- Email: [email protected]
Strongly enhanced IL-10 production using stable galectin-1 homodimers
Molecular Immunology 44 (2007) 506–513
Strongly enhanced IL-10 production using stable galectin-1 homodimers Judith van der Leij a , Anke van den Berg a , Geert Harms a , Hugo Eschbach a , Hans Vos a , Peter Zwiers b , Rob van Weeghel b , Herman Groen b , Sibrand Poppema a , Lydia Visser a,∗ a
Department of Pathology and Laboratory Medicine, University Medical Center Groningen and University of Groningen, Groningen, The Netherlands b IQ Corporation B.V., Groningen, The Netherlands Received 2 August 2005; received in revised form 14 February 2006; accepted 15 February 2006 Available online 3 April 2006
Abstract Galectin-1 is the homodimeric form of a protein, which is present in a dynamic equilibrium with the -galactoside monomeric form and has potent anti-inflammatory and immunomodulating effects. These favorable effects are probably related to the induction of apoptosis in activated T cells and the induction of IL-10, which have been demonstrated to be characteristic for the dimeric form of the protein. Based on these findings it can be speculated that the in vivo effects of galectin-1 can be improved by the generation of stable galectin-1 homodimers (dGal). To test this hypothesis we produced leucine-zipper based stable galectin-1 homodimers and tested its apoptosis inducing effects on MOLT-4 cells and its immunomodulatory effects in vitro on PBMC of five independent donors. Phosphatidylserine exposure and a drop in mitochondrial membrane potential was strongly enhanced on MOLT-4 cells upon treatment with dGal as compared to wtGal. The minimal effective concentration was 20-fold reduced as compared to the minimal effective wtGal concentration. dGal showed enhanced induction of IL-10 on total PBMC as compared to treatment with wild-type protein (wtGal). The minimal effective dGal concentration was 100-fold lower than that of wtGal. Of the purified cell populations monocytes are the strongest IL-10 producers, whereas T cells induce IL-10 at a lower level and no induction is observed in B cells. Besides induction of IL-10, dGal caused an increase in IL-1 production in all donors and a reduction of IL-2 production in 3 out of 5 donors, whereas no consistent changes were observed for other inflammatory cytokines. In summary, we demonstrated that dGal shows enhanced effects at strongly reduced concentrations. Application of dGal may therefore serve as an improved treatment of chronic inflammatory diseases. © 2006 Elsevier Ltd. All rights reserved. Keywords: Stable galectin-1 homodimers; IL10; Apoptosis
1. Introduction Galectin-1 is a homodimeric protein, consisting of two identical subunits from 14.5 kDa (Barondes et al., 1994; Cho and Cummings, 1995). The protein exists in a reversible monomerdimer equilibrium, with a dissociation constant (Kd ) of ∼7 M (Cho and Cummings, 1995). Galectin-1 binds to lactosamine residues present on N- and O-linked glycans, which are present on glycosylated T cell surface receptors like CD45, CD43, CD7, CD2, and CD3 (Walzel et al., 1999, 2000; Pace et al., 1999, 2000; Elola et al., 2005). The protein has a wide tissue distribution, and plays a role in many biological processes, including cell adhe∗
Corresponding author at: Department of Pathology and Laboratory Medicine, University Medical Center Groningen, Hanzeplein 1, P.O. Box 30.001, 9700 RB Groningen, The Netherlands. Tel.: +31 50 361 0402; fax: +31 50 363 2510. E-mail address: [email protected] (L. Visser). 0161-5890/$ – see front matter © 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.molimm.2006.02.011
sion (Rabinovich et al., 1999a; Rabinovich, 1999b), proliferation (Allione et al., 1998), and the induction of apoptosis in thymocytes and activated T cells (Perillo et al., 1995, 1997). Apoptosis induction in vitro is only observed using high concentrations of the galectin-1 protein (concentrations above 7 M) (Perillo et al., 1995), which suggests that the dimeric form of the protein is required to induce apoptosis. More evidence was provided in a study with a monomeric mutant form, which binds to cells but even at very high concentrations is unable to induce apoptosis of HL-60 and MOLT-4 cells (Dias-Baruffi et al., 2003). Using a covalently linked dimeric form of galectin-1, apoptosis induction in EL-4 thymoma cells and thymocytes from C57Bl/6 mice was increased three to 10-fold (Battig et al., 2004). Although the glycosylation status of CD45 is able to affect galectin-1 induced apoptosis (Nguyen et al., 2001), the presence of CD45 is not a prerequisite (Fajka-Boja et al., 2002). In contrast, CD7 appears to be necessary for induction of apoptosis mediated by galectin-1 (Pace et al., 2000).
J. van der Leij et al. / Molecular Immunology 44 (2007) 506–513
Recently, we demonstrated that dimeric, but not monomeric concentrations of the galectin-1 protein induce IL-10 production in CD4+ and CD8+ T cell subsets (van der Leij et al., 2004). It can be speculated that the IL-10 induction is responsible for the beneficial role of galectin-1 observed in several animal models of inflammatory (Santucci et al., 2000) or autoimmune diseases (Levi et al., 1983; Offner et al., 1990; Rabinovich et al., 1999c; Santucci et al., 2003). In these models, an increased level of leukocyte apoptosis and a downregulation of pro-inflammatory cytokines TNF-␣, IL-1, IL-12, IFN-␥, and upregulation of IL-5 is detected. The in vivo efficacy of galectin-1 as observed in previous studies may still be limited since processes as apoptosis and IL-10 induction require the dimeric form of the galectin-1 protein. We speculated that the use of stable galectin-1 homodimers might enhance the beneficial effects observed with high concentrations of wild-type galectin-1. These stable galectin-1 dimers might thereby provide a very attractive means for the treatment of inflammatory and autoimmune diseases. In this study we report the construction of stabilized galectin-1 homodimers (dGal) and the analysis of the effects on apoptosis, IL-10 and other cytokines upon treatment of peripheral blood mononuclear cells in comparison to the effects observed with high concentrations of normal/wild-type galectin-1 protein (wtGal). 2. Materials and methods 2.1. Medium The culture medium used throughout the experiments was RPMI 1640 supplemented with 2 mM L-glutamine, 10% heatinactivated fetal bovine serum (FBS), 100 U/ml penicillin, and 100 g/ml streptomycin (Cambrex, Verviers, Belgium). In culture experiments with the galectin-1 protein, the culture medium was additionally supplemented with 1.2 mM dithiothreitol (DTT). 2.2. Preparation of recombinant human galectin-1 protein Wild-type recombinant human galectin-1 protein (wtGal) was prepared as described previously (van der Leij et al., 2004). For the construction of stable galectin-1 homodimers (dGal), we used a FOS leucine zipper based construct, which was kindly provided by de Kruif and Logtenberg (1996). Between the FOS leucine zipper and galectin-1, a hinge region was placed functioning as a flexible linker. The FOS leucine zipper was flanked by CGG and GGC amino acids at the N- and C-terminus, respec-
507
tively, to covalently link the zippers by disulfide bonds between cysteine residues (Fig. 1). 2.3. Measurement of phosphatidylserine exposure and mitochondrial membrane potential Phosphatidylserine exposure on the cell surface was measured using a phosphatidylserine detection kit (IQP, Groningen, The Netherlands). Mitochondrial membrane potentials were assessed using 3,3 -dihexiloxa dicarbocyanine iodide (DiOC6) (Molecular Probes, Leiden, The Netherlands). MOLT-4 T cells were cultured at 37 ◦ C for 3 h (1 × 106 ml−1 in RPMI/10%FCS/1.2 mM DTT) with different concentrations of galectin-1 protein. To dissociate galectin-1 from the cell membrane cells were treated with 0.1 M lactose/PBS for 10 min at room temperature. For phosphatidyl exposure the cells were resuspended in 110 l calcium buffer (10 mM HEPES, 140 mM NaCl, 2.5 mM CaCl2 , pH 7.4) containing 2.5 l AnnexinV-Fitc. Cells were incubated for 20 min on ice, washed once with calcium buffer, suspended in 160 l calcium buffer containing 1 l propidium iodide (PI), and incubated for 10 min on ice. For measurement of the mitochondrial membrane potential cells were mixed with DiOC6 (final concentration 0.1 M) and incubated at 37 ◦ C for 30 min. After washing PI was added at a final concentration of 0.1 g/ml. Cells were immediately analyzed by flow cytometry, 10,000 events per sample, on a Coulter Epics-Elite flow cytometer (Coulter Corporation, Hialeah, FL, USA). Data were analyzed using WinList 4.0 software (Verity Software House Inc., Topsham, ME, USA). 2.4. Galectin-1 treatment and cytokine measurements Peripheral blood mononuclear cells (PBMC) were obtained from healthy volunteers and cells were isolated by Ficoll–Hypaque centrifugation. Isolated PBMC (1 × 106 ml−1 ) were stimulated for 6 or 24 h with different concentrations of galectin-1 protein, with or without ␣CD3 (10 ng/ml) activation. For inhibition assays, galectin-1 protein was pre-incubated for 30 min at room temperature with 0.1 M lactose before addition to cells. Cytokine production was determined in cell-free culture supernatants after 24 h stimulation, using multiplex ELISA. An 8-plex premixed antibody bead kit was purchased from BioSource International Inc. (BioSource, Etten-Leur, The Netherlands). Cytokines analyzed were IL-1, IL-2, IL-4, IL-5, IL-10, IL-12p70, IL-13, and IFN-␥. A standard
Fig. 1. Overview of FOS-hinge-galectin-1 region. Leucine residues (L) in the FOS zipper are indicated in bold. The FOS leucine zipper is flanked by CGG and GGC amino acids (indicated in italic) at the N- and C-terminus, respectively, to covalently link leucine zippers. Behind the FOS zipper, a hinge region was introduced (PKPSTPPGSSH, bold), followed by the galectin-1 protein (underlined).
508
J. van der Leij et al. / Molecular Immunology 44 (2007) 506–513
curve was generated and analyzed in duplicate according to manufacturer’s instructions. ELISA was performed according to manufacturer’s instructions, and samples were analyzed using the Luminex 100TM Instrument (Luminex Corporation, Austin, TX, USA). The concentrations of cytokines in the samples were determined using the standard curves, generated using a five-parameter algorithm in curve fitting software. In indicated experiments, cytokine productions for IL-10 and IL-1 were measured using individual ELISA kits according to manufacturer’s instructions (R&D Systems, Abingdon, UK). We used a nonparametric Mann–Whitney test to define if the differences were statistically significant. 2.5. Real-time PCR analysis for IL-10, IL-1β, and HPRT Total RNA was isolated from cell pellets using the absolutely RNA RT-PCR miniprep kit (Stratagene, La Jolla, CA, USA). RNA (200 ng) and was reverse transcribed with Superscript II reverse transcriptase (Invitrogen, Paisley, UK) in a volume of 20 l using random hexamers (300 ng). Primers (Invitrogen) and probes (Eurogentec, Seraing, Belgium) used for real-time PCR analysis were developed using primer design software (Applied Biosystems, Foster City, CA, USA). Primers used were: IL-10F 5 -atgaaggatcagctggacaactt-3 , IL-10R 5 -ccttgatgtctgggtcttggt-3 ; HPRTF 5 -ggcagtataatccaaagatggtcaa-3 , HPRTR 5 -gtctggcttatatccaacacttcgt-3 . Probe sequences labeled 5 with the FAM reporter dye and 3 with the TAMRA quencher dye molecules were: IL-10 5 acctgggttgccaagccttgtctg-3 , HPRT 5 -caagcttgctggtgaaaaggacccc-3 . An Assay-on-Demand Gene Expression ProductTM (Applied Biosystems) was used for IL-1β (Hs 00174097 m1). Reactions were performed in 384-well plates (Applied Biosystems) in a volume of 20 l containing real-time PCR mastermix (Eurogentec), 900 nM of each primer, 200 nM of an individual probe and 2 ng cDNA. PCR amplifications were performed using the ABI prism 7900HT sequence detection system (Applied Biosystems). Standard cycling conditions were used including a pre-amplification step of 50 ◦ C for 2 min, 95 ◦ C for 10 min, followed by amplification for 40 cycles of 95 ◦ C for 15 s and 60 ◦ C for 1 min. All samples were analyzed in triplicate. Mean cycle treshold values (Ct) and standard deviations (S.D.) were calculated for cytokine and housekeeping genes. The amount of cytokine target was normalized relative to the amount of housekeeping gene (Ct = Ctgene − CtHPRT ) and the S.D. √ of the Ct (S.D.(Ct)) was calculated S.D.(Ct) = ((S.D.gene )2 + (S.D.HPRT )2 ). The relative amount of cytokine was measured by determining the 2−Ct . The range is given as 2−(Ct+S.D.Ct) and 2−(Ct−S.D.Ct) . 2.6. Isolation of individual cell populations PBMC were obtained from four healthy donors and cells were isolated by Ficoll–Hypaque centrifugation. Monocytes and B cells were isolated each with negative isolation kits for monocytes and B cells (Dynal, Oslo, Norway). CD4 and CD8 positive lymphocytes were isolated by staining with fluorochromelabeled antibodies against CD4 and CD8 (IQProducts, Gronin-
gen, The Netherlands), and cells were sorted on the MoFlow cytometer (Cytomation, Fort Collins, CO, USA). Purity of isolated cell populations was checked routinely. PBMC and isolated cell populations were incubated with different concentrations of dGal (0, 0.2, 1 and 2 M) for 24 h at 1 × 106 cells/ml. Cell culture supernatants were harvested and measured for IL-10 production by ELISA. 2.7. Measurement of Lactate dehydogenase (LDH) release as an assessment of cell death PBMC were obtained from healthy volunteers and cells were isolated by Ficoll–Hypaque centrifugation. Isolated PBMC (1 × 106 ml−1 ) were stimulated for 24 h with different concentrations of galectin-1 protein, with ␣CD3 (10 ng/ml) activation in RPMI with 2% FBS. LDH release was measured with the CytoTox-ONE homogeneous membrane integrity assay (Promega, Madison, WI, USA). The percentage of dead cells was calculated by using the maximum LDH release (lysis of total cells). 3. Results 3.1. Enhanced phosphatidylserine exposure and drop in mitochondrial membrane potential using stable galectin-1 homodimers To test the efficiency of dGal to induce apoptosis MOLT4 T cells were treated with dGal concentrations ranging from 0.1 to 5 M and the phosphatidylserine exposure on the cell membrane as determined by annexin V staining as well as the mitochondrial membrane potential as measured by DiOC6 was compared to wtGal concentrations varying from 0.1 to 20 M. Fig. 2A represents the mean of three independent experiments and shows that wtGal induced annexin V binding to 59 ± 3% of the cells at 20 M, whereas the percentage of annexin V positive cells at 5 M was reduced to 42 ± 15%. Using 5 M dGal efficient binding of annexin V to the cell membrane was observed in 77 ± 7% of cells, and with 1 M dGal still 68 ± 9% of the treated cells stained positive for annexin V. In Fig. 2B the results for the DiOC6 staining show that the percentages of cells with a drop in mitochondrial membrane potential are lower then the percentages of AnnexinV positive cells in the same experiments indicating that the exposure of phospatidylserines is probably an earlier or more efficient event. Using 20 M wtGal a drop in membrane potential was shown in 21 ± 5% of cells while with a concentration of 1 M dGal 33 ± 2% of cells show a drop in the mitochondrial membrane potential. Both methods revealed a similar increase of efficiency of the dGal at 20-fold reduced concentrations than wtGal. These data demonstrate a markedly enhanced activity of dGal with respect to the induction of apoptosis. 3.2. Enhanced production of IL-10 using stable galectin-1 homodimers To test the efficiency of IL-10 induction of dGAL in comparison to wtGal, we incubated PBMC with various concentrations of either protein. Stimulation of PBMC with ␣CD3 in com-
J. van der Leij et al. / Molecular Immunology 44 (2007) 506–513
509
Fig. 2. Effect of wild-type galectin-1 and stable galectin-1 homodimers on phosphatidylserine exposure and mitochondrial membrane potential in MOLT-4 T-cells. MOLT-4 T-cells were incubated for 3 h with wild-type galectin-1 (Gal 0.1–20 M) or stable galectin homodimers (dGal 0.1–5 M) and analyzed by flow cytometry. Results are shown as the percentage of AnnexinV positive cells (A). Results are shown as the percentage of cells with a drop in mitochondrial membrane potential (−m) (B). Bars represent the mean and standard deviation of three independent experiments. The dimeric galectin-1 (dGal) is effective at 20-fold lower concentrations as compared to wild-type galectin-1 (wtGal).
bination with low concentrations of wtGal (0.2 or 2 M) did not induce IL-10 protein production, whereas in combination with 20 M wtGal an significant increased IL-10 protein production (p = 0.048) was observed with levels ranging from 279 to 500 pg/ml in four donors and 1367 pg/ml in one donor. Cells treated only with ␣CD3 showed low levels of IL-10 ranging from 58 to 306 pg/ml (Fig. 3A). Cells treated with ␣CD3 in combination with the lowest (0.2 M) dGal concentration already demonstrated significant induction of high IL-10 levels in all five donors (p = 0.027) ranging from 642 to 1171 pg/ml (Fig. 3A). Thus, dGal showed enhanced efficiency for the induction of IL10 production in activated lymphocytes at concentrations of at least 100-fold lower as compared to wtGal. Treatment of resting PBMC with varying concentrations of dGal also revealed enhanced induction of IL-10 at concentrations up to 100-fold lower than the wtGal (not shown). IL-10 mRNA levels were analyzed in the same samples after 6 and 24 h of treatment. After 6 h (Fig. 3B), low levels of IL-10 mRNA were observed in control and wtGal treated cells (relative mRNA quantity range 2–7), whereas IL-10 mRNA levels were enhanced in dGal treated cells at all tested concentrations (IL10 mRNA level range 3–25). After 24 h, effects of galectin-1 on IL-10 mRNA levels were similar to the results at 6 h, although overall the relative IL-10 transcript levels were lower (Fig. 3C). In all five donors, IL-10 mRNA levels were higher at all three tested dGal concentrations as compared to the highest wtGal concentration (Fig. 3B and C).
Fig. 3. Effect of wild-type galectin-1 and stable galectin-1 homodimers on IL10 production. IL-10 protein (A) and IL-10 mRNA (B and C) production by lymphocytes of five independent donors after 24 h (A and C) or 6 h (B) of incubation with ␣CD3 (10 ng/ml) in combination with different concentrations of wild-type galectin-1 protein (wtGal 0.2, 2 and 20 M) or stable galectin-1 homodimers (dGal 0.2, 1 and 2 M). In all five donors a significant increase in IL-10 is seen with 20 M wtGal and all three concentrations of dGal (p = 0.048, 0.027, 0.027 and 0.027, respectively). Lactose inhibition (D) efficiently blocks IL-10 production induced by 1 M dGal. The relative amount of IL-10 mRNA was measured by determining the 2−∆Ct . The range is given as 2−(∆Ct+S.D.∆Ct) and 2−(∆Ct−S.D.∆Ct) .
Pre-incubation with lactose inhibits binding of galectin-1 to the various known cell surface receptors and this also inhibits the wtGal induced production of IL-10 (van der Leij et al., 2004). To test whether lactose pre-treatment displays the same effects on dGal, resting and activated PBMC of different donors were
510
J. van der Leij et al. / Molecular Immunology 44 (2007) 506–513
treated with dGal in different concentrations with or without addition of 0.1 M lactose for 24 h. Fig. 3D demonstrates in a representative experiment that pre-incubation with lactose efficiently blocked IL-10 induction by 1 M dGal both in resting and activated PBMC, with 74% and 70%, respectively. 3.3. Cell populations producing IL-10 after incubation with galectin-1 To establish which cell populations produce IL-10 due to galectin-1 incubation we isolated monocytes, B cells and CD4 and CD8 positive lymphocytes. The purity of the monocytes was 80–90%, B cells were 95% pure with some residual monocytes. CD4 and CD8 positive cells were more than 98% pure. B cells could not be induced to produce IL-10 by dGal. After incubation with different concentrations of dGal (0, 0.2, 1 and 2 M) CD4 positive lymphocytes of four donors produced low concentrations of IL-10 (Fig. 4A). The amounts of IL-10 produced were concentration dependent. The yield of CD8 positive cells was low so CD8 positive cells were only treated with 0 and 2 M dGal (Fig. 4B). The measured concentrations of IL10 were low but in the same range as for the CD4 positive cells. Monocytes are able to produce higher concentrations of IL-10. Comparison of the IL-10 amounts to the concentrations produced by PBMC with the levels in monocytes (Fig. 4C) and T lymphocytes together of the same donors indicate that this is not sufficient to account for the total amount. At 2 M dGal the mean of the production of the four donors is 3610 ± 992 pg/ml in the PBMC and 1102 ± 269 pg/ml for the monocytes. 3.4. Effects of stable galectin-1 homodimers on other cytokines To test the potential of stable galectin-1 homodimers to modulate the production of other cytokines, we performed a multiplex ELISA on galectin-1 treated cells (24 h) of five independent donors for cytokines which have been reported to be modulated upon galectin-1 treatment in animal models. ␣CD3 treatment of PBMCs of two donors revealed only low IL2 levels and no effects were observed for wtGal or dGal. For the three other donors no changes in IL-2 production were observed with 0.2, 2 or 20 M wtGal. In two of these donors reduced IL-2 levels were present with all dGal concentrations (Fig. 5) and in one of these donors reduction of IL-2 was only observed for 0.2 and 1 M dGal. For IFN-␥, treatment with 20 M wtGal revealed decreased protein levels in 1/5 donors, this downregulation of IFN-␥ was not observed using dGal (results not shown). The levels of IL-4, IL5, IL-12p70, and IL-13 proteins were very low and no effects were observed using various concentrations of wtGal or dGal protein (data not shown). For IL-1 a high level of protein was observed upon dGal treatment, which was most pronounced at the highest concentration of dGal in all five donors. Treatment with ␣CD3 in combination with 2 M dGal resulted in IL-1 levels ranging from 3630 to 6300 pg/ml, whereas IL-1 production in cells treated only with ␣CD3 were <250 pg/ml. For ␣CD3 in combination with 20 M wtGal a weak increase in the level of IL-1 protein was detected with productions ranging
Fig. 4. Effect of stable galectin-1 homodimers on IL-10 production of different cell populations. Cells were incubated with different concentrations of stable galectin-1 (dGal 0, 0.2 and 2 M) for 24 h. B cells (purity 95%) did not produce IL-10. Induction of IL-10 in the CD4 positive cells (over 98% pure) of four donors is shown (A). In (B) the induction of IL-10 by the CD8 cells (over 98% pure) of four donors (only 0 and 2 M dGal were used) is shown. Comparison of PBMC (containing 35–64% monocytes) and monocyte (mono) (purity 80–90%) production of IL-10 is shown in (C) for all four donors individually. Monocytes and T cells both produce IL-10 but the combined levels do not reach the large amounts of IL-10 observed in PBMC.
Fig. 5. Effect of wild-type galectin-1 and stable galectin-1 homodimers on IL-2 protein level. Lymphocytes of five independent donors were incubated for 24 h with ␣CD3 (10 ng/ml) in combination with different concentrations of wild-type galectin-1 protein (wtGal 0.2, 2 and 20 M) or stable galectin-1 homodimers (dGal 0.2, 1 and 2 M). Statistically there is no significant difference.
J. van der Leij et al. / Molecular Immunology 44 (2007) 506–513
511
Fig. 7. Comparison of IL-1 production and cell death. PBMC of three donors were treated with wild-type galectin-1 (wtGal 0.2, 2 and 20 M) and stable galectin-1 homodimers (dGal 0.2, 1 and 2 M) in combination with ␣CD3 for 24 hours. a representative donor is shown for both IL-1 production (right Y axis) and percentage of death cells (left Y axis). The IL-1 production is not caused by the release of IL-1 due to cell death.
centrations of wtGal or dGal, while there is a strong increase in the IL-1 concentration with dGal. Therefore the induction of IL-1 is not likely to be due to necrosis. 4. Discussion
Fig. 6. Effect of wild-type galectin-1 and stable galectin-1 homodimers on IL1 production. IL-1 protein (A) and IL-1 mRNA (B and C) production by lymphocytes of five independent donors after 24 h (A and C) or 6 h (B) of incubation with ␣CD3 (10 ng/ml) in combination with different concentrations of wild-type galectin-1 protein (wtGal 0.2, 2 and 20 M) or stable galectin-1 homodimers (dGal 0.2, 1 and 2 M). In all five donors a significant increase in IL-1 production is seen with 20 M wtGal (p = 0.027) and all three concentrations (p = 0.027 for all three concentrations) dGal. The relative amount of IL-1 mRNA was measured by determining the 2−∆Ct . The range is given as 2−(∆Ct+S.D.∆Ct) and 2−(∆Ct−S.D.∆Ct) .
from 340 to 1880 pg/ml (Fig. 6A). To further analyze this unexpected effect of both wtGal and dGal treatment of PBMCs, we also studied the effect at the mRNA level. This revealed similar relative IL-1 transcript levels in all samples, including the untreated and wtGal or dGal treated samples (Fig. 6B and C). Comparison of the 6 and 24 h treated samples revealed reduction of IL-1 mRNA levels in the samples at 24 h in comparison to the 6 h samples. Release of IL-1 can be due to leakage of the cell membrane caused by cell death (necrosis), so a comparison was made between LDH release as a marker of necrosis and IL-1 production. In Fig. 7 is the percentage of dead cells and the IL-1 production given of one representative donor out of three. There is no increase in the amount of dead cells with increasing con-
Research over recent years has identified a beneficial role for galectin-1 treatment in inflammatory (Santucci et al., 2000) and autoimmune diseases (Levi et al., 1983; Offner et al., 1990; Rabinovich et al., 1999c; Santucci et al., 2003). The antiinflammatory results achieved in these animal models and the in vitro experiments with human cells were characterized by an increase of apoptosis of activated T cells and modulation of cytokine production (Perillo et al., 1995; Rabinovich et al., 1999c, 2002; Santucci et al., 2003). Based on enhanced activity of dGal with respect to apoptosis induction and IL-10 production it may be anticipated that treatment with dGal that will not dissociate upon dilution in vivo might provide an improved tool to enhance the anti-inflammatory properties of galectin-1. IL-10 is an important immunoregulatory cytokine that has not been studied in the animal models treated with galectin-1, but may have contributed to the galectin-1-induced beneficial effects observed. Since induction of apoptosis is mainly reported in thymocytes and CD7 positive T cell lines we used the MOLT-4 T cell line. We demonstrated that roughly 20-fold less dGal as compared to wtGal is required to obtain efficient apoptosis induction. Our results are an improvement on a recently published study, in which a three-fold difference was observed between covalent homodimers (obtained by linking two galectin-1 moieties by a two-glycine linker) and wild-type galectin-1 (Battig et al., 2004). We have demonstrated that dGal is more effective even at a100-fold reduced concentration with respect to IL-10 induction than wtGal, supporting the putative strength of dGal as a tool for treatment of immune system associated diseases. In comparison to the 100-fold reduced levels of dGal required for efficient
512
J. van der Leij et al. / Molecular Immunology 44 (2007) 506–513
IL-10 induction the apoptotic activities of dGal appear to be less pronounced. These results suggest that for apoptosis induction a critical threshold concentration of galectin-1 homodimers is required. This might be explained by the apoptotic pathway induced by galectin-1, which appears to be dependent of receptor cross-linking (Pace et al., 1999; Perillo et al., 1995). In an earlier study (van der Leij et al., 2004) we only tested T cell subpopulations for their ability to induce IL-10 production upon galectin-1 stimulation, since the best defined galectin-1 receptors are T cell markers. The IL-10 production we saw by CD4 and CD8 positive cells was lower than the total production seen by PBMC. In this study the production of IL-10 could be contributed to monocytes and T cells but not to B cells when we tested in purified cell populations. Similar to our previous findings, the IL-10 amounts produced by these PBMC subpopulations are considerably lower than the levels in the total mixture of cells of the same donors. These data suggest that cell–cell contact is a crucial factor that contributes to induction of IL-10 production. In this study, we did not observe consistent changes in IFN-␥ production following dGal or wtGal treatment. In one donor a reduced IFN-␥ protein level was observed after treatment with ␣CD3 in combination with 20 M wtGal, however, this effect was not observed using dGal treatment. In our previous study (van der Leij et al., 2004) we observed consistent downregulation of IFN-␥ production in sorted CD4+ and CD8+ T cells of five donors upon treatment with high concentrations of wtGal. This might imply that in total PBMC, IFN-␥ might rather be produced by other cell types, like natural killer cells, instead of CD4+ or CD8+ T cells. In a mouse model for rheumatoid arthritis, galectin-1 gene therapy resulted in downregulation of IFN-␥ in draining lymph nodes (>500 pg in control cells versus <50 pg in treated mice) (Rabinovich et al., 1999c). In an experimental model for colitis in mice, production of IFN-␥ is downregulated significantly in plasma and mucosa of mice treated with recombinant human galectin-1 (Santucci et al., 2003). Downregulation of IL-2 production was observed in three donors upon dGal treatment, whereas no changes were observed for wtGal. In two donors the overall production was very low and no changes were observed for wtGal or dGal. In a study by Vespa et al. (1999), treatment of murine T cell hybridoma BI-141 cells with ␣CD3 in combination with 20 M galectin-1 protein during 12 h resulted in a 10-fold reduction in IL-2 production. The difference in cell type, murine T cell hybridoma versus human blood mononuclear cells, and the timing, 12 h versus 24 h, may account for variations in IL-2 reduction observed between our study and the results presented by Vespa et al. (1999). For IL-4, IL-5, IL-12p70 and IL-13 cytokine levels present in the culture media were at the lower level of the detection limit and precluded a reliable interpretation of the data and the detection of a putative downregulating effect. No induction was observed for these cytokines for wtGal or dGal at any of the tested concentrations. In the literature no data are available about the effect of galectin-1 on the production of IL-4 and IL13. For IL-5, upregulation was detected in a mouse model for rheumatoid arthritis upon analyses of cell suspensions of draining lymph nodes cells stimulated for 72 h with collagen type II
(Rabinovich et al., 1999c). However, no significant differences in IL-5 production were observed in spleen cells between control and experimental groups. In an experimental mouse model for colitis, early administration of 0.4 or 1.0 mg/kg recombinant wild-type human galectin-1 resulted in a strong downregulation of IL-12 levels in plasma and mucosa of treated animals (Santucci et al., 2003). Surprisingly, a high level of IL-1 protein was observed in dGal treated but not wtGal treated cells. However, at the mRNA level similar levels of IL-1 transcripts were observed in dGaltreated, wtGal-treated and untreated samples. Thus, despite high protein levels observed in dGal treated cells it is unlikely that this effect can be attributed to dGal induced expression of IL1 at the mRNA level. We excluded that the release of IL-1 was due to cell death by simultaneous measurement of LDH release. It might be speculated that high IL-1 protein levels are obtained by secretion of IL-1 by monocytes and macrophages. In these cells inflammatory signals (such as LPS) promote synthesis and accumulation of the inactive pro-IL-1 form in the cytoplasm. In a second step, ATP triggers caspase-1 (IL-1 converting enzyme; ICE) mediated cleavage of pro-IL-1 and the subsequent release of the mature protein (Andrei et al., 2004; Wewers, 2004). A first explanation for the high IL-1 protein levels in dGal treated cells might be that dGal mimics the function of LPS and exogenous ATP in IL-1 production, whereas wtGal does not display this effect. Another possible explanation for the high IL-1 protein levels in dGal treated cells might be that lower amounts of IL-1 receptor are present in dGal treated cells as compared to wtGal or untreated cells. In a mouse model for colitis, early administration of 1 mg/kg recombinant wild-type human galectin-1 resulted in downregulation of IL-1 protein levels in plasma and in colon mucosa (Santucci et al., 2003). In our in vitro model we measure no change in IL-1 protein levels with wtGal and enhanced levels with dGal. The timing and complexity of different processes that play a role in vivo make the comparison with in vitro experiments difficult. In conclusion, we demonstrated that stable galectin-1 homodimers display enhanced activity with respect to IL-10 induction and structural changes related to apoptosis as compared to wild-type galectin-1 protein. In contrast to the variable changes observed for IL-2 and IFN-␥, the effects on IL-10 were consistent in all five donors. Although genetic heterogeneity in human might result in variation between the effectiveness of dGal induced effects on cytokine productions in individual persons, our results clearly demonstrate that treatment with dGal is an improved tool for the treatment of chronic inflammatory or autoimmune diseases. Acknowledgement This work was supported by a grant from the Graduate School, Groningen University Institute for Drug Exploration (GUIDE). References Allione, A., Wells, V., Forni, G., Mallucci, L., Novelli, F., 1998. Betagalactoside-binding protein (beta GBP) alters the cell cycle, up-regulates
J. van der Leij et al. / Molecular Immunology 44 (2007) 506–513 expression of the alpha- and beta-chains of the IFN-gamma receptor, and triggers IFN-gamma-mediated apoptosis of activated human T lymphocytes. J. Immunol. 161, 2114–2119. Andrei, C., Margiocco, P., Poggi, A., Lotti, L.V., Torrisi, M.R., Rubartelli, A., 2004. Phospholipases C and A2 control lysosome-mediated IL-1 beta secretion: Implications for inflammatory processes. Proc. Natl. Acad. Sci. U.S.A. 101, 9745–9750. Barondes, S.H., Castronovo, V., Cooper, D.N., Cummings, R.D., Drickamer, K., Feizi, T., Gitt, M.A., Hirabayashi, J., Hughes, C., Kasai, K., 1994. Galectins: a family of animal beta-galactoside-binding lectins. Cell 76, 597– 598. Battig, P., Saudan, P., Gunde, T., Bachmann, M.F., 2004. Enhanced apoptotic activity of a structurally optimized form of galectin-1. Mol. Immunol. 41, 9–18. Cho, M., Cummings, R.D., 1995. Galectin-1, a beta-galactoside-binding lectin in Chinese hamster ovary cells. I. Physical and chemical characterization. J. Biol. Chem. 270, 5198–5206. de Kruif, J., Logtenberg, T., 1996. Leucine zipper dimerized bivalent and bispecific scFv antibodies from a semi-synthetic antibody phage display library. J. Biol. Chem. 271, 7630–7634. Dias-Baruffi, M., Zhu, H., Cho, M., Karmakar, S., McEver, R.P., Cummings, R.D., 2003. Dimeric galectin-1 induces surface exposure of phosphatidylserine and phagocytic recognition of leukocytes without inducing apoptosis. J. Biol. Chem. 278, 41282–41293. Elola, M.T., Chiesa, M.E., Alberti, A.F., Mordoh, J., Fink, N.E., 2005. Galectin-1 receptors in different cell types. J. Biomed. Sci. 12, 13–29. Fajka-Boja, R., Szemes, M., Ion, G., Legradi, A., Caron, M., Monostori, E., 2002. Receptor tyrosine phosphatase, CD45 binds galectin-1 but does not mediate its apoptotic signal in T cell lines. Immunol. Lett. 82, 149–154. Levi, G., Tarrab-Hazdai, R., Teichberg, V.I., 1983. Prevention and therapy with electrolectin of experimental autoimmune myasthenia gravis in rabbits. Eur. J. Immunol. 13, 500–507. Nguyen, J.T., Evans, D.P., Galvan, M., Pace, K.E., Leitenberg, D., Bui, T.N., Baum, L.G., 2001. CD45 modulates galectin-1-induced T cell death: regulation by expression of core 2 O-glycans. J. Immunol. 167, 5697–5707. Offner, H., Celnik, B., Bringman, T.S., Casentini-Borocz, D., Nedwin, G.E., Vandenbark, A.A., 1990. Recombinant human beta-galactoside binding lectin suppresses clinical and histological signs of experimental autoimmune encephalomyelitis. J. Neuroimmunol. 28, 177–184. Pace, K.E., Lee, C., Stewart, P.L., Baum, L.G., 1999. Restricted receptor segregation into membrane microdomains occurs on human T cells during apoptosis induced by galectin-1. J. Immunol. 163, 3801–3811. Pace, K.E., Hahn, H.P., Pang, M., Nguyen, J.T., Baum, L.G., 2000. CD7 delivers a pro-apoptotic signal during galectin-1-induced T cell death. J. Immunol. 165, 2331–2334.
513
Perillo, N.L., Uittenbogaart, C.H., Nguyen, J.T., Baum, L.G., 1997. Galectin-1, an endogenous lectin produced by thymic epithelial cells, induces apoptosis of human thymocytes. J. Exp. Med. 185, 1851– 1858. Perillo, N.L., Pace, K.E., Seilhamer, J.J., Baum, L.G., 1995. Apoptosis of T cells mediated by galectin-1. Nature 378, 736–739. Rabinovich, G.A., Ariel, A., Hershkoviz, R., Hirabayashi, J., Kasai, K.I., Lider, O., 1999a. Specific inhibition of T-cell adhesion to extracellular matrix and proinflammatory cytokine secretion by human recombinant galectin-1. Immunology 97, 100–106. Rabinovich, G.A., 1999b. Galectins: an evolutionarily conserved family of animal lectins with multifunctional properties; a trip from the gene to clinical therapy. Cell Death Differ. 6, 711–721. Rabinovich, G.A., Daly, G., Dreja, H., Tailor, H., Riera, C.M., Hirabayashi, J., Chernajovsky, Y., 1999c. Recombinant galectin-1 and its genetic delivery suppress collagen-induced arthritis via T cell apoptosis. J. Exp. Med. 190, 385–398. Rabinovich, G.A., Ramhorst, R.E., Rubinstein, N., Corigliano, A., Daroqui, M.C., Kier-Joffe, E.B., Fainboim, L., 2002. Induction of allogenic T-cell hyporesponsiveness by galectin-1-mediated apoptotic and non-apoptotic mechanisms. Cell Death Differ. 9, 661–670. Santucci, L., Fiorucci, S., Cammilleri, F., Servillo, G., Federici, B., Morelli, A., 2000. Galectin-1 exerts immunomodulatory and protective effects on concanavalin A-induced hepatitis in mice. Hepatology 31, 399– 406. Santucci, L., Fiorucci, S., Rubinstein, N., Mencarelli, A., Palazzetti, B., Federici, B., Rabinovich, G.A., Morelli, A., 2003. Galectin-1 suppresses experimental colitis in mice. Gastroenterology 124, 1381–1394. van der Leij, J., van den Berg, A., Blokzijl, T., Harms, G., van Goor, H., Zwiers, P., van Weeghel, R., Poppema, S., Visser, L., 2004. Dimeric galectin-1 induces IL-10 production in T-lymphocytes: an important tool in the regulation of the immune response. J. Pathol. 204, 511– 518. Vespa, G.N., Lewis, L.A., Kozak, K.R., Moran, M., Nguyen, J.T., Baum, L.G., Miceli, M.C., 1999. Galectin-1 specifically modulates TCR signals to enhance TCR apoptosis but inhibit IL-2 production and proliferation. J. Immunol. 162, 799–806. Walzel, H., Schulz, U., Neels, P., Brock, J., 1999. Galectin-1, a natural ligand for the receptor-type protein tyrosine phosphatase CD45. Immunol. Lett. 67, 193–202. Walzel, H., Blach, M., Hirabayashi, J., Kasai, K.I., Brock, J., 2000. Involvement of CD2 and CD3 in galectin-1 induced signaling in human Jurkat T-cells. Glycobiology 10, 131–140. Wewers, M.D., 2004. IL-1beta: an endosomal exit. Proc. Natl. Acad. Sci. U.S.A. 101, 10241–10242.
Strongly enhanced IL-10 production using stable galectin-1 homodimers Judith van der Leij a , Anke van den Berg a , Geert Harms a , Hugo Eschbach a , Hans Vos a , Peter Zwiers b , Rob van Weeghel b , Herman Groen b , Sibrand Poppema a , Lydia Visser a,∗ a
Department of Pathology and Laboratory Medicine, University Medical Center Groningen and University of Groningen, Groningen, The Netherlands b IQ Corporation B.V., Groningen, The Netherlands Received 2 August 2005; received in revised form 14 February 2006; accepted 15 February 2006 Available online 3 April 2006
Abstract Galectin-1 is the homodimeric form of a protein, which is present in a dynamic equilibrium with the -galactoside monomeric form and has potent anti-inflammatory and immunomodulating effects. These favorable effects are probably related to the induction of apoptosis in activated T cells and the induction of IL-10, which have been demonstrated to be characteristic for the dimeric form of the protein. Based on these findings it can be speculated that the in vivo effects of galectin-1 can be improved by the generation of stable galectin-1 homodimers (dGal). To test this hypothesis we produced leucine-zipper based stable galectin-1 homodimers and tested its apoptosis inducing effects on MOLT-4 cells and its immunomodulatory effects in vitro on PBMC of five independent donors. Phosphatidylserine exposure and a drop in mitochondrial membrane potential was strongly enhanced on MOLT-4 cells upon treatment with dGal as compared to wtGal. The minimal effective concentration was 20-fold reduced as compared to the minimal effective wtGal concentration. dGal showed enhanced induction of IL-10 on total PBMC as compared to treatment with wild-type protein (wtGal). The minimal effective dGal concentration was 100-fold lower than that of wtGal. Of the purified cell populations monocytes are the strongest IL-10 producers, whereas T cells induce IL-10 at a lower level and no induction is observed in B cells. Besides induction of IL-10, dGal caused an increase in IL-1 production in all donors and a reduction of IL-2 production in 3 out of 5 donors, whereas no consistent changes were observed for other inflammatory cytokines. In summary, we demonstrated that dGal shows enhanced effects at strongly reduced concentrations. Application of dGal may therefore serve as an improved treatment of chronic inflammatory diseases. © 2006 Elsevier Ltd. All rights reserved. Keywords: Stable galectin-1 homodimers; IL10; Apoptosis
1. Introduction Galectin-1 is a homodimeric protein, consisting of two identical subunits from 14.5 kDa (Barondes et al., 1994; Cho and Cummings, 1995). The protein exists in a reversible monomerdimer equilibrium, with a dissociation constant (Kd ) of ∼7 M (Cho and Cummings, 1995). Galectin-1 binds to lactosamine residues present on N- and O-linked glycans, which are present on glycosylated T cell surface receptors like CD45, CD43, CD7, CD2, and CD3 (Walzel et al., 1999, 2000; Pace et al., 1999, 2000; Elola et al., 2005). The protein has a wide tissue distribution, and plays a role in many biological processes, including cell adhe∗
Corresponding author at: Department of Pathology and Laboratory Medicine, University Medical Center Groningen, Hanzeplein 1, P.O. Box 30.001, 9700 RB Groningen, The Netherlands. Tel.: +31 50 361 0402; fax: +31 50 363 2510. E-mail address: [email protected] (L. Visser). 0161-5890/$ – see front matter © 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.molimm.2006.02.011
sion (Rabinovich et al., 1999a; Rabinovich, 1999b), proliferation (Allione et al., 1998), and the induction of apoptosis in thymocytes and activated T cells (Perillo et al., 1995, 1997). Apoptosis induction in vitro is only observed using high concentrations of the galectin-1 protein (concentrations above 7 M) (Perillo et al., 1995), which suggests that the dimeric form of the protein is required to induce apoptosis. More evidence was provided in a study with a monomeric mutant form, which binds to cells but even at very high concentrations is unable to induce apoptosis of HL-60 and MOLT-4 cells (Dias-Baruffi et al., 2003). Using a covalently linked dimeric form of galectin-1, apoptosis induction in EL-4 thymoma cells and thymocytes from C57Bl/6 mice was increased three to 10-fold (Battig et al., 2004). Although the glycosylation status of CD45 is able to affect galectin-1 induced apoptosis (Nguyen et al., 2001), the presence of CD45 is not a prerequisite (Fajka-Boja et al., 2002). In contrast, CD7 appears to be necessary for induction of apoptosis mediated by galectin-1 (Pace et al., 2000).
J. van der Leij et al. / Molecular Immunology 44 (2007) 506–513
Recently, we demonstrated that dimeric, but not monomeric concentrations of the galectin-1 protein induce IL-10 production in CD4+ and CD8+ T cell subsets (van der Leij et al., 2004). It can be speculated that the IL-10 induction is responsible for the beneficial role of galectin-1 observed in several animal models of inflammatory (Santucci et al., 2000) or autoimmune diseases (Levi et al., 1983; Offner et al., 1990; Rabinovich et al., 1999c; Santucci et al., 2003). In these models, an increased level of leukocyte apoptosis and a downregulation of pro-inflammatory cytokines TNF-␣, IL-1, IL-12, IFN-␥, and upregulation of IL-5 is detected. The in vivo efficacy of galectin-1 as observed in previous studies may still be limited since processes as apoptosis and IL-10 induction require the dimeric form of the galectin-1 protein. We speculated that the use of stable galectin-1 homodimers might enhance the beneficial effects observed with high concentrations of wild-type galectin-1. These stable galectin-1 dimers might thereby provide a very attractive means for the treatment of inflammatory and autoimmune diseases. In this study we report the construction of stabilized galectin-1 homodimers (dGal) and the analysis of the effects on apoptosis, IL-10 and other cytokines upon treatment of peripheral blood mononuclear cells in comparison to the effects observed with high concentrations of normal/wild-type galectin-1 protein (wtGal). 2. Materials and methods 2.1. Medium The culture medium used throughout the experiments was RPMI 1640 supplemented with 2 mM L-glutamine, 10% heatinactivated fetal bovine serum (FBS), 100 U/ml penicillin, and 100 g/ml streptomycin (Cambrex, Verviers, Belgium). In culture experiments with the galectin-1 protein, the culture medium was additionally supplemented with 1.2 mM dithiothreitol (DTT). 2.2. Preparation of recombinant human galectin-1 protein Wild-type recombinant human galectin-1 protein (wtGal) was prepared as described previously (van der Leij et al., 2004). For the construction of stable galectin-1 homodimers (dGal), we used a FOS leucine zipper based construct, which was kindly provided by de Kruif and Logtenberg (1996). Between the FOS leucine zipper and galectin-1, a hinge region was placed functioning as a flexible linker. The FOS leucine zipper was flanked by CGG and GGC amino acids at the N- and C-terminus, respec-
507
tively, to covalently link the zippers by disulfide bonds between cysteine residues (Fig. 1). 2.3. Measurement of phosphatidylserine exposure and mitochondrial membrane potential Phosphatidylserine exposure on the cell surface was measured using a phosphatidylserine detection kit (IQP, Groningen, The Netherlands). Mitochondrial membrane potentials were assessed using 3,3 -dihexiloxa dicarbocyanine iodide (DiOC6) (Molecular Probes, Leiden, The Netherlands). MOLT-4 T cells were cultured at 37 ◦ C for 3 h (1 × 106 ml−1 in RPMI/10%FCS/1.2 mM DTT) with different concentrations of galectin-1 protein. To dissociate galectin-1 from the cell membrane cells were treated with 0.1 M lactose/PBS for 10 min at room temperature. For phosphatidyl exposure the cells were resuspended in 110 l calcium buffer (10 mM HEPES, 140 mM NaCl, 2.5 mM CaCl2 , pH 7.4) containing 2.5 l AnnexinV-Fitc. Cells were incubated for 20 min on ice, washed once with calcium buffer, suspended in 160 l calcium buffer containing 1 l propidium iodide (PI), and incubated for 10 min on ice. For measurement of the mitochondrial membrane potential cells were mixed with DiOC6 (final concentration 0.1 M) and incubated at 37 ◦ C for 30 min. After washing PI was added at a final concentration of 0.1 g/ml. Cells were immediately analyzed by flow cytometry, 10,000 events per sample, on a Coulter Epics-Elite flow cytometer (Coulter Corporation, Hialeah, FL, USA). Data were analyzed using WinList 4.0 software (Verity Software House Inc., Topsham, ME, USA). 2.4. Galectin-1 treatment and cytokine measurements Peripheral blood mononuclear cells (PBMC) were obtained from healthy volunteers and cells were isolated by Ficoll–Hypaque centrifugation. Isolated PBMC (1 × 106 ml−1 ) were stimulated for 6 or 24 h with different concentrations of galectin-1 protein, with or without ␣CD3 (10 ng/ml) activation. For inhibition assays, galectin-1 protein was pre-incubated for 30 min at room temperature with 0.1 M lactose before addition to cells. Cytokine production was determined in cell-free culture supernatants after 24 h stimulation, using multiplex ELISA. An 8-plex premixed antibody bead kit was purchased from BioSource International Inc. (BioSource, Etten-Leur, The Netherlands). Cytokines analyzed were IL-1, IL-2, IL-4, IL-5, IL-10, IL-12p70, IL-13, and IFN-␥. A standard
Fig. 1. Overview of FOS-hinge-galectin-1 region. Leucine residues (L) in the FOS zipper are indicated in bold. The FOS leucine zipper is flanked by CGG and GGC amino acids (indicated in italic) at the N- and C-terminus, respectively, to covalently link leucine zippers. Behind the FOS zipper, a hinge region was introduced (PKPSTPPGSSH, bold), followed by the galectin-1 protein (underlined).
508
J. van der Leij et al. / Molecular Immunology 44 (2007) 506–513
curve was generated and analyzed in duplicate according to manufacturer’s instructions. ELISA was performed according to manufacturer’s instructions, and samples were analyzed using the Luminex 100TM Instrument (Luminex Corporation, Austin, TX, USA). The concentrations of cytokines in the samples were determined using the standard curves, generated using a five-parameter algorithm in curve fitting software. In indicated experiments, cytokine productions for IL-10 and IL-1 were measured using individual ELISA kits according to manufacturer’s instructions (R&D Systems, Abingdon, UK). We used a nonparametric Mann–Whitney test to define if the differences were statistically significant. 2.5. Real-time PCR analysis for IL-10, IL-1β, and HPRT Total RNA was isolated from cell pellets using the absolutely RNA RT-PCR miniprep kit (Stratagene, La Jolla, CA, USA). RNA (200 ng) and was reverse transcribed with Superscript II reverse transcriptase (Invitrogen, Paisley, UK) in a volume of 20 l using random hexamers (300 ng). Primers (Invitrogen) and probes (Eurogentec, Seraing, Belgium) used for real-time PCR analysis were developed using primer design software (Applied Biosystems, Foster City, CA, USA). Primers used were: IL-10F 5 -atgaaggatcagctggacaactt-3 , IL-10R 5 -ccttgatgtctgggtcttggt-3 ; HPRTF 5 -ggcagtataatccaaagatggtcaa-3 , HPRTR 5 -gtctggcttatatccaacacttcgt-3 . Probe sequences labeled 5 with the FAM reporter dye and 3 with the TAMRA quencher dye molecules were: IL-10 5 acctgggttgccaagccttgtctg-3 , HPRT 5 -caagcttgctggtgaaaaggacccc-3 . An Assay-on-Demand Gene Expression ProductTM (Applied Biosystems) was used for IL-1β (Hs 00174097 m1). Reactions were performed in 384-well plates (Applied Biosystems) in a volume of 20 l containing real-time PCR mastermix (Eurogentec), 900 nM of each primer, 200 nM of an individual probe and 2 ng cDNA. PCR amplifications were performed using the ABI prism 7900HT sequence detection system (Applied Biosystems). Standard cycling conditions were used including a pre-amplification step of 50 ◦ C for 2 min, 95 ◦ C for 10 min, followed by amplification for 40 cycles of 95 ◦ C for 15 s and 60 ◦ C for 1 min. All samples were analyzed in triplicate. Mean cycle treshold values (Ct) and standard deviations (S.D.) were calculated for cytokine and housekeeping genes. The amount of cytokine target was normalized relative to the amount of housekeeping gene (Ct = Ctgene − CtHPRT ) and the S.D. √ of the Ct (S.D.(Ct)) was calculated S.D.(Ct) = ((S.D.gene )2 + (S.D.HPRT )2 ). The relative amount of cytokine was measured by determining the 2−Ct . The range is given as 2−(Ct+S.D.Ct) and 2−(Ct−S.D.Ct) . 2.6. Isolation of individual cell populations PBMC were obtained from four healthy donors and cells were isolated by Ficoll–Hypaque centrifugation. Monocytes and B cells were isolated each with negative isolation kits for monocytes and B cells (Dynal, Oslo, Norway). CD4 and CD8 positive lymphocytes were isolated by staining with fluorochromelabeled antibodies against CD4 and CD8 (IQProducts, Gronin-
gen, The Netherlands), and cells were sorted on the MoFlow cytometer (Cytomation, Fort Collins, CO, USA). Purity of isolated cell populations was checked routinely. PBMC and isolated cell populations were incubated with different concentrations of dGal (0, 0.2, 1 and 2 M) for 24 h at 1 × 106 cells/ml. Cell culture supernatants were harvested and measured for IL-10 production by ELISA. 2.7. Measurement of Lactate dehydogenase (LDH) release as an assessment of cell death PBMC were obtained from healthy volunteers and cells were isolated by Ficoll–Hypaque centrifugation. Isolated PBMC (1 × 106 ml−1 ) were stimulated for 24 h with different concentrations of galectin-1 protein, with ␣CD3 (10 ng/ml) activation in RPMI with 2% FBS. LDH release was measured with the CytoTox-ONE homogeneous membrane integrity assay (Promega, Madison, WI, USA). The percentage of dead cells was calculated by using the maximum LDH release (lysis of total cells). 3. Results 3.1. Enhanced phosphatidylserine exposure and drop in mitochondrial membrane potential using stable galectin-1 homodimers To test the efficiency of dGal to induce apoptosis MOLT4 T cells were treated with dGal concentrations ranging from 0.1 to 5 M and the phosphatidylserine exposure on the cell membrane as determined by annexin V staining as well as the mitochondrial membrane potential as measured by DiOC6 was compared to wtGal concentrations varying from 0.1 to 20 M. Fig. 2A represents the mean of three independent experiments and shows that wtGal induced annexin V binding to 59 ± 3% of the cells at 20 M, whereas the percentage of annexin V positive cells at 5 M was reduced to 42 ± 15%. Using 5 M dGal efficient binding of annexin V to the cell membrane was observed in 77 ± 7% of cells, and with 1 M dGal still 68 ± 9% of the treated cells stained positive for annexin V. In Fig. 2B the results for the DiOC6 staining show that the percentages of cells with a drop in mitochondrial membrane potential are lower then the percentages of AnnexinV positive cells in the same experiments indicating that the exposure of phospatidylserines is probably an earlier or more efficient event. Using 20 M wtGal a drop in membrane potential was shown in 21 ± 5% of cells while with a concentration of 1 M dGal 33 ± 2% of cells show a drop in the mitochondrial membrane potential. Both methods revealed a similar increase of efficiency of the dGal at 20-fold reduced concentrations than wtGal. These data demonstrate a markedly enhanced activity of dGal with respect to the induction of apoptosis. 3.2. Enhanced production of IL-10 using stable galectin-1 homodimers To test the efficiency of IL-10 induction of dGAL in comparison to wtGal, we incubated PBMC with various concentrations of either protein. Stimulation of PBMC with ␣CD3 in com-
J. van der Leij et al. / Molecular Immunology 44 (2007) 506–513
509
Fig. 2. Effect of wild-type galectin-1 and stable galectin-1 homodimers on phosphatidylserine exposure and mitochondrial membrane potential in MOLT-4 T-cells. MOLT-4 T-cells were incubated for 3 h with wild-type galectin-1 (Gal 0.1–20 M) or stable galectin homodimers (dGal 0.1–5 M) and analyzed by flow cytometry. Results are shown as the percentage of AnnexinV positive cells (A). Results are shown as the percentage of cells with a drop in mitochondrial membrane potential (−m) (B). Bars represent the mean and standard deviation of three independent experiments. The dimeric galectin-1 (dGal) is effective at 20-fold lower concentrations as compared to wild-type galectin-1 (wtGal).
bination with low concentrations of wtGal (0.2 or 2 M) did not induce IL-10 protein production, whereas in combination with 20 M wtGal an significant increased IL-10 protein production (p = 0.048) was observed with levels ranging from 279 to 500 pg/ml in four donors and 1367 pg/ml in one donor. Cells treated only with ␣CD3 showed low levels of IL-10 ranging from 58 to 306 pg/ml (Fig. 3A). Cells treated with ␣CD3 in combination with the lowest (0.2 M) dGal concentration already demonstrated significant induction of high IL-10 levels in all five donors (p = 0.027) ranging from 642 to 1171 pg/ml (Fig. 3A). Thus, dGal showed enhanced efficiency for the induction of IL10 production in activated lymphocytes at concentrations of at least 100-fold lower as compared to wtGal. Treatment of resting PBMC with varying concentrations of dGal also revealed enhanced induction of IL-10 at concentrations up to 100-fold lower than the wtGal (not shown). IL-10 mRNA levels were analyzed in the same samples after 6 and 24 h of treatment. After 6 h (Fig. 3B), low levels of IL-10 mRNA were observed in control and wtGal treated cells (relative mRNA quantity range 2–7), whereas IL-10 mRNA levels were enhanced in dGal treated cells at all tested concentrations (IL10 mRNA level range 3–25). After 24 h, effects of galectin-1 on IL-10 mRNA levels were similar to the results at 6 h, although overall the relative IL-10 transcript levels were lower (Fig. 3C). In all five donors, IL-10 mRNA levels were higher at all three tested dGal concentrations as compared to the highest wtGal concentration (Fig. 3B and C).
Fig. 3. Effect of wild-type galectin-1 and stable galectin-1 homodimers on IL10 production. IL-10 protein (A) and IL-10 mRNA (B and C) production by lymphocytes of five independent donors after 24 h (A and C) or 6 h (B) of incubation with ␣CD3 (10 ng/ml) in combination with different concentrations of wild-type galectin-1 protein (wtGal 0.2, 2 and 20 M) or stable galectin-1 homodimers (dGal 0.2, 1 and 2 M). In all five donors a significant increase in IL-10 is seen with 20 M wtGal and all three concentrations of dGal (p = 0.048, 0.027, 0.027 and 0.027, respectively). Lactose inhibition (D) efficiently blocks IL-10 production induced by 1 M dGal. The relative amount of IL-10 mRNA was measured by determining the 2−∆Ct . The range is given as 2−(∆Ct+S.D.∆Ct) and 2−(∆Ct−S.D.∆Ct) .
Pre-incubation with lactose inhibits binding of galectin-1 to the various known cell surface receptors and this also inhibits the wtGal induced production of IL-10 (van der Leij et al., 2004). To test whether lactose pre-treatment displays the same effects on dGal, resting and activated PBMC of different donors were
510
J. van der Leij et al. / Molecular Immunology 44 (2007) 506–513
treated with dGal in different concentrations with or without addition of 0.1 M lactose for 24 h. Fig. 3D demonstrates in a representative experiment that pre-incubation with lactose efficiently blocked IL-10 induction by 1 M dGal both in resting and activated PBMC, with 74% and 70%, respectively. 3.3. Cell populations producing IL-10 after incubation with galectin-1 To establish which cell populations produce IL-10 due to galectin-1 incubation we isolated monocytes, B cells and CD4 and CD8 positive lymphocytes. The purity of the monocytes was 80–90%, B cells were 95% pure with some residual monocytes. CD4 and CD8 positive cells were more than 98% pure. B cells could not be induced to produce IL-10 by dGal. After incubation with different concentrations of dGal (0, 0.2, 1 and 2 M) CD4 positive lymphocytes of four donors produced low concentrations of IL-10 (Fig. 4A). The amounts of IL-10 produced were concentration dependent. The yield of CD8 positive cells was low so CD8 positive cells were only treated with 0 and 2 M dGal (Fig. 4B). The measured concentrations of IL10 were low but in the same range as for the CD4 positive cells. Monocytes are able to produce higher concentrations of IL-10. Comparison of the IL-10 amounts to the concentrations produced by PBMC with the levels in monocytes (Fig. 4C) and T lymphocytes together of the same donors indicate that this is not sufficient to account for the total amount. At 2 M dGal the mean of the production of the four donors is 3610 ± 992 pg/ml in the PBMC and 1102 ± 269 pg/ml for the monocytes. 3.4. Effects of stable galectin-1 homodimers on other cytokines To test the potential of stable galectin-1 homodimers to modulate the production of other cytokines, we performed a multiplex ELISA on galectin-1 treated cells (24 h) of five independent donors for cytokines which have been reported to be modulated upon galectin-1 treatment in animal models. ␣CD3 treatment of PBMCs of two donors revealed only low IL2 levels and no effects were observed for wtGal or dGal. For the three other donors no changes in IL-2 production were observed with 0.2, 2 or 20 M wtGal. In two of these donors reduced IL-2 levels were present with all dGal concentrations (Fig. 5) and in one of these donors reduction of IL-2 was only observed for 0.2 and 1 M dGal. For IFN-␥, treatment with 20 M wtGal revealed decreased protein levels in 1/5 donors, this downregulation of IFN-␥ was not observed using dGal (results not shown). The levels of IL-4, IL5, IL-12p70, and IL-13 proteins were very low and no effects were observed using various concentrations of wtGal or dGal protein (data not shown). For IL-1 a high level of protein was observed upon dGal treatment, which was most pronounced at the highest concentration of dGal in all five donors. Treatment with ␣CD3 in combination with 2 M dGal resulted in IL-1 levels ranging from 3630 to 6300 pg/ml, whereas IL-1 production in cells treated only with ␣CD3 were <250 pg/ml. For ␣CD3 in combination with 20 M wtGal a weak increase in the level of IL-1 protein was detected with productions ranging
Fig. 4. Effect of stable galectin-1 homodimers on IL-10 production of different cell populations. Cells were incubated with different concentrations of stable galectin-1 (dGal 0, 0.2 and 2 M) for 24 h. B cells (purity 95%) did not produce IL-10. Induction of IL-10 in the CD4 positive cells (over 98% pure) of four donors is shown (A). In (B) the induction of IL-10 by the CD8 cells (over 98% pure) of four donors (only 0 and 2 M dGal were used) is shown. Comparison of PBMC (containing 35–64% monocytes) and monocyte (mono) (purity 80–90%) production of IL-10 is shown in (C) for all four donors individually. Monocytes and T cells both produce IL-10 but the combined levels do not reach the large amounts of IL-10 observed in PBMC.
Fig. 5. Effect of wild-type galectin-1 and stable galectin-1 homodimers on IL-2 protein level. Lymphocytes of five independent donors were incubated for 24 h with ␣CD3 (10 ng/ml) in combination with different concentrations of wild-type galectin-1 protein (wtGal 0.2, 2 and 20 M) or stable galectin-1 homodimers (dGal 0.2, 1 and 2 M). Statistically there is no significant difference.
J. van der Leij et al. / Molecular Immunology 44 (2007) 506–513
511
Fig. 7. Comparison of IL-1 production and cell death. PBMC of three donors were treated with wild-type galectin-1 (wtGal 0.2, 2 and 20 M) and stable galectin-1 homodimers (dGal 0.2, 1 and 2 M) in combination with ␣CD3 for 24 hours. a representative donor is shown for both IL-1 production (right Y axis) and percentage of death cells (left Y axis). The IL-1 production is not caused by the release of IL-1 due to cell death.
centrations of wtGal or dGal, while there is a strong increase in the IL-1 concentration with dGal. Therefore the induction of IL-1 is not likely to be due to necrosis. 4. Discussion
Fig. 6. Effect of wild-type galectin-1 and stable galectin-1 homodimers on IL1 production. IL-1 protein (A) and IL-1 mRNA (B and C) production by lymphocytes of five independent donors after 24 h (A and C) or 6 h (B) of incubation with ␣CD3 (10 ng/ml) in combination with different concentrations of wild-type galectin-1 protein (wtGal 0.2, 2 and 20 M) or stable galectin-1 homodimers (dGal 0.2, 1 and 2 M). In all five donors a significant increase in IL-1 production is seen with 20 M wtGal (p = 0.027) and all three concentrations (p = 0.027 for all three concentrations) dGal. The relative amount of IL-1 mRNA was measured by determining the 2−∆Ct . The range is given as 2−(∆Ct+S.D.∆Ct) and 2−(∆Ct−S.D.∆Ct) .
from 340 to 1880 pg/ml (Fig. 6A). To further analyze this unexpected effect of both wtGal and dGal treatment of PBMCs, we also studied the effect at the mRNA level. This revealed similar relative IL-1 transcript levels in all samples, including the untreated and wtGal or dGal treated samples (Fig. 6B and C). Comparison of the 6 and 24 h treated samples revealed reduction of IL-1 mRNA levels in the samples at 24 h in comparison to the 6 h samples. Release of IL-1 can be due to leakage of the cell membrane caused by cell death (necrosis), so a comparison was made between LDH release as a marker of necrosis and IL-1 production. In Fig. 7 is the percentage of dead cells and the IL-1 production given of one representative donor out of three. There is no increase in the amount of dead cells with increasing con-
Research over recent years has identified a beneficial role for galectin-1 treatment in inflammatory (Santucci et al., 2000) and autoimmune diseases (Levi et al., 1983; Offner et al., 1990; Rabinovich et al., 1999c; Santucci et al., 2003). The antiinflammatory results achieved in these animal models and the in vitro experiments with human cells were characterized by an increase of apoptosis of activated T cells and modulation of cytokine production (Perillo et al., 1995; Rabinovich et al., 1999c, 2002; Santucci et al., 2003). Based on enhanced activity of dGal with respect to apoptosis induction and IL-10 production it may be anticipated that treatment with dGal that will not dissociate upon dilution in vivo might provide an improved tool to enhance the anti-inflammatory properties of galectin-1. IL-10 is an important immunoregulatory cytokine that has not been studied in the animal models treated with galectin-1, but may have contributed to the galectin-1-induced beneficial effects observed. Since induction of apoptosis is mainly reported in thymocytes and CD7 positive T cell lines we used the MOLT-4 T cell line. We demonstrated that roughly 20-fold less dGal as compared to wtGal is required to obtain efficient apoptosis induction. Our results are an improvement on a recently published study, in which a three-fold difference was observed between covalent homodimers (obtained by linking two galectin-1 moieties by a two-glycine linker) and wild-type galectin-1 (Battig et al., 2004). We have demonstrated that dGal is more effective even at a100-fold reduced concentration with respect to IL-10 induction than wtGal, supporting the putative strength of dGal as a tool for treatment of immune system associated diseases. In comparison to the 100-fold reduced levels of dGal required for efficient
512
J. van der Leij et al. / Molecular Immunology 44 (2007) 506–513
IL-10 induction the apoptotic activities of dGal appear to be less pronounced. These results suggest that for apoptosis induction a critical threshold concentration of galectin-1 homodimers is required. This might be explained by the apoptotic pathway induced by galectin-1, which appears to be dependent of receptor cross-linking (Pace et al., 1999; Perillo et al., 1995). In an earlier study (van der Leij et al., 2004) we only tested T cell subpopulations for their ability to induce IL-10 production upon galectin-1 stimulation, since the best defined galectin-1 receptors are T cell markers. The IL-10 production we saw by CD4 and CD8 positive cells was lower than the total production seen by PBMC. In this study the production of IL-10 could be contributed to monocytes and T cells but not to B cells when we tested in purified cell populations. Similar to our previous findings, the IL-10 amounts produced by these PBMC subpopulations are considerably lower than the levels in the total mixture of cells of the same donors. These data suggest that cell–cell contact is a crucial factor that contributes to induction of IL-10 production. In this study, we did not observe consistent changes in IFN-␥ production following dGal or wtGal treatment. In one donor a reduced IFN-␥ protein level was observed after treatment with ␣CD3 in combination with 20 M wtGal, however, this effect was not observed using dGal treatment. In our previous study (van der Leij et al., 2004) we observed consistent downregulation of IFN-␥ production in sorted CD4+ and CD8+ T cells of five donors upon treatment with high concentrations of wtGal. This might imply that in total PBMC, IFN-␥ might rather be produced by other cell types, like natural killer cells, instead of CD4+ or CD8+ T cells. In a mouse model for rheumatoid arthritis, galectin-1 gene therapy resulted in downregulation of IFN-␥ in draining lymph nodes (>500 pg in control cells versus <50 pg in treated mice) (Rabinovich et al., 1999c). In an experimental model for colitis in mice, production of IFN-␥ is downregulated significantly in plasma and mucosa of mice treated with recombinant human galectin-1 (Santucci et al., 2003). Downregulation of IL-2 production was observed in three donors upon dGal treatment, whereas no changes were observed for wtGal. In two donors the overall production was very low and no changes were observed for wtGal or dGal. In a study by Vespa et al. (1999), treatment of murine T cell hybridoma BI-141 cells with ␣CD3 in combination with 20 M galectin-1 protein during 12 h resulted in a 10-fold reduction in IL-2 production. The difference in cell type, murine T cell hybridoma versus human blood mononuclear cells, and the timing, 12 h versus 24 h, may account for variations in IL-2 reduction observed between our study and the results presented by Vespa et al. (1999). For IL-4, IL-5, IL-12p70 and IL-13 cytokine levels present in the culture media were at the lower level of the detection limit and precluded a reliable interpretation of the data and the detection of a putative downregulating effect. No induction was observed for these cytokines for wtGal or dGal at any of the tested concentrations. In the literature no data are available about the effect of galectin-1 on the production of IL-4 and IL13. For IL-5, upregulation was detected in a mouse model for rheumatoid arthritis upon analyses of cell suspensions of draining lymph nodes cells stimulated for 72 h with collagen type II
(Rabinovich et al., 1999c). However, no significant differences in IL-5 production were observed in spleen cells between control and experimental groups. In an experimental mouse model for colitis, early administration of 0.4 or 1.0 mg/kg recombinant wild-type human galectin-1 resulted in a strong downregulation of IL-12 levels in plasma and mucosa of treated animals (Santucci et al., 2003). Surprisingly, a high level of IL-1 protein was observed in dGal treated but not wtGal treated cells. However, at the mRNA level similar levels of IL-1 transcripts were observed in dGaltreated, wtGal-treated and untreated samples. Thus, despite high protein levels observed in dGal treated cells it is unlikely that this effect can be attributed to dGal induced expression of IL1 at the mRNA level. We excluded that the release of IL-1 was due to cell death by simultaneous measurement of LDH release. It might be speculated that high IL-1 protein levels are obtained by secretion of IL-1 by monocytes and macrophages. In these cells inflammatory signals (such as LPS) promote synthesis and accumulation of the inactive pro-IL-1 form in the cytoplasm. In a second step, ATP triggers caspase-1 (IL-1 converting enzyme; ICE) mediated cleavage of pro-IL-1 and the subsequent release of the mature protein (Andrei et al., 2004; Wewers, 2004). A first explanation for the high IL-1 protein levels in dGal treated cells might be that dGal mimics the function of LPS and exogenous ATP in IL-1 production, whereas wtGal does not display this effect. Another possible explanation for the high IL-1 protein levels in dGal treated cells might be that lower amounts of IL-1 receptor are present in dGal treated cells as compared to wtGal or untreated cells. In a mouse model for colitis, early administration of 1 mg/kg recombinant wild-type human galectin-1 resulted in downregulation of IL-1 protein levels in plasma and in colon mucosa (Santucci et al., 2003). In our in vitro model we measure no change in IL-1 protein levels with wtGal and enhanced levels with dGal. The timing and complexity of different processes that play a role in vivo make the comparison with in vitro experiments difficult. In conclusion, we demonstrated that stable galectin-1 homodimers display enhanced activity with respect to IL-10 induction and structural changes related to apoptosis as compared to wild-type galectin-1 protein. In contrast to the variable changes observed for IL-2 and IFN-␥, the effects on IL-10 were consistent in all five donors. Although genetic heterogeneity in human might result in variation between the effectiveness of dGal induced effects on cytokine productions in individual persons, our results clearly demonstrate that treatment with dGal is an improved tool for the treatment of chronic inflammatory or autoimmune diseases. Acknowledgement This work was supported by a grant from the Graduate School, Groningen University Institute for Drug Exploration (GUIDE). References Allione, A., Wells, V., Forni, G., Mallucci, L., Novelli, F., 1998. Betagalactoside-binding protein (beta GBP) alters the cell cycle, up-regulates
J. van der Leij et al. / Molecular Immunology 44 (2007) 506–513 expression of the alpha- and beta-chains of the IFN-gamma receptor, and triggers IFN-gamma-mediated apoptosis of activated human T lymphocytes. J. Immunol. 161, 2114–2119. Andrei, C., Margiocco, P., Poggi, A., Lotti, L.V., Torrisi, M.R., Rubartelli, A., 2004. Phospholipases C and A2 control lysosome-mediated IL-1 beta secretion: Implications for inflammatory processes. Proc. Natl. Acad. Sci. U.S.A. 101, 9745–9750. Barondes, S.H., Castronovo, V., Cooper, D.N., Cummings, R.D., Drickamer, K., Feizi, T., Gitt, M.A., Hirabayashi, J., Hughes, C., Kasai, K., 1994. Galectins: a family of animal beta-galactoside-binding lectins. Cell 76, 597– 598. Battig, P., Saudan, P., Gunde, T., Bachmann, M.F., 2004. Enhanced apoptotic activity of a structurally optimized form of galectin-1. Mol. Immunol. 41, 9–18. Cho, M., Cummings, R.D., 1995. Galectin-1, a beta-galactoside-binding lectin in Chinese hamster ovary cells. I. Physical and chemical characterization. J. Biol. Chem. 270, 5198–5206. de Kruif, J., Logtenberg, T., 1996. Leucine zipper dimerized bivalent and bispecific scFv antibodies from a semi-synthetic antibody phage display library. J. Biol. Chem. 271, 7630–7634. Dias-Baruffi, M., Zhu, H., Cho, M., Karmakar, S., McEver, R.P., Cummings, R.D., 2003. Dimeric galectin-1 induces surface exposure of phosphatidylserine and phagocytic recognition of leukocytes without inducing apoptosis. J. Biol. Chem. 278, 41282–41293. Elola, M.T., Chiesa, M.E., Alberti, A.F., Mordoh, J., Fink, N.E., 2005. Galectin-1 receptors in different cell types. J. Biomed. Sci. 12, 13–29. Fajka-Boja, R., Szemes, M., Ion, G., Legradi, A., Caron, M., Monostori, E., 2002. Receptor tyrosine phosphatase, CD45 binds galectin-1 but does not mediate its apoptotic signal in T cell lines. Immunol. Lett. 82, 149–154. Levi, G., Tarrab-Hazdai, R., Teichberg, V.I., 1983. Prevention and therapy with electrolectin of experimental autoimmune myasthenia gravis in rabbits. Eur. J. Immunol. 13, 500–507. Nguyen, J.T., Evans, D.P., Galvan, M., Pace, K.E., Leitenberg, D., Bui, T.N., Baum, L.G., 2001. CD45 modulates galectin-1-induced T cell death: regulation by expression of core 2 O-glycans. J. Immunol. 167, 5697–5707. Offner, H., Celnik, B., Bringman, T.S., Casentini-Borocz, D., Nedwin, G.E., Vandenbark, A.A., 1990. Recombinant human beta-galactoside binding lectin suppresses clinical and histological signs of experimental autoimmune encephalomyelitis. J. Neuroimmunol. 28, 177–184. Pace, K.E., Lee, C., Stewart, P.L., Baum, L.G., 1999. Restricted receptor segregation into membrane microdomains occurs on human T cells during apoptosis induced by galectin-1. J. Immunol. 163, 3801–3811. Pace, K.E., Hahn, H.P., Pang, M., Nguyen, J.T., Baum, L.G., 2000. CD7 delivers a pro-apoptotic signal during galectin-1-induced T cell death. J. Immunol. 165, 2331–2334.
513
Perillo, N.L., Uittenbogaart, C.H., Nguyen, J.T., Baum, L.G., 1997. Galectin-1, an endogenous lectin produced by thymic epithelial cells, induces apoptosis of human thymocytes. J. Exp. Med. 185, 1851– 1858. Perillo, N.L., Pace, K.E., Seilhamer, J.J., Baum, L.G., 1995. Apoptosis of T cells mediated by galectin-1. Nature 378, 736–739. Rabinovich, G.A., Ariel, A., Hershkoviz, R., Hirabayashi, J., Kasai, K.I., Lider, O., 1999a. Specific inhibition of T-cell adhesion to extracellular matrix and proinflammatory cytokine secretion by human recombinant galectin-1. Immunology 97, 100–106. Rabinovich, G.A., 1999b. Galectins: an evolutionarily conserved family of animal lectins with multifunctional properties; a trip from the gene to clinical therapy. Cell Death Differ. 6, 711–721. Rabinovich, G.A., Daly, G., Dreja, H., Tailor, H., Riera, C.M., Hirabayashi, J., Chernajovsky, Y., 1999c. Recombinant galectin-1 and its genetic delivery suppress collagen-induced arthritis via T cell apoptosis. J. Exp. Med. 190, 385–398. Rabinovich, G.A., Ramhorst, R.E., Rubinstein, N., Corigliano, A., Daroqui, M.C., Kier-Joffe, E.B., Fainboim, L., 2002. Induction of allogenic T-cell hyporesponsiveness by galectin-1-mediated apoptotic and non-apoptotic mechanisms. Cell Death Differ. 9, 661–670. Santucci, L., Fiorucci, S., Cammilleri, F., Servillo, G., Federici, B., Morelli, A., 2000. Galectin-1 exerts immunomodulatory and protective effects on concanavalin A-induced hepatitis in mice. Hepatology 31, 399– 406. Santucci, L., Fiorucci, S., Rubinstein, N., Mencarelli, A., Palazzetti, B., Federici, B., Rabinovich, G.A., Morelli, A., 2003. Galectin-1 suppresses experimental colitis in mice. Gastroenterology 124, 1381–1394. van der Leij, J., van den Berg, A., Blokzijl, T., Harms, G., van Goor, H., Zwiers, P., van Weeghel, R., Poppema, S., Visser, L., 2004. Dimeric galectin-1 induces IL-10 production in T-lymphocytes: an important tool in the regulation of the immune response. J. Pathol. 204, 511– 518. Vespa, G.N., Lewis, L.A., Kozak, K.R., Moran, M., Nguyen, J.T., Baum, L.G., Miceli, M.C., 1999. Galectin-1 specifically modulates TCR signals to enhance TCR apoptosis but inhibit IL-2 production and proliferation. J. Immunol. 162, 799–806. Walzel, H., Schulz, U., Neels, P., Brock, J., 1999. Galectin-1, a natural ligand for the receptor-type protein tyrosine phosphatase CD45. Immunol. Lett. 67, 193–202. Walzel, H., Blach, M., Hirabayashi, J., Kasai, K.I., Brock, J., 2000. Involvement of CD2 and CD3 in galectin-1 induced signaling in human Jurkat T-cells. Glycobiology 10, 131–140. Wewers, M.D., 2004. IL-1beta: an endosomal exit. Proc. Natl. Acad. Sci. U.S.A. 101, 10241–10242.