,A). We confirmed that GFP-Blimp-1 cells had increased levels of galectin-1 mRNA (Fig. 1,B) and protein (Fig. 1 C).

We next examined expression of galectin-1 in response to stimulation by various cytokines and mitogens. We found that induction of galectin-1 mRNA was concomitant with induction of Blimp-1 mRNA following T cell-independent LPS stimulation in primary mouse splenic B cells (Fig. 1,D) and that galectin-1 was secreted into the culture medium in this context (Fig. 1,E). Two T cell stimuli, IL-2 + IL-5 and IL-4 + CD40L, also induced galectin-1 mRNA expression (Fig. 1,F). Splenic B cells in which the Blimp-1 gene Prdm1 is deleted show severe impairment of IgM secretion and induction of CD138 when stimulated with LPS in vitro (22). We examined the levels of secreted galectin-1 from Prdm1^f/fCD19^Cre+/+ cells, Blimp-1 knockout (KO) splenic B cells, and from Prdm1^f/fCD19^+/+ (control) splenic B cells in response to LPS stimulation. We found that the levels of secreted galectin-1 were dramatically diminished in Prdm1-deficient B cell culture (Fig. 1 G). Thus, galectin-1 is induced during plasma cell differentiation in a Blimp-1-dependent manner.

IL-21 + anti-CD40 combined with IL-2 + anti-IgM treatment induces human peripheral B cells to form Ab-secreting cells (33). We took advantage of this fact to examine the correlation between galectin-1 and Blimp-1 mRNA expression during human peripheral B cell differentiation. Indeed, human peripheral B cells showed significant induction of IgG production (data not shown), induction of CD38 plasma cell surface marker (see Fig. 5D), and induction of Blimp-1 mRNA (Fig. 1,H, right panel) under this culture condition on day 6 (33). As expected, we observed that the induction of Blimp-1 mRNA was accompanied by induction of galectin-1 mRNA on day 6 (Fig. 1 H, left panel).

To directly test the function of galectin-1 during plasma cell differentiation, we transduced B cells with a bicistronic retrovirus carrying the cDNAs encoding FLAG-tagged galectin-1 and YFP (pGC-NGal1-YFP). Expression of FLAG-tagged galectin-1 was confirmed by immunoblotting using an Ab to FLAG (Fig. 2,A). Notably, ectopic expression of galectin-1 in BCL-1 cells, a mouse B cell lymphoma line capable of producing IgM in response to cytokines, led to increased IgM secretion compared with control pGC-YFP-expressing cells (Fig. 2,B). Similar results were obtained for splenic B cells ectopically expressing galectin-1 that were stimulated with LPS + anti-IgM, a treatment that causes B cell proliferation but not differentiation (Fig. 2,C). Furthermore, the ectopic expression of galectin-1 enhanced the formation of IgM-secreting cells induced by suboptimal doses of LPS in splenic B cells (Fig. 2 D). A similar effect of ectopic expression of galectin-1 on promoting Ig production was also found in CESS cells (data not shown). These data suggest that galectin-1 induced during plasma cell differentiation may have a role in promoting Ig production.

Galectin-1 has been suggested to function in a variety of cellular compartments (34). In most cases, the intracellular function of galectin-1 is independent of its carbohydrate-binding ability (2, 34). To further address which cellular localization of galectin-1 promotes Ig production, we generated a mutant galectin-1, W69G (7, 35), a β-galactoside binding and export mutant containing a point mutation that changes Trp at residue position 69 to Gly (W69G). The W69G mutant, which remained intracellular, allowed us to examine whether extracellular galectin-1 is responsible for promoting Ig production. We found that BCL-1 cells and splenic B cells expressing W69G galectin-1 failed to increase Ig production compared with cells expressing wild-type galectin-1 (Fig. 2,B–D), although the protein levels of intracellular FLAG-tagged wild type and W69G galectin-1 in transduced cells were comparable (Fig. 2,A). Furthermore, exogenously added lactose, a known galectin-1 ligand, blocked the enhanced IgM production by pGC-NGal1-YFP transduction in BCL-1 cells; in contrast, sucrose, which does not bind galectin-1, failed to influence IgM production (Fig. 2 E). These data suggest that galectin-1 enhances Ig secretion in a carbohydrate- and an extracellular-dependent manner.

To further investigate the molecular mechanism by which galectin-1 promotes Ig secretion, we examined the levels of total H (μ)-chain transcripts and the ratio of the secreted form (μs) to membrane form (μm) of μ-chain transcripts in pGC-NGal1-YFP- or pGC-YFP-transduced BCL-1 cells. The levels of each form of mRNA in sham and IL-2 + IL-5-treated cells were included as negative and positive controls, respectively. We found that pGC-NGal1-YFP-transduced BCL-1 cells showed a 2.0-fold increase in the total amount of μ-chain transcripts compared with that in pGC-YFP-transduced cells (Fig. 2,F). This extent of increase of the μ-chain transcript was similar to that of BCL-1 cells stimulated with IL-2 + IL-5 (1.7-fold), suggesting that galectin-1 up-regulates μ-chain mRNA expression. Changes in the ratio of μm to μs mRNA upon ectopic expression of galectin-1 were also monitored. We found that ectopic expression of galectin-1 caused only a 1.5-fold increase in the ratio of μm to μs mRNA expression, in contrast to the dramatic change in the ratio (10-fold) upon IL-2 + IL-5 stimulation of BCL-1 cells (Fig. 2 G), suggesting that galectin-1 does not dramatically influence splicing of the μ-chain mRNA. These results suggest that galectin-1 promotes μ-chain mRNA transcription during plasma cell differentiation.

To further establish the role of extracellular galectin-1 in Ig secretion, we tested whether exogenously added recombinant galectin-1 could enhance IgM production. The purified recombinant galectin-1 was found to be endotoxin free (data not shown), and it had biological function because it increased the cell surface exposure of PS in a dose-dependent manner, as determined by annexin V staining, in MOLT-4 T cells as reported (Fig. 3,A) (14). We found that, similar to the effect of ectopic galectin-1 expression, exogenously added recombinant galectin-1 increased IgM production in splenic B cells (Fig. 3,B). The effect of recombinant galectin-1 on promoting Ig production was blocked by the addition of lactose in a dose-dependent manner (Fig. 3,C). However, unlike its ability to induce PS exposure in the T cell lineage, exogenously added recombinant galectin-1 did not increase the percentage of annexin V-positive or trypan blue-stained splenic B cells (Fig. 3, D and E). In fact, incubation with exogenous recombinant galectin-1 reduced spontaneous cell death of splenic B cells, as determined by trypan blue staining (Fig. 3 E). This effect could be due to the activity of recombinant galectin-1 to activate B cells. Collectively, these data demonstrate that extracellular galectin-1 can promote Ig production, but it does not induce PS exposure in B cells.

Because galectin-1 works downstream of Blimp-1 and is capable of promoting Ig production, we sought to determine whether ectopic expression of galectin-1 could rescue the defect of Ig secretion and of CD138 induction in Prdm1-deficient B cells in response to LPS. As shown in Fig. 4, in contrast to the effect of reintroduction of Blimp-1, we found that the percentage of CD138⁺ cells and the amounts of IgM-secreting cells from infected populations remained diminished when introducing galectin-1 to LPS-stimulated Prdm1-deficient B cells, indicating that ectopic expression of galectin-1 could not compensate for Blimp-1 deficiency.

Because we observed that galectin-1 is induced and has a role in promoting Ig production, we wished to examine the pattern of galectin-1 binding during plasma cell differentiation. Plasmids encoding recombinant galectin-1-hIgGFc^m (Gal-1-Fc) or W69G- hIgGFc^m (W69G-Fc) were generated, and the proteins were purified with a protein A column (30). The fusion protein contained full-length galectin-1 or W69G in the N-terminal and Fc^m portion of human IgG in the C-terminal, which allows it to be detected by anti-human IgG. Fc^m used here prevented binding to cell surface Fc receptors as described (30); W69G-Fc was a control to exclude the possibility of nonspecific binding by Gal-1-Fc. The expression of purified recombinant fusion protein was monitored (Fig. 5,A). We used Gal-1-Fc to monitor the binding of galectin-1 to B cells and found that Gal-1-Fc bound to purified mouse splenic B cells (Fig. 5,B). This binding was dependent on galactoside recognition, as neither incubation with W69G-Fc alone (Fig. 5,B, top) nor with Gal-1-Fc in the presence of lactose in the binding reaction (Fig. 5,B, bottom) showed binding. Unexpectedly, following LPS-mediated differentiation for 3 days, we found reduced binding of galectin-1 to B220^lowCD138⁺ plasma cells compared with the B220^highCD138⁻ undifferentiated populations (Fig. 5,C). We also examined whether galectin-1 bound to human B cells when stimulated to differentiate by IL-2 + IL-21 and anti-CD40 + anti-IgM. Likewise, we observed that Gal-1-Fc bound to unstimulated mature B cells (Fig. 5,D, top), but after 6 days of stimulation almost no Gal-1-Fc binding to CD38^highIgD⁻ plasma cells was observed (Fig. 5 D, bottom). Therefore, our data suggest that galectin-1 binds better to mature B cells than to plasma cells.

To further understand which glycan ligands are recognized by galectin-1, we screened Gal-1-Fc for carbohydrate ligands by the glycan microarray screening service available at the Consortium of Functional Glycomics. As expected, we found that Gal-1-Fc had the highest affinity for sugar compounds containing multiple Galβ1–4 N-acetylglucosamine (GlcNAc) (LacNAc), Galβ1–4Glc (lactose), or Galβ1–3GlcNAc units (Fig. 6,A) (36). Glycans with an average signal intensity for Gal-1-Fc binding >2-fold that of the baseline (an intensity index of >2.00) are shown in Table I. α-2,6-Sialylation of the galactose residue or fucosylation of the N-acetylglucosamine abolished Gal-1-Fc binding (data not shown).

Based on the sugar array profile (Fig. 6,A and Table I) and the fact that galectins recognize multiple LacNAc units present on the branches of N- or O-linked glycans (37, 38), we asked if LacNAc could block the binding of galectin-1 to B cells. We used lactose as a positive control in this experiment. Sucrose and N-acetylgalactosamine (GalNAc)β1–4GlcNAc (Lac-di-NAc) were used as negative controls. Notably, we found that the binding of Gal-1-Fc to murine splenic B cells was efficiently blocked by LacNAc in a dose-dependent manner (Fig. 6,B). In fact, LacNAc was a more potent inhibitor than lactose in blocking galectin-1 binding to B cells, as LacNAc at 1 mM greatly reduced Gal-1-Fc binding, whereas a dose of 100 mM lactose was required to achieve a similar blockade (Fig. 6,B). Sucrose at 100 mM and Lac-di-NAc at 1 mM did not affect Gal-1-Fc binding (Fig. 6,B). We next determined whether LacNAc could block murine splenic B cell differentiation. As shown in Fig. 6,C, LacNAc at 10 mM and 1 mM significantly reduced LPS-mediated induction of IgM in splenic B cells, whereas Lac-di-NAc did not affect IgM production. Because B cell stimulation followed by differentiation requires cell division (39), we labeled purified splenic B cells with the cell division tracking dye CFSE and stained cells for CD138 expression after 3 days in culture with LPS plus various doses of LacNAc or Lac-di-NAc to monitor the effects of sugar compounds on B cell proliferation and induction of CD138 following LPS stimulation. Results in Fig. 6,D show that LacNAc at higher doses (10 mM and 1 mM) interfered more obviously with the induction of CD138^high. The effect of LacNAc was not completely due to its ability to alter LPS-induced cell proliferation because reduced frequencies of CD138⁺ cells were observed at concentrations of LacNAc (1 mM and 0.1 mM) that did not affect CFSE dye dilution (Fig. 6 D). Taken together, these data show a correlation between abrogation of galectin-1 binding to mature B cells by LacNAc and the effect of LacNAc on the inhibition of plasma cell differentiation.

Galectins have a variety of roles in regulating immune cell homeostasis; however, the role of galectin-1 in plasma cell differentiation in particular has been unclear. We demonstrate here for the first time that, during plasma cell differentiation, induction of galectin-1 is Blimp-1 dependent. Galectin-1 increases the levels of μ-chain transcripts and Ig production during plasma cell differentiation. Because exported galectin-1 was responsible for this activity and did not bind to cells with plasma cell surface markers, the data suggest that the effect of extracellular galectin-1 is to promote Ig production by B cells after their activation, rather than to induce Ig secretion by plasma cells. Our findings are summarized in Fig. 7.

Galectin-1 functions in various cellular compartments. The intracellular activities of galectin-1 appear to be independent of its lectin activity (2). We showed that a galectin-1 mutant, W69G, is impaired in its ability to promote Ig production, that exogenously added lactose blocks the induction of Ig caused by ectopic expression of galectin-1, and that exogenously added recombinant galectin-1 increases Ig production, indicating that binding to cell surface counter-receptors is required for galectin-1 function during plasma cell differentiation. Additionally, blocking the binding of galectin-1 to murine splenic B cells by LacNAc abrogated LPS-stimulated plasma cell differentiation (Fig. 6); this also supports the notion that galectin-1 requires carbohydrate recognition for its activity during plasma cell differentiation. Numerous galectin-1 receptors, including CD45, CD7, CD43, CD2, CD3, CD4, CD7, integrins, fibronectin, laminin, and pre-BCR have been reported (4). Our finding that galectin-1 preferentially binds to mature B cells (Fig. 5) suggests that these cells contain potential galectin-1 receptors. One potential receptor is CD45; it has been implicated as the galectin-1 receptor in Burkitt’s lymphoma B cells (40), and its expression is down-regulated in plasma cells (41). Alternatively, galectin-1 may preferentially bind to mature B cells because only they retain certain glycosylation modification patterns that galectin-1 recognizes. In support of the latter hypothesis, a recent publication indicated that Th1, Th2, and Th17 cells have differential surface glycosylation patterns, revealed by plant lectin binding, thus rendering them differentially susceptible to galectin-1-mediated cell death (13). Nevertheless, further work is needed to elucidate the key surface glycoconjugates that mediate the effect of galectin-1 during plasma cell differentiation. Additionally, our results cannot exclude the possibilities that other galectins might be involved in modulating Ig production because several galectins bind LacNAc with differential affinity (37, 38). However, the role and mechanism of regulation of the expression of other galectins during B cell activation/differentiation remain unclear. Galectin-3 is induced during B cell activation by IL-4 or anti-CD40, but its induction seems to favor the memory B cell fate (42). Therefore, it remains to be investigated whether other galectin family members participate in modulating B cell activation/differentiation.

Are there other possible functions for galectin-1 secreted by activating/differentiating B cells in addition to promoting Ig production? Galectin-1 can induce phenotypic and functional maturation in human monocyte-derived dendritic cells (43). Also, several lines of study have indicated that galectin-1 is involved in regulating apoptosis in activated T cells (10, 44). Thus, galectin-1 may act as a surveillance factor by which activated B cells eliminate or change the properties of activated T cells during immune responses in secondary lymphoid tissues. Unlike its known role in increasing the number of annexin V-positive T cells, we found that galectin-1 had no significant role in regulating apoptosis in mature B cells: neither ectopic expression of galectin-1 by retroviral vector (data not shown) nor exogenously added recombinant galectin-1 increased cell death or annexin V-positive cells in B cells in this study (Fig. 3, C and D). It has been shown that in early stages of B cell development, the galectin-1/pre-BCR complex functions in pre-B cell proliferation (15), which also suggests that galectin-1 does not cause B cell death.

The mechanism underlying the ability of galectin-1 to promote Ig production remains unclear. Intracellular galectin-1 has been implicated in regulating pre-mRNA splicing in a complex with galectin-3 (45, 46, 47). Blimp-1 also has a possible role in regulating the switch from μm to μs (22). Our results suggest that it is unlikely that the role of Blimp-1 in affecting splicing of μm mRNA to μs mRNA is through the action of galectin-1 because we did not find significant change in the ratio of μs/μm transcripts following ectopic expression of galectin-1 (Fig. 2,G). We found, however, that total levels of μ-chain mRNA were elevated (Fig. 2,F), suggesting that galectin-1 may trigger a signaling pathway(s) to increase μ-chain expression. One clue in this regard is that the effect of galectin-1 in promoting Ig production is not by up-regulating the expression of Blimp-1 or the activated form of XBP-1 (XBP-1s) because we did not find changes in the levels of Blimp-1 or XBP-1s mRNA in BCL-1 cells ectopically expressing galectin-1 (data not shown). Thus, galectin-1 may signal through the unidentified counter-receptors that function independently of triggering Blimp-1 induction during B cell activation/differentiation. Although galectin-1 enhanced Ig production ∼2-fold, we found that exogenous galectin-1 could not rescue the deficiency of Blimp-1 in splenic B cell culture (Fig. 4). This observation is similar to a previous finding that ectopic expression of XBP-1s cannot compensate for a Prdm1 deficiency (22). These data suggest that multiple Blimp-1 targets are required to coordinately regulate Ig production. In conclusion, we have established a previously undefined role and mechanism for Blimp-1-dependent induction of galectin-1 in promoting Ig production during plasma cell differentiation.

The authors thank Drs. Chi-Huey Wong and Tse Wen Chang for discussions and critical reading of the manuscript, Hui-Kai Kuo and Wen-Wen Chen for technical assistance, Consortium of Functional Glycomics (Grant GM62116) for the gift of the sugar compounds and performing the Gal-1-Fc glycan array, and Dr. Rachel Ettinger (National Institutes of Health, Bethesda, MD) for human peripheral B cell culture advice.

The authors have no financial conflicts of interest.