Galectin-1 Couples Glycobiology to Inflammation in Osteoarthritis through the Activation of an NF-κB−Regulated Gene Network This information is current as of September 27, 2021. Stefan Toegel, Daniela Weinmann, Sabine André, Sonja M. Walzer, Martin Bilban, Sebastian Schmidt, Catharina Chiari, Reinhard Windhager, Christoph Krall, Idriss M. Bennani-Baiti and Hans-Joachim Gabius

J Immunol 2016; 196:1910-1921; Prepublished online 20 Downloaded from January 2016; doi: 10.4049/jimmunol.1501165 http://www.jimmunol.org/content/196/4/1910 http://www.jimmunol.org/

Supplementary http://www.jimmunol.org/content/suppl/2016/01/19/jimmunol.150116 Material 5.DCSupplemental References This article cites 57 articles, 6 of which you can access for free at: http://www.jimmunol.org/content/196/4/1910.full#ref-list-1 by guest on September 27, 2021 Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2016 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Galectin-1 Couples Glycobiology to Inflammation in Osteoarthritis through the Activation of an NF-kB–Regulated Gene Network

Stefan Toegel,* Daniela Weinmann,* Sabine Andre´,† Sonja M. Walzer,* Martin Bilban,‡ Sebastian Schmidt,† Catharina Chiari,* Reinhard Windhager,* Christoph Krall,x Idriss M. Bennani-Baiti,{ and Hans-Joachim Gabius†

Osteoarthritis is a degenerative joint disease that ranks among the leading causes of adult disability. Mechanisms underlying os- teoarthritis pathogenesis are not yet fully elucidated, putting limits to current disease management and treatment. Based on the phenomenological evidence for dysregulation within the glycome of chondrocytes and the network of a family of adhesion/growth- regulatory lectins, that is, galectins, we tested the hypothesis that Galectin-1 is relevant for causing degeneration. Immunohisto- Downloaded from chemical analysis substantiated that Galectin-1 upregulation is associated with osteoarthritic cartilage and subchondral bone his- topathology and severity of degeneration (p < 0.0001, n = 29 patients). In vitro, the lectin was secreted and it bound to osteoarthritic chondrocytes inhibitable by cognate sugar. Glycan-dependent Galectin-1 binding induced a set of disease markers, including matrix metalloproteinases and activated NF-kB, hereby switching on an inflammatory gene signature (p <10216). Inhibition of distinct components of the NF-kB pathway using dedicated inhibitors led to dose-dependent impairment of Galectin- 1–mediated transcriptional activation. Enhanced secretion of effectors of degeneration such as three matrix metalloproteinases http://www.jimmunol.org/ underscores the data’s pathophysiological relevance. This study thus identifies Galectin-1 as a master regulator of clinically relevant inflammatory-response genes, working via NF-kB. Because inflammation is critical to cartilage degeneration in osteo- arthritis, this report reveals an intimate relation of glycobiology to osteoarthritic cartilage degeneration. The Journal of Immu- nology, 2016, 196: 1910–1921.

steoarthritis, also known as osteoarthrosis or degenerative being. Beyond direct coverage of all hospitalization-related expenses,

arthritis, is the most prevalent form of arthritis and a osteoarthritis is therefore responsible for another $13 billion in lost by guest on September 27, 2021 O leading contributor to physical pain and disability, hereby productivity in the United States (2). Despite this enormously heavy shortening adult working life and causing considerable socioeco- human and economic toll, molecular mechanisms governing the nomic costs worldwide (1). The Center for Disease Control esti- pathogenesis of osteoarthritis are not yet fully understood. Advances mates osteoarthritis to afflict 13.9% of men and women aged 25–65 in clarifying molecular causes would have the potential to markedly and 33.6% of those older than 65. More than 27 million Americans improve disease management and treatment. Due to the emerging suffer from osteoarthritis, incurring a projected hospitalization cost significance of glycan-protein (lectin) recognition in diverse aspects of .$42 billion (2). Up to now, no cure for the progressive degen- of pathophysiology (3, 4), in this work we focus on a multifunctional eration of joints in articular cartilage and subchondral bone, mostly lectin of the galectin family. seen in knees, hips, hands, and spine, has been identified. In addition Galectins are potent regulators of adhesion and growth control to diminishing the joints’ flexibility and shock-buffering capacity, the via triggering outside-in signaling cascades after binding to distinct pain that accompanies osteoarthritis compromises the person’s well- cell surface glycans, with relevance for inflammation (5–7). Obvi- ously, coordinated changes in lectin and glycan presentation un- *Karl Chiari Lab for Orthopaedic Biology, Department of Orthopaedics, Medical derlie the involvement of glycobiology in pathogenesis. With respect University of Vienna, 1090 Vienna, Austria; †Institute of Physiological Chemistry, to osteoarthritis, we have characterized the human chondrocyte Faculty of Veterinary Medicine, Ludwig-Maximilians-University Munich, 80539 Munich, Germany; ‡Department of Laboratory Medicine and Core Facility Geno- glycophenotype, showing that it was profoundly altered in patients mics, Core Facilities, Medical University of Vienna, 1090 Vienna, Austria; xCenter suffering from cartilage degeneration (8, 9). Turning to galectins by for Medical Statistics, Informatics, and Intelligent Systems, Medical University { immunohistochemical screening, initial evidence had suggested that Vienna, 1090 Vienna, Austria; and B2 Scientific Group, 1180 Vienna, Austria Galectin-1 was associated to the level of degeneration in human Received for publication May 20, 2015. Accepted for publication December 9, 2015. osteoarthritic cartilage (10). This lectin is known to regulate cell-to- This work was supported by European Community Contract 260600 (GlycoHIT) and by the Verein zur Foerderung des biologisch-technologischen Fortschritts in der cell and cell-to-matrix communication via glycan bridging and Medizin e.V. (Heidelberg, Germany). subsequent signaling, acting as anti-inflammatory effector by in- Address correspondence and reprint requests to Prof. Stefan Toegel, Medical Uni- ducing apoptosis of activated T cells (6). In relevant tumors, it is a versity of Vienna, Waehringer Guertel 18-20, 1090 Vienna, Austria. E-mail address: marker useful for differential diagnosis between chondroblastic [email protected] osteosarcoma and conventional chondrosarcoma (7, 11). In porcine The online version of this article contains supplemental material. articular chondrocytes, Galectin-1 modulates cellular interactions Abbreviations used in this article: MMP, matrix metalloproteinase; MS, Mankin with constituents of the natural extracellular matrix as well as with score; RT-qPCR, quantitative RT-PCR; TFBS, transcription factor binding site. an artificial matrix (12, 13), and, obviously relevant for degenera- Copyright Ó 2016 by The American Association of Immunologists, Inc. 0022-1767/16/$30.00 tion, Galectin-1–dependent signaling upregulates .100-fold matrix www.jimmunol.org/cgi/doi/10.4049/jimmunol.1501165 The Journal of Immunology 1911 metalloproteinase (MMP)-13 gene transcription while at the same resin. Absence of contamination by bacterial LPS was ensured by a re- time downmodulating type II collagen (COL2A1) gene transcription spective protocol (26), as purity was rigorously checked by one- and two- (13). Reactivity to laminin, a well-known galectin counterreceptor dimensional gel electrophoresis and mass-spectrometric fingerprinting, as previously reported (27). Fluorescent labeling under activity-preserving in the extracellular matrix (14, 15), was documented for Galectin-1 conditions was performed as described, and carbohydrate-binding activ- in the intervertebral disc (16), and a positive Galectin-1 gradient ity was determined by cell-binding assays without or with cognate sugar from the proliferative to the hypertrophic zone was reported in used as an inhibitor (28). growth-plate cartilage (13). Cell culture Work on clinical aspects of rheumatoid arthritis has uncovered a significant contribution of galectins to its pathogenesis (17, 18). In Osteoarthritic and nonosteoarthritic chondrocytes were isolated from femo- ral condyles and tibial plateaus of articular cartilage following established pro- tissue specimens of these patients, Galectin-1 is mostly detected at tocols (8). Isolated chondrocytes were cultured in DMEM (Life Technologies) the synovial sublining layer and downregulated in the synovial fluid containing 10% FCS (Biochrom) and 100 mg/ml gentamicin (Life Technologies) (7, 19–22). Presence of specific autoantibodies, a common phe- in a humidified atmosphere of 5% CO2/95% air at 37˚C. To preserve the nomenon for galectins in autoimmune diseases (23), may account chondrocyte phenotype, only cells of passage 0 (osteoarthritic chondrocytes) for this decrease. Interestingly, the absence of Galectin-1 exacer- and passage 1 (nonosteoarthritic chondrocytes) were used for assays. Upon 90% confluency, chondrocyte cultures were starved overnight bated disease manifestation in a mouse model of collagen-induced using DMEM supplemented with gentamicin and treated with recombinant arthritis, in line with previous evidence of disease amelioration upon Galectin-1—in the presence or absence of the cognate sugar lactose (Sigma- its delivery to sites of inflammation (24). Collectively, these data Aldrich) to block binding to glycans—in starvation medium for the indi- k suggest that Galectin-1 may play a role in osteoarthritis pathogen- cated time periods. For inhibition of NF- B pathway components, inhibitors acting at distinct sites (Calbiochem; please see Fig. 7 for details on their mo- esis and physiology, which, however, might be different from the lecular targets and used concentrations) were added to the chondrocyte cul- Downloaded from protective (anti-inflammatory) activity in rheumatoid arthritis. tures 1 h prior to the stimulation with recombinant Galectin-1. Additionally, In this study, we first solidify and markedly extend the evidence chondrocyte cultures were stimulated with IL-1b,TNF-a,orIL-8(all for a significant correlation of Galectin-1 presence with the degree from BioLegend) to induce proinflammatory conditions in vitro. of osteoarthritic cartilage degeneration in patients, and then pre- Immunofluorescent detection of Galectin-1 binding sites sent data on Galectin-1 secretion, its lactose-inhibitable binding to A50ml cell suspension (3 3 105 cells, prepared by trypsinization of chondrocytes, and definitive effector capacity upregulating disease http://www.jimmunol.org/ markers, such as degradative enzymes on the mRNA and protein monolayer cultures) was incubated in presence and absence of 0.1 M lactose together with 50 ml PBS containing 10 mg Galectin-1-FITC for 10 levels. Next, transcriptomic analyses for osteoarthritic chondrocytes min at 4˚C, limiting internalization of cell-bound Galectin-1. After re- disclose that Galectin-1 exerts control over a gene signature relevant moving any unbound lectin by washing with PBS, cells were immedi- for degeneration in osteoarthritis. We demonstrate that, upon cell ately mounted for microscopy without fixation. Confocal images of surface binding in a lactose-inhibitable manner, Galectin-1 acts as fluorescence-labeled cells were obtained using a Carl Zeiss LSM 700 Laser Scanning Microscope at original magnification 363 and Zen master regulator of clinically relevant inflammatory-response genes software. centrally involving NF-kB, as unraveled by the systematic testing of aseriesoftarget-specificinhibitors. Quantitative RT-PCR by guest on September 27, 2021 Isolation of total RNA, cDNA synthesis, and SYBR Green–based quanti- Materials and Methods tative RT-PCR (RT-qPCR) experiments followed established protocols Clinical specimens (29, 30). Briefly, total RNA was extracted from Galectin-1–treated osteo- arthritic chondrocytes and examined for quality and quantity on the Agi- Osteoarthritic cartilage was obtained from osteoarthritis patients under- lent 2100 Bioanalyzer Nano LabChip (Agilent Technologies), prior to going total knee arthroplasty, whereas nonosteoarthritic cartilage specimens reverse transcription into cDNA. The protocols thoroughly followed of the knee joint were collected postmortem from individuals without the minimal guidelines for the design and documentation of quantitative macroscopic signs of arthritic cartilage degeneration. Clinical specimens PCR experiments, as recently outlined (31). A detailed quantitative PCR were obtained with written informed consent and in full accordance with the checklist is readily provided by the authors upon request. terms of the ethics committee of the Medical University of Vienna. ELISA Histological assessment The levels of ADAMTS-4 (BosterBio), Galectin-1, proMMP-1, total Specimens of articular knee cartilage from 29 osteoarthritis patients MMP-3, and proMMP-13 (all from R&D Systems), secreted into the cell (11 males, 18 females; 41–81 y) were selected macroscopically to cover a culture medium by Galectin-1–treated osteoarthritic chondrocytes, were wide range of degeneration stages in each patient. Paraffin sections of determined by commercial ELISAs, according to the manufacturers’ proto- mildly and severely degenerated cartilage were stained with Safranin O cols. ELISA standard curve ranges were 0.625–40 ng/ml in the case of (Sigma-Aldrich) (10). Thereafter, a varying number of regions (2–10 per ADAMTS-4, 0.313–20 ng/mg in the case of Galectin-1, 0.156–10 ng/ml in patient, depending on availability of regions differing in degeneration) was the cases of proMMP-1 and total MMP-3, and 78–5000 pg/ml in the case of graded according to the Mankin score (MS), as previously described (10). In proMMP-13. The lowest concentration that could be reliably determined was total, 231 regions were histologically graded prior to the processing of set as the detection limit. consecutive sections for immunohistochemical staining, as described below. Immunohistochemistry Microarray Immunohistochemical staining for Galectin-1 followed recently established Osteoarthritic chondrocytes were isolated from five male patients (47–78 y). m protocols (10). Briefly, tissue sections were incubated with rabbit polyclonal Following starvation, cells were incubated in the presence of 50 g/ml Ab against human Galectin-1 (25). Ag-dependent staining was developed recombinant Galectin-1 for 24 h. Total RNA was extracted and exten- using 3,39-diaminobenzidine tetrahydrochloride hydrate (FLUKA), whereas sively analyzed for quantity, purity, and integrity using the Agilent 2100 counterstaining was performed using Mayer’s hemalum solution (Merck). Bioanalyzer and the Nanodrop 1000. RNA integrity numbers were between The percentage of Galectin-1–positive chondrocytes in the regions of interest 9.1 and 10, and A260/A280 values were between 2.02 and 2.11. GeneChip was determined independently by two observers, as described (10). analysis was performed using 200 ng total RNA per sample. Preparation of terminally labeled cRNA, hybridization to genome-wide human PrimeView Production and quality control of recombinant Galectin-1 and GeneChips (Affymetrix), and scanning of the arrays were carried out FITC-labeled Galectin-1 (Galectin-1-FITC) according to the manufacturer’s protocols. Robust multichip average signal extraction, normalization, and filtering were performed as described using Human Galectin-1 was obtained by recombinant production and purified the comprehensive R- and bioconductor-based web service for microarray from bacterial extracts by affinity chromatography on lactosylated data analysis CARMAweb (32). A variation filter was applied for selecting Sepharose 4B, prepared by ligand coupling to divinyl sulfone-activated informative (i.e., significantly varying) genes. The filtering criteria for the 1912 GALECTIN-1 ACTIVATES INFLAMMATION IN OSTEOARTHRITIS exemplary data sets required an interquartile range of .0.5 and at least one Results . sample with expression intensity of 100. The microarray data discussed in Presence of Galectin-1 in articular chondrocytes correlates this publication have been deposited in the Gene Expression Omnibus da- tabase and are publicly accessible through Gene Expression Omnibus Series with osteoarthritis-related cartilage degeneration accession number GSE68760 (http://www.ncbi.nlm.nih.gov/geo/query/acc. To test the hypothesis for a role of Galectin-1 as bioeffector of cgi?token=snqzskeenhgznef&acc=GSE68760). osteoarthritis, we first assessed in detail the immunopositivity of Bioinformatic analyses chondrocytes for Galectin-1 in relation to the degeneration status Quantification of differential gene expression was carried out using a of osteoarthritic cartilage in 29 patients. Fig. 1A shows three moderated two-sample t test, as previously described (33). The gene set prototypical cartilage regions of a representative patient, as follows: enrichment analyses were performed, as previously reported (34). Briefly, one region with almost intact surface and tissue morphology (MS1); gene set enrichment analyses studies were conducted in the Molecular Signatures Database, and the MSigDB C2, C3, and C5 collections were searched for significant correlations. Overrepresentation of a given transcription factor binding site (TFBS) in any given gene set was ascertained by comparing the TFBS occurrence in the gene set with its occurrence in the human genome and in all known human promoters. The p values were obtained by hypergeometric testing and were further cor- rected for multiple testing using the Benjamini-Hochberg method. The threshold to consider p values as significant was 0.05. Promoter analysis for identification of putative binding sites for transcription factors was carried out, as described (35). Downloaded from Genes studied To avoid gene homonymy or species ambiguity (36), genes discussed in this work are listed below with their unique National Center for Biotechnology Information GeneID: ADAMTS4 (GeneID: 9507), AGC1 (176), CCL4 (6351), CCL5 (6352), COL2A1 (1280), CXCL1 (2919), CXCL2 (2920), CXCL3 (2921), CXCL5 (6374), FRZB (2487), GREM1 (26585), IL1B (3553), IL6 (3569), IL8 (3576), IRAK (3654), LGALS1 (3956), MMP1 http://www.jimmunol.org/ (4312), MMP3 (4314), MMP13 (4322), NFKB1 (4790), RELA (5970), SELE (6401), ST3GAL1 (6482), ST6GAL1 (6480), and TNFA (7124). Western blot Proteins were extracted from chondrocytes after treatment with recombi- nant Galectin-1, as indicated. Cells were thoroughly washed with PBS and lysed with radioimmunoprecipitation assay lysis buffer (Santa Cruz), supplemented with PMSF, sodium orthovanadate, and a protease inhibitor mixture, for 1 h on ice. The cell lysate was then further processed in an ice- cold ultrasonic bath, thoroughly vortexed, and centrifuged. Proteins in the by guest on September 27, 2021 supernatants were separated using 5% acrylamide stacking and 10% ac- rylamide separating gels and transferred to a nitrocellulose membrane by electroblotting (Protran; Schleicher & Schuell). Binding sites for proteins on the membrane were blocked overnight at 4˚C with PBS (Life Tech- nologies) containing 5% milk powder (Roth). Membranes were incubated for 2 h with anti-IkBa (N-terminal Ag, 1:1,000, mouse monoclonal; Cell Signaling), anti–phospho-IkBa (Ser32/36, 1:1,000, mouse monoclonal; Cell Signaling), anti–NF-kB p65 (1:1,000, rabbit monoclonal; Cell Signaling), or anti–phospho-NF-kB p65 (Ser536, 1:1,000, rabbit monoclonal; Cell Signaling) together with anti–b-actin (1:5,000, mouse monoclonal; Sigma- Aldrich) in Odyssey Blocking Buffer (LI-COR Biosciences), which had been diluted with PBS containing 0.1% Tween 20 (Bio-Rad). Thereafter, the membrane was incubated for 1 h with DyLight 800 nm-labeled goat anti- rabbit IgG and IRDye 680LT goat anti-mouse IgG, both diluted 1:15,000 in PBS containing 0.1% Tween 20. Finally, the immunoreactive protein bands were detected and quantified using the Odyssey Imager with Image Studio Software (LI-COR Biosciences). The ratio between p-IkBa and IkBa or FIGURE 1. Localization of Galectin-1 in human osteoarthritic knee between p-p65 and p65 (all normalized for b-actin) was expressed as relative cartilage and correlation of chondrocyte positivity with cartilage de- quantity in comparison with the untreated control set to 1. generation. (A) Histological sections of articular cartilage/subchondral bone (n = 29 patients) were stained with Safranin O and categorized Statistical analyses according to their degeneration status using the MS. Shown is an ex- emplary series from one representative donor presenting mildly (MS1), Statistical analysis of a correlation between Galectin-1 immunopositivity and the MS was performed using SAS 4.3. Pearson’s correlation coeffi- moderately (MS6), and severely (MS11) degenerated regions. Overview cients were calculated for each patient separately, and the Wilcoxon images were photomerged using Adobe Photoshop from single photo- signed-rank test was performed to test whether the median correlation graphs. Scale bars, 500 mm. (B) Immunohistochemistry of the articular coefficient was different from 0. Statistical comparison of the microarray cartilage sections shown in (A), using an Ab against Galectin-1 without data was performed using the limma algorithm (32), and multiple testing cross-reactivity to other human galectins. Scale bar, 50 mm. (C)From correction based on the false discovery rate was performed to produce each of the 29 patients, 2–10 (8 6 1.7) regions were histologically graded adjusted p values. Statistical analyses of RT-qPCR, ELISA, and quantita- according to the MS and consecutive sections were stained immuno- tive Western blot data were performed using SPSS 20.0. Normality of the histochemically for Galectin-1. The percentage of Galectin-1–positive data was analyzed using the Kolmogorov-Smirnov test, and statistical chondrocytes was assessed in each region. Shown are both the scatter- significance of the data was delineated using paired or unpaired t tests or Wilcoxon signed-rank test. All analysis units (n) given in figure and table plots of Galectin-1 positivity versus MS together with the regression lines legends—unless otherwise stated—refer to the number of independent for each patient (left panel) and with the regression line over all patients observations (biological replicates) underlying the respective descriptive (right panel). DZ, deep zone; MZ, middle zone; SB, subchondral bone; statistics and statistical tests. SZ, superficial zone; T, tidemark region. The Journal of Immunology 1913 one with considerable surface discontinuity and matrix depletion (MS6); and one with vertical clefts, matrix loss, formation of chondrocyte clusters, and blood vessels penetrating the tidemark (MS11). Immunohistochemical analysis of respective consecutive sections revealed that the presence of Galectin-1 was restricted to blood vessels and a subset of subchondral trabecular osteocytes in the MS1 region (Fig. 1B). In full agreement with previous findings in nonosteoarthritic cartilage from tumor patients (10), articular chondrocytes were immunonegative from the superficial to the deep zone in this intact cartilage region. In the MS6 region, by contrast, chondrocytes of the superficial zone were predominantly positive for Galectin-1, whereas no positivity was found in chondrocytes of the middle and deep zone. Most interestingly, we observed strong Galectin-1–dependent staining in all chondrocytes in the superficial zone of the MS11 region, together with focal staining in chon- drocytes in the middle zone. As depicted in Fig. 1B, the presence of Galectin-1 in middle-zone chondrocytes was preferentially local- ized near deep vertical clefts or fissures. Likewise, chondrocytes assembled in clusters were always positive for Galectin-1, even in Downloaded from the absence of vertical clefts (data not shown). From these data, extending our initial previous findings (10), we reasoned that the presence of Galectin-1 in osteoarthritic chon- drocytes is intimately connected with osteoarthritis severity and associated with histopathologic features of osteoarthritic cartilage

and subchondral bone. Therefore, we determined the percentage http://www.jimmunol.org/ of Galectin-1–positive chondrocytes in each region of the 29 test series and performed correlation analyses between Galectin-1 immunopositivity and the MS for each patient (Fig. 1C). Statis- tical analysis disclosed a strong correlation between Galectin-1 levels and degeneration severity (p , 0.0001). Sex-specific stratification of data excluded any significant difference between female and male patients (data not shown). Taken together, these results demonstrate that Galectin-1 accumu- lates in articular osteoarthritic chondrocytes of the superficial/middle by guest on September 27, 2021 zones and that the percentage of Galectin-1–positive chondrocytes correlates with increasing degeneration status of cartilage. As next step, we explored whether chondrocytes have the capacity to secrete Galectin-1 and whether expression and secretion of Galectin-1 might be induced by early signals of proinflammatory conditions. Secretion of Galectin-1 by isolated chondrocytes and lack of regulatory response to proinflammatory cytokines FIGURE 2. Galectin-1 is expressed and secreted by cultured chon- As shown in Fig. 2A, isolated osteoarthritic chondrocytes secreted drocytes and binds to cell surfaces via glycans. (A) Osteoarthritic chon- 1.65 6 2.27 ng/ml Galectin-1 into the supernatant within 24 h of drocytes (n = 4 patients) were starved overnight prior to treatment with b a b monolayer culture (n = 4 patients). Presence of the proinflammatory 10 ng/ml IL-1 , 10 ng/ml TNF- , a combination of 10 ng/ml IL-1 and a cytokines IL-1b,TNF-a, or IL-8 did not significantly alter the 10 ng/ml TNF- , or 100 ng/ml IL-8 for 24 h and collection of cell culture supernatants. Secreted Galectin-1 levels (ng/ml) were determined using amount of secreted Galectin-1 (Fig. 2A; p . 0.05). This finding was ELISA. Results are shown as mean 6 SD. (B) Osteoarthritic chondrocytes extended by RT-qPCR analysis of LGALS1 mRNA to gene expression (n = 4 patients) were starved overnight prior to treatment with 10 ng/ml IL-1b, levels in osteoarthritic chondrocytes (Fig. 2B) and nonosteoarthritic 10 ng/ml TNF-a, a combination of 10 ng/ml IL-1b and10ng/mlTNF-a,or chondrocytes (Fig. 2C). The data indicate that the expression and 100ng/mlIL-8for24handisolationoftotalRNA.LGALS1 mRNA levels secretion of Galectin-1 by isolated chondrocytes did not appear to be were determined using RT-qPCR. Results are expressed as fold changes sensitive to proinflammatory conditions. Of note, the same concen- (mean 6 SD) versus untreated controls set to 1. (C) Nonosteoarthritic chon- tration of TNF-a caused a strong signal enhancement for Galectin-1 drocytes (n = 3 individuals) were starved overnight prior to treatment with in Western blots of extracts from monocyte-dependent PBMCs, be- 10 ng/ml IL-1b or 10 ng/ml TNF-a for24handisolationoftotalRNA. ing as effective as stimulation by anti-CD3 (37). This underscores the LGALS1 mRNA levels were determined using RT-qPCR. Results are 6 D need to consider the cell type when studying the induction of Galectin-1. expressed as fold changes (mean SD) versus untreated controls set to 1. ( ) Cultured osteoarthritic chondrocytes were trypsinized and resuspended prior To act as an effector, extracellular Galectin-1 needs to bind to to labeling with Galectin-1-FITC (green) at 4˚C in the absence or presence of osteoarthritic chondrocytes. We thus next investigated Galectin-1– 0.1 M lactose. After 10 min of incubation, cells were washed and analyzed chondrocyte cell surface binding using Galectin-1-FITC. using laser-scanning microscopy, with the focus plane set to the center of cells. Fluorescence images as well as the corresponding differential interfer- Galectin-1 binds to osteoarthritic chondrocytes via glycans ence contrast images and overlay images are presented. Shown are the results and affects the expression of functional osteoarthritis markers from chondrocytes of one patient, representative for three independent ex- Chondrocytes labeled with Galectin-1-FITC exhibited fluo- periments (n =3patients). rescence staining of cell membranes (Fig. 2D). As documented, 1914 GALECTIN-1 ACTIVATES INFLAMMATION IN OSTEOARTHRITIS addition of cognate sugar (lactose) precluded binding of Thus, we next measured the effects of Galectin-1 on functional Galectin-1 to the surface of chondrocytes. This evidence for marker genes in cultured human chondrocytes with LPS-free glycan-dependent binding intimated that Galectin-1 may trig- Galectin-1 (26, 27). ADAMTS4 (Fig. 3A), MMP1 (Fig. 3B), ger an outside-in signaling by glycoconjugate cross-linking MMP3 (Fig. 3C), and MMP13 (Fig. 3D) mRNA levels were (lattice formation). This process may modulate chondrocyte found to be significantly upregulated by Galectin-1 in a gene expression. concentration-dependent manner. Interestingly, Galectin-1 sig- Downloaded from http://www.jimmunol.org/ by guest on September 27, 2021

FIGURE 3. Galectin-1 affects the expression of genes for matrix-related proteins and enzymes via binding to glycans. (A–G)Osteoarthriticchon- drocytes (n = 5 patients) were starved overnight prior to treatment with 1–50 mg/ml Galectin-1 for 24 h and isolation of total RNA and collection of supernatants. RT-qPCR was performed for (A) ADAMTS4,(B) MMP1,(C) MMP3,(D) MMP13,(E) AGC1, and (F) COL2A1 using validated assays. Results are expressed as fold changes (mean 6 SD) versus untreated controls set to 1. *p , 0.05 (paired t test). (G) Cell culture supernatants were monitored by ELISA for ADAMTS-4 (ng/ml), proMMP-1 (ng/ml), total MMP-3 (ng/ml), and proMMP-13 (pg/ml); *p , 0.05 (paired t test). (H–M) Osteoarthritic chondrocytes (n = 3 patients) were starved overnight prior to treatment with 50 mg/ml Galectin-1 for 24 h in presence or absence of 0.1 or 0.2 M b-lactose. mRNA levels of (H) ADAMTS4,(I) MMP1,(J) MMP3,(K) MMP13,(L) AGC1, and (M) COL2A1 were assessed using RT-qPCR and expressed as percentage (mean 6 SD) of Galectin-1–treated chondrocytes. #p , 0.05 (paired t test versus Galectin-1–treated chondrocytes). The Journal of Immunology 1915 Downloaded from

FIGURE 4. The 20 most up- and downregulated genes in osteoarthritic chondrocytes using Galectin-1 as ef- http://www.jimmunol.org/ fector. (A) Osteoarthritic chondrocytes (n = 5 patients; numbered with “1–5”) were starved overnight prior to treatment with 50 mg/ml Galectin-1 for 24 h. Following microarray analysis, heat maps of the 20 most upregu- lated genes and the 20 most downregulated genes were plotted. (B) For these genes, the ratio between mRNA levels in Galectin-1–treated versus untreated chon- drocytes across all five patients was calculated. The p values, corrected for multiple hypothesis testing by the

Benjamini-Hochberg method (BH), are also given. (C) by guest on September 27, 2021 The results of the microarray experiments were ascer- tained using RT-qPCR in the same cDNA samples as used in the microarray analysis. Values are given as fold changes with respect to untreated controls set to 1. Means and SDs were computed, normality of the data was analyzed using Kolmogorov-Smirnov (KS) test, and statistical significance of the data was examined using paired t test and Wilcoxon signed-rank test. 1916 GALECTIN-1 ACTIVATES INFLAMMATION IN OSTEOARTHRITIS nificantly and concentration dependently reduced COL2A1 and Galectin-1 establishes a chondrocyte mRNA signature AGC1 mRNA expression (Fig. 3E, 3F). At the protein level, associated with inflammation and NF-kB signaling Galectin-1–treated chondrocytes secreted large amounts of Indeed, genome-wide analysis revealed a set of differentially regulated , proMMP-1, MMP-3, and proMMP-13 (p 0.05; Fig. 3G), again genes of particular interest. Among the most upregulated genes in the in a concentration-dependent manner, whereas ADAMTS-4 protein presence of Galectin-1, a group of genes for chemokine ligands, in- levels were not significantly modified by Galectin-1. In agreement cluding CXCL3, CCL4, CCL5, CXCL2, CXCL1,andCXCL5, together with inhibition of Galectin-1 binding to chondrocytes in the pres- with other genes such as SELE and IL8, was found (Fig. 4A, 4B). On ence of lactose (Fig. 2), effects of Galectin-1 on mRNA levels were the other side of the spectrum, bone morphogenetic protein antagonist dose dependently reduced to basal levels by lactose in the cases of GREM1 was among the most downregulated genes. RT-qPCR ex- ADAMTS4, MMP1, MMP3,andMMP13 (Fig. 3H, 3I, respectively), also confirming the lack of LPS contamination in the Galectin-1 periments confirmed that IL8, IL6, SELE,andST3GAL1 were strongly preparation used. In contrast, however, the modest Galectin-1–in- upregulated, whereas GREM1, FRZB,andST6GAL1 (an enzyme in- duced reduction of mRNA expression COL2A1 and AGC1 was not troducing a block to Galectin-1 binding of N-glycans) were signifi- significantly reversed by lactose (Fig. 3J, 3K). cantly downregulated by Galectin-1 (Fig. 4C). Because some genes are Together, these findings indicated that, by binding to chon- highly sensitive to gene dosage, we also list in Supplemental Table I drocytes in a lactose-inhibitable manner, Galectin-1 upregulates moderately modulated genes, including Galectin-1–regulated glyco- presence of functional disease markers involved in matrix deg- genes (enzymes and lectins) and osteoarthritis-related genes. These radation. To take this activity to the genome level and characterize genes may also contribute to disease progression alone or in com- the route to these disease-associated effects, we carried out bination and may thus be worth pursuing in future studies. Downloaded from genome-wide and molecular analyses, to uncover the capacity of The genome-wide analyses showed that treatment of primary Galectin-1 to differentially modulate genes with relevance to human chondrocytes with Galectin-1 induces an inflammation gene osteoarthritis pathogenesis. signature with very high significance (p , 10216). The Galectin- http://www.jimmunol.org/ by guest on September 27, 2021

FIGURE 5. Bioinformatic investigation of genes induced in chondrocytes treated with Galectin-1. Data shown are Venn diagram rep- resentations. The size of circles is indicative of the gene set size, and the size of intersections, also indicated with a number in bold, depicts the number of shared genes between the two gene sets. (A) C2 analysis of perturbation signatures associated with Galectin-1 treatment: the 12 most significant signatures are listed from the most (1) to least significant (12) based on the extent of gene overlap between the chon- drocytes/Galectin-1 gene set and paired gene signatures. (B) C2 analysis of canonical and KEGG pathways associated with Galectin-1 treatment: the eight most significant signatures are ranked from the most (1) to least significant (8). (C) C2 analysis of reactome pathways as- sociated with Galectin-1 treatment: the top four signatures are listed by decreasing significance from 1 to 4. The Journal of Immunology 1917

Table I. C5-GSEA analysis of Galectin-1–treated chondrocytes

Gene Set Name Description Genes in Overlap (k) Genes in Gene Set (K) k/K p Value Inflammatory response Genes annotated by the GO term 21 129 0.1628 1.87 3 1026 GO:0006954. The immediate defensive reaction (by vertebrate tissue) to infection or injury caused by chemical or physical agents. The process is characterized by local vasodilation, extravasation of plasma into intercellular spaces, and accumulation of WBCs and macrophages. Response to wounding Genes annotated by the GO term 25 190 0.1316 1.06 3 1025 GO:0009611. A change in state or activity of a cell or an organism (in terms of movement, secretion, enzyme production, gene expression, etc.) as a result of a stimulus indicating damage to the organism. Response to extrastimulus Genes annotated by the GO term 29 312 0.0929 1.17 3 1023 GO:0009605. A change in state or activity of a cell or an organism (in terms of movement, secretion, enzyme production, gene expression, etc.) as a result of an external Downloaded from stimulus. Defense response Genes annotated by the GO term 26 270 0.0963 1.26 3 1023 GO:0006952. Reactions, triggered in response to the presence of a foreign body or the occurrence of an injury, which result in restriction of damage to the organism attacked or prevention/recovery from the http://www.jimmunol.org/ infection caused by the attack. Locomotory behavior Genes annotated by the GO term 11 95 0.1158 8.77 3 1023 GO:0007626. The specific movement from place to place of an organism in response to external or internal stimuli. Locomotion of a whole organism in a manner dependent upon some combination of that organism’s internal state and external conditions. K, number of genes included in biological processes; k, number of overlapping genes induced by Galectin-1. by guest on September 27, 2021

1–imparted inflammation signature showed up in three different The activity of Galectin-1 in osteoarthritic chondrocytes is types of analyses, including those of reactome pathways, chemical mediated via the NF-kB pathway and genetic perturbation signatures, and canonical and KEGG Based on the bioinformatic findings, we next sought to bio- signaling pathways (Fig. 5). An additional analysis of signaling chemically test the role of NF-kB in the signal transduction molecules downstream of Galectin-1 reveals a major overlap triggered by Galectin-1. Therefore, the direct impact of Galectin-1 with several biological processes (e.g., response to wound or de- on the activation of the canonical NF-kB pathway was assessed fense mechanisms), wherein inflammation is a core component over time (15 min–16 h). As presented in Fig. 6A (first lane), (Table I). Analysis of TFBS overrepresented in promoters of Galectin-1 significantly induced the phosphorylation of IkBa in genes selectively upregulated following induction by Galectin-1 a time-dependent manner, with a peak at 1 h after starting of revealed strong representation of an NF-kB (and NF-kB subunits the treatment and subsequent reduction to basal levels. Taking the RELA and NFKB1; p , 5.9 3 1023– 1.55 3 1026) signature, levels of IkBa into account (Fig. 6A, second lane), the ratio of suggesting that the Galectin-1–induced switch to a prodegener- p-IkBa/IkBa strongly and time dependently increased at 1 and 4 h ative expression program in human osteoarthritic chondrocytes (Fig. 6B; n = 3 patients; p , 0.05), pointing to phosphorylation and largely involves NF-kB (Table II). proteolytic degradation of IkBa as a result of exposure to Galectin-

Table II. C3-GSEA analysis of Galectin-1–treated chondrocytes

Transcription Factor Transcription Factor Binding Motif k/K p Value c-rel SGGRNTTTCC 0.0820 1.54 3 1026 NF-kB GGGAMTTYCC 0.0757 1.55 3 1025 NF-kB (RelA/p65) GGGRATTTCC 0.0717 8.51 3 1025 NF-kB GGGNNTTTCC 0.0896 1.21 3 1024 NF-kB KGGRAANTCCC 0.0647 6.65 3 1024 NF-kB GGGGAMTTTCC 0.0591 1.66 3 1023 IRF1 SAAAAGYGAAACC 0.0560 3.79 3 1023 NF-kB GGGACTTTCCA 0.0532 5.90 3 1023 SOX9 AACAATRG 0.0549 6.15 3 1023 AP1 TGANTCA 0.0366 7.17 3 1023 The top 10 overrepresented TFBS are listed in the order of significance. k/K: extent of overlap with all genes known to possess this motif. Nucleotide nomenclature: K: G or T; M: A or C; N: A, C, G, or T; R: A or G; S: C or G; Y: C or T. 1918 GALECTIN-1 ACTIVATES INFLAMMATION IN OSTEOARTHRITIS

7C); 2) the IKK complex components IKKa,IKKb,andNEMO (Fig. 7E, 7F); 3) IkBa phosphorylation (Fig. 7G) and proteasomal degeneration of IkBa (Fig. 7H); 4) translocation of NF-kBtothe nucleus (Fig. 7J, 7K); and 5) the histone acetyltransferases (CBP and p300) and histone deacetylases (Fig. 7L, 7M). In comparison, in- hibition of IRAK and TPL-2 resulted in a modest, albeit significant, reduction of Galectin-1 activity (Fig. 7D, 7I, respectively). Evi- dently, the activation of IRAK and the TPL-2/ERK pathway via NF-kB components might not represent major routes of Galectin-1 signaling in osteoarthritic chondrocytes. Of note, activation for IL1B transcription by Galectin-1 was reduced to basal levels in the presence of the cognate sugar lactose in a dose-dependent manner (data not shown). Taken together, these experiments reveal that blocking of the NF-kB pathway using several different strategies significantly diminished Galectin-1–induced IL1B expression. Therefore, canonical NF-kB signaling is likely to play a key role in the gene- regulatory activity of Galectin-1 in osteoarthritic chondrocytes, and this signaling is mediated by glycan-dependent Galectin-1 bind-

ing to the cell surface. Downloaded from

Discussion Until the 1990s, osteoarthritis was thought of mostly as a me- chanical, “wear and tear” condition that ultimately leads to cartilage degradation. Several studies have thereafter supported a shift to the

concept that inflammation is central to disease progression in both http://www.jimmunol.org/ early and late osteoarthritis (38). As to degradation, members of the MMP family, active in breakdown of extracellular matrix in both physiological and pathological processes (39), along with proin- flammatory genes are induced via the NF-kB pathway in osteoar- thritic chondrocytes. It is now accepted that the combined activities of these pathways are crucial to cause cartilage damage in osteo- arthritis (40). The upstream processes are, however, not yet fully identified, and molecular effectors initiating the deleterious sig- FIGURE 6. Galectin-1 activates the NF-kB pathway in osteoarthritic chon- by guest on September 27, 2021 drocytes. (A–C) Quantitative Western blot analyses of the phosphorylation of naling cascade are actively sought, as they may offer both etio- IkBa and p65 in osteoarthritic chondrocytes (n = 3 patients), which were starved logical insights and novel therapeutic opportunities. Molecular overnight and treated with 50 mg/ml Galectin-1 for 15 min, 1 h, 4 h, and 16 h. targets that would allow interference with disease onset or pro- b-Actin was used as loading control. (A) Shown are the blots of one repre- gression may indeed have beneficial impact globally on the lives of sentative patient for phospho-IkBa (p-IkBa), IkBa, phospho-p65 (p-p65), p65, hundreds of millions of affected individuals. and b-actin. (B) Shown are the ratios between p-IkBa and IkBa (normalized for In previous studies, we had revealed that several members of the b-actin) over time. Data are expressed as relative quantity in comparison with galectin family are expressed in vivo in osteoarthritis (10), and that , the untreated control set to 1. *p 0.05 (paired t test versus untreated control). specific features of the chondrocyte glycophenotype are altered C b ( ) Shown are the ratios between p-p65 and p65 (normalized for -actin) over during osteoarthritic conditions (8, 9). In this study, we focused on time. Data are expressed as relative quantity in comparison with the untreated Galectin-1, by immunohistochemical analysis and work with hu- control set to 1. *p , 0.05 (paired t test versus untreated control). man osteoarthritic chondrocytes. When comparing regions of dif- ferent degrees of tissue damage, Galectin-1 expression in articular 1. Fittingly, this treatment led to the phosphorylation of p65 in a chondrocytes is significantly correlated with the level of cartilage time-dependent manner with a peak at 1 h (Fig. 6A, third lane), degeneration (p , 0.0001). In particular, Galectin-1 accumulated whereas p65 levels were mildly reduced at 4 h (fourth lane). Cor- in cartilage zones most closely associated with matrix depletion, respondingly, p-p65/p65 ratios were significantly and time depen- chondrocyte clustering, and surface erosion, whereas it was absent dently increased at 1 and 4 h (Fig. 6C; n = 3 patients; p , 0.05), in chondrocytes of the deep zone. The pattern of Galectin-1 ex- indicating the activation of p65, which sets the stage for nuclear pression distribution in osteoarthritic cartilage was also congruent translocation and NF-kB–mediated gene activation. with the previously reported staining profiles of MMP1, MMP3, We next evaluated whether the canonical NF-kB pathway MMP8, and MMP13 as well as IL-1b and TNF-a (41). Taken to- (depicted in Fig. 7A) was of functional relevance to the activity gether, these results suggest a functional link between Galectin-1 of Galectin-1 in osteoarthritic chondrocytes. We thus specifically and the disease manifestation, possibly through upregulation of blocked several components of the NF-kB pathway using dedi- proinflammatory cytokines and/or the MMPs. In this concept, cated inhibitors (targets indicated in Fig. 7A) and determined the Galectin-1 would act as master regulator on a set of disease markers effect of Galectin-1 on the transcription of IL1B as a representa- in an auto- and paracrine manner via glycan binding. tive example of an NF-kB–dependent gene. To explore whether Galectin-1 has this capacity, we first Fig. 7B–M shows the percentage of Galectin-1 activity in the ascertained the following: 1) Galectin-1 secretion from osteoar- presence of the inhibitors, with Galectin-1 activity in the absence of thritic chondrocytes using ELISA as well as 2) glycan-dependent inhibitors set to 100%. Dose-dependent and marked impairment of binding of Galectin-1 to the chondrocyte surface by Galectin-1- Galectin-1–mediated IL1B transcription was observed following FITC. Testing chondrocytes from clinical specimens, we next specific inhibition of the following: 1) PDK-1 and TAK-1 (Fig. 7B, addressed the question whether exposure to Galectin-1 could The Journal of Immunology 1919 Downloaded from http://www.jimmunol.org/

FIGURE 7. Targeted inhibition of NF-kB mediators diminishes Galectin-1–mediated transcription of IL1B.(A) Scheme of the canonical NF-kB acti- vation pathway. Upon binding of Galectin-1 to osteoarthritic chondrocytes, the NF-kB signaling cascade is induced, leading to transcription of NF-kB– related genes such as IL1B. Multiple components and processes of this pathway were blocked by specific inhibitors, as indicated. Dashed arrows indicate

currently unknown mechanisms of signal transduction. (B–M) Osteoarthritic chondrocytes were starved overnight and treated with 50 mg/ml Galectin-1 in by guest on September 27, 2021 the presence or absence of inhibitors for 24 h prior to RT-qPCR analyses of IL1B mRNA levels. The graphs show the percentage of Galectin-1 activity in the presence of the inhibitors (Galectin-1 activity in the absence of inhibitors equals 100%). Labels on x-axes indicate the used concentrations of the inhibitors (mM), which were selected to be nontoxic and to span across the 50% inhibition benchmark whenever possible. The experiment was repeated three times with cells from three different patients. Shown are the mean 6 SD (two technical replicates) from one representative experiment. Statistics were performed in comparison with Galectin-1 activity in the absence of inhibitors (100 6 8.5%), *p , 0.05 (unpaired t test). contribute to known functional characteristics of cartilage de- and tissues react this way when stimulated with TNF-a,IL-1, generation. Biochemical and molecular analyses lend credence to LPS, and IKK inhibitors (47, 48), pointing to Galectin-1 as a this concept, as Galectin-1 markedly upregulated gene expression bona fide proinflammatory coordinator in human chondrocytes. of matrix-degrading enzymes (e.g., ADAMTS4, MMP1, MMP3, Moreover, the fact that the top six TFBS most significantly as- MMP13), and significantly downmodulated extracellular matrix sociated with genes induced by Galectin-1 are binding sites for components such as the AGC1 aggrecan and COL2A1 collagen. NF-kB further intimates the notion that NF-kB is a mediator of Relevant for cartilage breakdown, ELISA measurements showed the Galectin-1–triggered proinflammatory activities. that overexpression of the MMP genes is associated with signifi- Importantly, the lectin itself is not under the control of early cantly increased secretion of proMMP-1 (6.8-fold), MMP-3 (9.1- proinflammatory signals in chondrocytes, as shown in this work. fold), and proMMP-13 (113.8-fold). These results, on the levels As the reactivity of mononuclear cells to TNF-a attests (37), of production of mRNA and protein, pinpoint Galectin-1 as a po- extrapolations from experience with a certain cell type are not tent inducer of extracellular presence of MMPs in osteoarthritic valid. Because the expression of Galectin-1 can be triggered chondrocytes. Similar regulatory events had been reported for by various regulators with a strictly cell-type special activity Galectin-3 and MMP-1 and MMP-9 in murine B16F10 melanoma (Supplemental Table II) and can be under the control of a wide cells and for MMP-9 in corneal keratinocytes (42–44), for MMP-7 variety of transcription factors (data not shown), the search for the upregulation by Galectin-7 in murine and human lymphoma cells, pathophysiological elicitor of Galectin-1 overexpression in oste- and for coexpression of Galectin-7 and MMP-9 in sections of oarthritic cartilage is warranted. human laryngeal cancer (45, 46). In this context, our finding of To directly substantiate the involvement of NF-kB downstream enhanced bioavailability of effectors of degeneration prompted of Galectin-1, we showed that Galectin-1 induced the activation of us to perform transcriptomic analyses. IkBa and p65. Underscoring the salient message of the funda- Using genome-wide analyses, Galectin-1 was revealed to mental difference of Galectin-1 activity in diverse cell types, consistently and strongly trigger an inflammatory expression Galectin-1 impaired proteasomal IkBa degradation in mononu- profile in patients’ chondrocytes. This signature highly overlaps clear cells to attenuate the immune response (37) and reduced the with expression profiles in inflammation. Various types of cells phosphorylation of p65/IKKa/b in human LS-180 colorectal ad- 1920 GALECTIN-1 ACTIVATES INFLAMMATION IN OSTEOARTHRITIS enocarcinoma cells (49). Moreover, several NF-kB pathway in- References hibitors impaired Galectin-1–mediated induction of IL1B gene 1. Glyn-Jones, S., A. J. R. Palmer, R. Agricola, A. J. Price, T. L. Vincent, expression. These findings are consistent with reports implicating H. Weinans, and A. J. Carr. 2015. Osteoarthritis. Lancet 386: 376–387. 2. Centers for Disease Control and Prevention Public Health Service U. S. Galectin-1 in the positive regulation of NF-kB gene targets, in Department of Health and Human Services. 2010. Osteoarthritis and you: contrast to the well-documented anti-inflammatory role. In fact, patient information from the CDC. J. Pain Palliat. Care Pharmacother. 24: 430–431. in human mesenchymal stem cells, Galectin-1 activates NF-kB– 3. Gabius, H.-J. 2015. The magic of the sugar code. Trends Biochem. Sci. 40: 341. dependent fibronectin, laminin-5, and MMP-2 expression (50); 4. Solı´s, D., N. V. Bovin, A. P. Davis, J. Jime´nez-Barbero, A. Romero, R. Roy, K. Smetana, Jr., and H.-J. Gabius. 2015. A guide into glycosciences: how in stimulated rat pancreatic stellate cells, Galectin-1 increased chemistry, biochemistry and biology cooperate to crack the sugar code. Biochim. the production of chemokines (51, 52); and in tissue of squa- Biophys. Acta 1850: 186–235. 5. Kaltner, H., and H.-J. Gabius. 2012. A toolbox of lectins for translating the sugar mous cell, head and neck carcinomas factors associated with code: the galectin network in phylogenesis and tumors. Histol. Histopathol. 27: NF-kB activation and for poor prognoses such as TRIM23 and 397–416. 6. Liu, F.-T., R.-Y. Yang, and D. K. Hsu. 2012. Galectins in acute and chronic MAP3K2 (53). inflammation. Ann. N. Y. Acad. Sci. 1253: 80–91. The genome-wide expression profiling further revealed a 7. Smetana, K., Jr., S. Andre´, H. Kaltner, J. Kopitz, and H.-J. Gabius. 2013. number of regulated genes that add to the range of Galectin-1– Context-dependent multifunctionality of galectin-1: a challenge for defining the lectin as therapeutic target. Expert Opin. Ther. Targets 17: 379–392. dependent activities in osteoarthritic chondrocytes. First, we 8. Toegel, S., M. Pabst, S. Q. Wu, J. Grass, M. B. Goldring, C. Chiari, A. Kolb, found that GREM1 and FRZB were significantly downregulated F. Altmann, H. Viernstein, and F. M. Unger. 2010. Phenotype-related differential by Galectin-1. As both factors had recently been suggested to be a-2,6- or a-2,3-sialylation of glycoprotein N-glycans in human chondrocytes. Osteoarthritis Cartilage 18: 240–248. natural brakes on terminal hypertrophic differentiation in healthy 9. Toegel, S., D. Bieder, S. Andre´, F. Altmann, S. M. Walzer, H. Kaltner, Downloaded from joint cartilage (54), Galectin-1–dependent regulation of GREM1 J. G. Hofstaetter, R. Windhager, and H.-J. Gabius. 2013. Glycophenotyping of osteoarthritic cartilage and chondrocytes by RT-qPCR, mass spectrometry, his- and FRZB might contribute to the chondrocyte hypertrophy-like tochemistry with plant/human lectins and lectin localization with a glycoprotein. changes reported in osteoarthritic cartilage (55). In addition, Arthritis Res. Ther. 15: R147. Galectin-1 presence modulated expression of genes for compo- 10. Toegel, S., D. Bieder, S. Andre´, K. Kayser, S. M. Walzer, G. Hobusch, R. Windhager, and H.-J. Gabius. 2014. Human osteoarthritic knee cartilage: nents of the extracellular matrix such as integrins, for Galectin-3 fingerprinting of adhesion/growth-regulatory galectins in vitro and in situ indi- and members of other lectin families, including E-selectin, cates differential upregulation in severe degeneration. Histochem. Cell Biol. 142: http://www.jimmunol.org/ which serves as homing receptor for leukocyte infiltration, as 373–388. 11. Gomez-Brouchet, A., F. Mourcin, P.-A. Gourraud, C. Bouvier, G. De Pinieux, well as the sialyltransferases ST3GAL1 and ST6GAL1, which S. Le Guelec, P. Brousset, M.-B. Delisle, and C. Schiff. 2010. Galectin-1 is a can reshape the osteoarthritic cartilage glycophenotype and Galectin-1 powerful marker to distinguish chondroblastic osteosarcoma and conventional reactivity. chondrosarcoma. Hum. Pathol. 41: 1220–1230. 12. Marcon, P., E. Marsich, A. Vetere, P. Mozetic, C. Campa, I. Donati, F. Vittur, Having provided these new insights into Galectin-1 functionality A. Gamini, and S. Paoletti. 2005. The role of Galectin-1 in the interaction between chondrocytes and a lactose-modified chitosan. Biomaterials 26: in osteoarthritic chondrocytes, a clear direction emerges for further 4975–4984. work. In addition to this family member, presence of Galectins-3, -4, 13. Marsich, E., P. Mozetic, F. Ortolani, M. Contin, M. Marchini, A. Vetere, and -8 had been revealed in clinical specimens of osteoarthritis (10). S. Pacor, S. Semeraro, F. Vittur, and S. Paoletti. 2008. Galectin-1 in cartilage: expression, influence on chondrocyte growth and interaction with ECM com- As it remains an open question whether they can form a network of ponents. Matrix Biol. 27: 513–525. by guest on September 27, 2021 regulators, it is an attractive challenge to work with mixtures mim- 14. Zhou, Q., and R. D. Cummings. 1993. L-14 lectin recognition of laminin and its promotion of in vitro cell adhesion. Arch. Biochem. Biophys. 300: 6–17. icking the galectin levels in the pathophysiological environment. On 15. Andre´, S., S. Kojima, N. Yamazaki, C. Fink, H. Kaltner, K. Kayser, and H.-J. Gabius. the biochemical side, counterreceptor characterization, for the protein/ 1999. Galectins-1 and -3 and their ligands in tumor biology: non-uniform properties lipid and its glycosylation, will define the first steps of the trigger in cell-surface presentation and modulationofadhesiontomatrixglycoproteinsfor various tumor cell lines, in biodistribution of free and liposome-bound galectins and cascade, from initial binding to signaling. In fact, Galectin-1 is in their expression by breast and colorectal carcinomas with/without metastatic known to target only few glycoconjugates despite the abundance of propensity. J. Cancer Res. Clin. Oncol. 125: 461–474. 16. Jing, L., S. So, S. W. Lim, W. J. Richardson, R. D. Fitch, L. A. Setton, and b-galactosides on the cell surface, the a5b1-integrin being especially INK4a J. Chen. 2012. Differential expression of galectin-1 and its interactions with cells relevant for growth regulation (56), and p16 -dependent and laminins in the intervertebral disc. J. Orthop. Res. 30: 1923–1931. reprogramming of counterreceptor expression and glycosylation 17. Li, S., Y. Yu, C. D. Koehn, Z. Zhang, and K. Su. 2013. Galectins in the path- ogenesis of rheumatoid arthritis. J. Clin. Cell. Immunol. 4: 1000164. may serve as precedent (27, 28, 57). That the counterreceptor(s), 18. Ahmed, T. J., M. K. Kaneva, C. Pitzalis, D. Cooper, and M. Perretti. 2014. on the level of the protein and/or its glycosylation, may be subject to Resolution of inflammation: examples of peptidergic players and pathways. NF-kB–dependent regulation, as is the case for CD7 (a glycoprotein Drug Discov. Today 19: 1166–1171. 19. Ohshima, S., S. Kuchen, C. A. Seemayer, D. Kyburz, A. Hirt, S. Klinzing, counterreceptor in apoptosis induction of activated T cells) in mu- B. A. Michel, R. E. Gay, F.-T. Liu, S. Gay, and M. Neidhart. 2003. Galectin 3 and rine L7 lymphoma cells (58), is a tempting assumption. These results its binding protein in rheumatoid arthritis. Arthritis Rheum. 48: 2788–2795. may disclose clues to identify means to interfere with Galectin-1’s 20. Neidhart, M., C. A. Seemayer, K. M. Hummel, B. A. Michel, R. E. Gay, and S. Gay. 2003. Functional characterization of adherent synovial fluid cells in chondrocyte-specific disease-promoting effects, for example, by en- rheumatoid arthritis: destructive potential in vitro and in vivo. Arthritis Rheum. hancing a2,6-sialylation to preclude Galectin-1 binding or blocking 48: 1873–1880. 21. Dasuri, K., M. Antonovici, K. Chen, K. Wong, K. Standing, W. Ens, H. El-Gabalawy, Galectin-1/counterreceptor expression. This strategy based on the and J. A. Wilkins. 2004. The synovial proteome: analysis of fibroblast-like syno- reported discovery may lead to an innovative therapeutic approach viocytes. Arthritis Res. Ther. 6: R161–R168. for osteoarthritis. 22. Xibille´-Friedmann, D., C. Bustos Rivera-Bahena, J. Rojas-Serrano, R. Burgos- Vargas, and J.-L. Montiel-Herna´ndez. 2013. A decrease in galectin-1 (Gal-1) levels correlates with an increase in anti-Gal-1 antibodies at the synovial level in patients with rheumatoid arthritis. Scand. J. Rheumatol. 42: 102–107. Acknowledgments 23. Sarter, K., C. Janko, S. Andre´,L.E.Mun˜oz, C. Schorn, S. Winkler, J. Rech, Insightful discussions with Drs. J. Domingo-Ekark, B. Friday, and H. Kaltner, H. M. Lorenz, M. Schiller, et al. 2013. Autoantibodies against W. Notelecs are gratefully acknowledged. We thank B. Rodriguez-Molina, galectins are associated with antiphospholipid syndrome in patients with sys- temic lupus erythematosus. Glycobiology 23: 12–22. R. Gruebl-Barabas, and M. Cezanne for excellent technical assistance. 24. Iqbal, A. J., D. Cooper, A. Vugler, B. R. Gittens, A. Moore, and M. Perretti. S. Nuernberger and C. Albrecht are gratefully acknowledged for support. 2013. Endogenous galectin-1 exerts tonic inhibition on experimental arthritis. J. Immunol. 191: 171–177. 25. Kaltner, H., K. Seyrek, A. Heck, F. Sinowatz, and H.-J. Gabius. 2002. Galectin-1 and galectin-3 in fetal development of bovine respiratory and digestive tracts: Disclosures comparison of cell type-specific expression profiles and subcellular localization. The authors have no financial conflicts of interest. Cell Tissue Res. 307: 35–46. The Journal of Immunology 1921

26. Sarter, K., S. Andre´, H. Kaltner, M. Lensch, C. Schulze, V. Urbonaviciute, 43. Dange, M. C., N. Srinivasan, S. K. More, S. M. Bane, A. Upadhya, A. D. Ingle, G. Schett, M. Herrmann, and H.-J. Gabius. 2009. Detection and chromatographic R. P. Gude, R. Mukhopadhyaya, and R. D. Kalraiya. 2014. Galectin-3 expressed removal of lipopolysaccharide in preparations of multifunctional galectins. on different lung compartments promotes organ specific metastasis by facili- Biochem. Biophys. Res. Commun. 379: 155–159. tating arrest, extravasation and organ colonization via high affinity ligands on 27. Andre´, S., H. Sanchez-Ruderisch, H. Nakagawa, M. Buchholz, J. Kopitz, melanoma cells. Clin. Exp. Metastasis 31: 661–673. P. Forberich, W. Kemmner, C. Bo¨ck, K. Deguchi, K. M. Detjen, et al. 2007. € INK4a 44. Mauris, J., A. M. Woodward, Z. Cao, N. Panjwani, and P. Argueso. 2014. Mo- Tumor suppressor p16 : modulator of glycomic profile and galectin-1 ex- lecular basis for MMP9 induction and disruption of epithelial cell-cell contacts pression to increase susceptibility to carbohydrate-dependent induction of by galectin-3. J. Cell Sci. 127: 3141–3148. anoikis in pancreatic carcinoma cells. FEBS J. 274: 3233–3256. 45. Demers, M., T. Magnaldo, and Y. St-Pierre. 2005. A novel function for galectin- 28. Amano, M., H. Eriksson, J. C. Manning, K. M. Detjen, S. Andre´, S. Nishimura, 7: promoting tumorigenesis by up-regulating MMP-9 gene expression. Cancer J. Lehtio¨, and H.-J. Gabius. 2012. Tumour suppressor p16INK4a: anoikis- Res. 65: 5205–5210. favouring decrease in N/O-glycan/cell surface sialylation by down-regulation 46. Saussez, S., S. Cludts, A. Capouillez, G. Mortuaire, K. Smetana, Jr., H. Kaltner, of enzymes in sialic acid biosynthesis in tandem in a pancreatic carcinoma model. FEBS J. 279: 4062–4080. S. Andre´, X. Leroy, H.-J. Gabius, and C. Decaestecker. 2009. Identification of 29. Toegel, S., V. E. Plattner, S. Q. Wu, M. B. Goldring, C. Chiari, A. Kolb, matrix metalloproteinase-9 as an independent prognostic marker in laryngeal and F. M. Unger, S. Nehrer, F. Gabor, H. Viernstein, and M. Wirth. 2009. Lectin hypopharyngeal cancer with opposite correlations to adhesion/growth-regulatory binding patterns reflect the phenotypic status of in vitro chondrocyte models. galectins-1 and -7. Int. J. Oncol. 34: 433–439. In Vitro Cell. Dev. Biol. Anim. 45: 351–360. 47. Sana, T. R., M. J. Janatpour, M. Sathe, L. M. McEvoy, and T. K. McClanahan. 30. Toegel, S., S. Q. Wu, M. Otero, M. B. Goldring, P. Leelapornpisid, C. Chiari, 2005. Microarray analysis of primary endothelial cells challenged with different A. Kolb, F. M. Unger, R. Windhager, and H. Viernstein. 2012. Caesalpinia inflammatory and immune cytokines. Cytokine 29: 256–269. sappan extract inhibits IL1b-mediated overexpression of matrix metal- 48. Tian, B., D. E. Nowak, M. Jamaluddin, S. Wang, and A. R. Brasier. 2005. loproteinases in human chondrocytes. Genes Nutr. 7: 307–318. Identification of direct genomic targets downstream of the nuclear factor-kB 31. Bustin, S. A., J.-F. Beaulieu, J. Huggett, R. Jaggi, F. S. Kibenge, P. A. Olsvik, transcription factor mediating tumor necrosis factor signaling. J. Biol. Chem. L. C. Penning, and S. Toegel. 2010. MIQE pre´cis: practical implementation of 280: 17435–17448. minimum standard guidelines for fluorescence-based quantitative real-time PCR 49. Satelli, A., and U. S. Rao. 2011. Galectin-1 is silenced by promoter hyper- experiments. BMC Mol. Biol. 11: 74. methylation and its re-expression induces apoptosis in human colorectal cancer Downloaded from 32. Ramadoss, P., B. J. Abraham, L. Tsai, Y. Zhou, R. H. Costa-e-Sousa, F. Ye, cells. Cancer Lett. 301: 38–46. M. Bilban, K. Zhao, and A. N. Hollenberg. 2014. Novel mechanism of positive 50. Yun, S. P., S.-J. Lee, Y. H. Jung, and H. J. Han. 2014. Galectin-1 stimulates b b versus negative regulation by thyroid hormone receptor 1(TR 1) identified motility of human umbilical cord blood-derived mesenchymal stem cells by by genome-wide profiling of binding sites in mouse liver. J. Biol. Chem. 289: downregulation of smad2/3-dependent collagen 3/5 and upregulation of NF-kB- 1313–1328. dependent fibronectin/laminin 5 expression. Cell Death Dis. 5: e1049. 33. Bennani-Baiti, I. M., D. N. Aryee, J. Ban, I. Machado, M. Kauer, 51. Fitzner, B., H. Walzel, G. Sparmann, J. Emmrich, S. Liebe, and R. Jaster. 2005. K. Muhlbacher,€ G. Amann, A. Llombart-Bosch, and H. Kovar. 2011. Notch Galectin-1 is an inductor of pancreatic stellate cell activation. Cell. Signal. 17:

signalling is off and is uncoupled from HES1 expression in Ewing’s sarcoma. http://www.jimmunol.org/ J. Pathol. 225: 353–363. 1240–1247. 34. Bennani-Baiti, I. M., A. Cooper, E. R. Lawlor, M. Kauer, J. Ban, D. N. T. Aryee, 52. Masamune, A., M. Satoh, J. Hirabayashi, K. Kasai, K. Satoh, and T. Shimosegawa. and H. Kovar. 2010. Intercohort gene expression co-analysis reveals chemokine 2006. Galectin-1 induces chemokine production and proliferation in pancreatic stellate cells. Am. J. Physiol. Gastrointest. Liver Physiol. 290: G729–G736. receptors as prognostic indicators in Ewing’s sarcoma. Clin. Cancer Res. 16: 3769–3778. 53. Valach, J., Z. Fı´k, H. Strnad, M. Chovanec, J. Plza´k, Z. Cada, P. Szabo, 35. Kaltner, H., A. S. Raschta, J. C. Manning, and H.-J. Gabius. Copy-number J. Sa´chova´, M. Hroudova´, M. Urbanova´, et al. 2012. Smooth muscle actin- variation of functional galectin genes: studying animal galectin-7 (p53-induced expressing stromal fibroblasts in head and neck squamous cell carcinoma: in- gene 1 in man) and tandem-repeat-type galectins-4 and -9. Glycobiology 23: creased expression of galectin-1 and induction of poor prognosis factors. Int. J. 1152–1163. Cancer 131: 2499–2508. 36. Bennani-Baiti, B., and I. M. Bennani-Baiti. 2012. Gene symbol precision. Gene 54. Leijten, J. C. H., J. Emons, C. Sticht, S. van Gool, E. Decker, A. Uitterlinden, 491: 103–109. G. Rappold, A. Hofman, F. Rivadeneira, S. Scherjon, et al. 2012. Gremlin 1,

37. Toscano, M. A., L. Campagna, L. L. Molinero, J. P. Cerliani, D. O. Croci, frizzled-related protein, and Dkk-1 are key regulators of human articular carti- by guest on September 27, 2021 J. M. Ilarregui, M. B. Fuertes, I. M. Nojek, J. P. Fededa, N. W. Zwirner, et al. lage homeostasis. Arthritis Rheum. 64: 3302–3312. k 2011. Nuclear factor (NF)- B controls expression of the immunoregulatory 55. van der Kraan, P. M., and W. B. van den Berg. 2012. Chondrocyte hypertrophy glycan-binding protein galectin-1. Mol. Immunol. 48: 1940–1949. and osteoarthritis: role in initiation and progression of cartilage degeneration? 38. Sellam, J., and F. Berenbaum. 2010. The role of synovitis in pathophysiology Osteoarthritis Cartilage 20: 223–232. and clinical symptoms of osteoarthritis. Nat. Rev. Rheumatol. 6: 625–635. 56. Gabius, H.-J., H. Kaltner, J. Kopitz, and S. Andre´. 2015. The glycobiology of the 39. Troeberg, L., and H. Nagase. 2012. Proteases involved in cartilage matrix deg- CD system: a dictionary for translating marker designations into glycan/lectin radation in osteoarthritis. Biochim. Biophys. Acta 1824: 133–145. 40. Marcu, K. B., M. Otero, E. Olivotto, R. M. Borzi, and M. B. Goldring. 2010. NF- structure and function. Trends Biochem. Sci. 40: 360–376. k 57. Sanchez-Ruderisch, H., C. Fischer, K. M. Detjen, M. Welzel, A. Wimmel, B signaling: multiple angles to target OA. Curr. Drug Targets 11: 599–613. INK4a 41. Tetlow, L. C., D. J. Adlam, and D. E. Woolley. 2001. Matrix metalloproteinase J. C. Manning, S. Andre´, and H.-J. Gabius. 2010. Tumor suppressor p16 : and proinflammatory cytokine production by chondrocytes of human osteoar- downregulation of galectin-3, an endogenous competitor of the pro-anoikis ef- thritic cartilage: associations with degenerative changes. Arthritis Rheum. 44: fector galectin-1, in a pancreatic carcinoma model. FEBS J. 277: 3552–3563. 585–594. 58. Koh, H. S., C. Lee, K. S. Lee, C. S. Ham, R. H. Seong, S. S. Kim, and S. H. Jeon. 42. Wang, Y.-G., S.-J. Kim, J.-H. Baek, H.-W. Lee, S.-Y. Jeong, and K.-H. Chun. 2008. CD7 expression and galectin-1-induced apoptosis of immature thymocytes 2012. Galectin-3 increases the motility of mouse melanoma cells by regulating are directly regulated by NF-kB upon T-cell activation. Biochem. Biophys. Res. matrix metalloproteinase-1 expression. Exp. Mol. Med. 44: 387–393. Commun. 370: 149–153. Supplemental Table I. Additional Galectin-1-regulated genes determined by microarray analysis (see Figure 4 for details). The ratio between mRNA levels in Galectin-1-treated versus untreated chondrocytes as well as p-values, corrected for multiple hypothesis testing by the Benjamini-Hochberg method (BH), are given.

Lectins Ratio Symbol Entrez ID Gene name BH treated/untreated SELE 6401 E-selectin (CD62E) 3E-07 323.3 OLR1 4973 oxidized low density lipoprotein (lectin-like) receptor 1 2E-02 5.5 HSPC159 29094 galectin-related protein (GRP) 2E-02 3.2 KLRD1 3824 killer cell lectin-like receptor subfamily D, member 1 1E-02 2.1 CLEC4E 26253 C-type lectin domain family 4, member E 7E-03 1.9 KLRK1 22914 killer cell lectin-like receptor subfamily K, member 1 5E-02 1.6 LGALS3 3958 lectin, galactoside-binding, soluble, 3 (galectin-3) 3E-02 1.5 KLRC1 3821 killer cell lectin-like receptor subfamily C, member 1 4E-02 1.5 KLRG1 10219 killer cell lectin-like receptor subfamily G, member 1 3E-02 0.7 SELP 6403 P-selectin (granule membrane protein 140kDa, antigen CD62P) 1E-02 0.4 MLEC 9761 malectin 3E-02 0.4 CLEC3B 7123 C-type lectin domain family 3, member B 2E-03 0.3 CLEC3A 10143 C-type lectin domain family 3, member A 9E-04 0.2 COLEC12 81035 collectin sub-family member 12 2E-03 0.2

N)glycosylation Ratio Symbol Entrez ID Gene name BH treated/untreated B4GALT3 8703 UDP-Gal:βGlcNAc β1,4-galactosyltransferase, polypeptide 3 4E-03 1.9 B4GALT5 9334 UDP-Gal:βGlcNAc β1,4-galactosyltransferase, polypeptide 5 1E-02 1.9 ST3GAL4 6484 ST3 β-galactoside α2,3-sialyltransferase 4 4E-02 1.5 FUT10 84750 fucosyltransferase 10 (α1,3-fucosyltransferase) 5E-02 0.7 asparagine-linked glycosylation 10, α1,2-glucosyltransferase homolog (S. ALG10;ALG10B 144245;84920 pombe); asparagine-linked glycosylation 10, α1,2-glucosyltransferase homolog 4E-02 0.7 B (yeast) FUT11 170384 fucosyltransferase 11 (α1,3-fucosyltransferase) 5E-02 0.6 asparagine-linked glycosylation 9, α1,2-mannosyltransferase homolog (S. ALG9 79796 3E-02 0.4 cerevisiae) FUT8 2530 fucosyltransferase 8 (α1,6-fucosyltransferase) 1E-02 0.4 ALG10B 144245 asparagine-linked glycosylation 10, α1,2-glucosyltransferase homolog B (yeast) 2E-02 0.4 B4GALT2 8704 UDP-Gal:βGlcNAc β1,4- galactosyltransferase, polypeptide 2 2E-02 0.3 asparagine-linked glycosylation 6, α1,3-glucosyltransferase homolog (S. ALG6 29929 2E-02 0.3 cerevisiae) asparagine-linked glycosylation 9, α1,2-mannosyltransferase homolog (S. ALG9;FDXACB1 79796;91893 3E-02 0.3 cerevisiae); ferredoxin-fold anticodon binding domain containing 1 ST6GAL1 6480 ST6 β-galactosamide α2,6-sialyltranferase 1 2E-02 0.1

O)glycosylation Ratio Symbol Entrez ID Gene name BH treated/untreated ST3GAL1 6482 ST3 β-galactoside α2,3-sialyltransferase 1 2E-04 4.7 POMGNT1 55624 protein O-linked mannose β1,2-N-acetylglucosaminyltransferase 1E-02 1.8 B3GAT1 27087 β1,3-glucuronyltransferase 1 (glucuronosyltransferase P) 4E-02 1.8 B3GAT2 135152 β1,3-glucuronyltransferase 2 (glucuronosyltransferase S) 4E-02 0.6 B3GNT9 84752 UDP-GlcNAc:βGal β1,3-N-acetylglucosaminyltransferase 9 1E-02 0.4 B3GALTL 145173 β1,3-galactosyltransferase-like 3E-02 0.4 B3GNT1 11041 UDP-GlcNAc:βGal β1,3-N-acetylglucosaminyltransferase 1 2E-02 0.3

Glycolipids Ratio Symbol Entrez ID Gene name BH treated/untreated B3GNT5 84002 UDP-GlcNAc:βGal β1,3-N-acetylglucosaminyltransferase 5 2E-02 2.8 A4GALT 53947 α1,4-galactosyltransferase 2E-02 2.0 ST6 (α-N-acetyl-neuraminyl-2,3-β-galactosyl-1,3)-N-acetylgalactosaminide α-2,6- ST6GALNAC5 81849 3E-02 0.7 sialyltransferase 5 GBGT1 26301 globoside α1,3-N-acetylgalactosaminyltransferase 1 2E-02 0.7 ST6 (α-N-acetyl-neuraminyl-2,3-β-galactosyl-1,3)-N-acetylgalactosaminide α-2,6- ST6GALNAC6 30815 1E-02 0.6 sialyltransferase 6 B3GALNT1 8706 β1,3-N-acetylgalactosaminyltransferase 1 (globoside blood group) 1E-02 0.1

Mannosidases Ratio Symbol Entrez ID Gene name BH treated/untreated MAN1C1 57134 α-mannosidase, class 1C, member 1 3E-04 0.2 MANSC1 54682 MANSC domain containing 1 2E-02 0.2 MAN2B2 23324 α-mannosidase, class 2B, member 2 8E-03 0.4 MANBA 4126 α-mannosidase, β A, lysosomal 1E-02 0.4 MAN1B1 11253 α-mannosidase, class 1B, member 1 1E-02 0.4 MAN2A2 4122 α-mannosidase, class 2A, member 2 9E-03 0.4 MAN1A2 10905 α-mannosidase, class 1A, member 2 2E-02 0.5

Sialic5acid5(donor)5synthesis/processing Ratio Symbol Entrez ID Gene name BH treated/untreated NANS 54187 N-acetylneuraminic acid synthase 4E-02 1.8 CMAS 55907 cytidine monophosphate N-acetylneuraminic acid synthetase 2E-02 0.6 SIAE 54414 sialic acid acetylesterase 1E-02 0.4

Cartilage)5and5OA)related5genes Ratio Symbol Entrez ID Gene name BH treated/untreated COL7A1 1294 collagen, type VII, alpha 1 1E-03 4.73 COL5A3 50509 collagen, type V, alpha 3 5E-02 1.94 COL22A1 169044 collagen, type XXII, alpha 1 6E-03 1.83 COL21A1 81578 collagen, type XXI, alpha 1 2E-02 0.53 COL1A1 1277 collagen, type I, alpha 1 1E-02 0.51 PCOLCE 5118 procollagen C-endopeptidase enhancer 3E-03 0.47 COL12A1 1303 collagen, type XII, alpha 1 5E-03 0.46 COL6A1 1291 collagen, type VI, alpha 1 1E-02 0.42 COL5A1 1289 collagen, type V, alpha 1 2E-02 0.40 COL8A1 1295 collagen, type VIII, alpha 1 1E-02 0.38 PCOLCE2 26577 procollagen C-endopeptidase enhancer 2 2E-02 0.37 COL5A1 1289 collagen, type V, alpha 1 1E-02 0.35 COL1A2 1278 collagen, type I, alpha 2 5E-03 0.33 COL5A1 1289 collagen, type V, alpha 1 1E-02 0.33 COL5A2 1290 collagen, type V, alpha 2 2E-02 0.29 COL6A3 1293 collagen, type VI, alpha 3 2E-02 0.25 COL15A1 1306 collagen, type XV, alpha 1 2E-02 0.24 COL8A2 1296 collagen, type VIII, alpha 2 3E-03 0.23 COL12A1 1303 Collagen, type XII, alpha 1 2E-03 0.19 COL10A1 1300 collagen, type X, alpha 1 2E-02 0.19 COL14A1 7373 collagen, type XIV, alpha 1 1E-02 0.18 COL1A2 1278 collagen, type I, alpha 2 8E-03 0.17 COL11A1 1301 collagen, type XI, alpha 1 8E-03 0.15

ACAN 176 aggrecan 5E-02 0.54

MMP12 4321 matrix metallopeptidase 12 (macrophage elastase) 7E-05 15.85 MMP10 4319 matrix metallopeptidase 10 (stromelysin 2) 5E-04 10.61 matrix metallopeptidase 9 (gelatinase B, 92kDa gelatinase, 92kDa type IV MMP9 4318 3E-02 3.72 collagenase) MMP19 4327 matrix metallopeptidase 19 3E-02 2.41 MMP25 64386 matrix metallopeptidase 25 4E-02 2.11 MMP7 4316 matrix metallopeptidase 7 (matrilysin, uterine) 5E-02 1.45

FRZB 2487 frizzled-related protein 1E-02 0.30 WNT5A 7474 wingless-type MMTV integration site family, member 5A 2E-04 7.35 WNT5B 81029 wingless-type MMTV integration site family, member 5B 2E-02 0.54 WISP2 8839 WNT1 inducible signaling pathway protein 2 4E-03 0.23

nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor, NFKBIA 4792 2E-04 21.85 alpha nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor, NFKBIE 4794 2E-03 11.79 epsilon NFKB1 4790 nuclear factor of kappa light polypeptide gene enhancer in B-cells 1 2E-03 11.45 NFKB2 4791 nuclear factor of kappa light polypeptide gene enhancer in B-cells 2 (p49/p100) 9E-03 7.90 NFKBIZ 64332 nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor, zeta 5E-04 5.80 NFKBIB 4793 nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor, beta 2E-02 5.06 RELA 5970 v-rel reticuloendotheliosis viral oncogene homolog A (avian) 5E-04 3.64

LAMB3 3914 laminin, beta 3 3E-04 23.61 LAMC2 3918 laminin, gamma 2 3E-02 3.48 LAMA3 3909 laminin, alpha 3 3E-02 2.36 LAMB2 3913 laminin, beta 2 (laminin S) 3E-02 0.64 LAMC1 3915 laminin, gamma 1 (formerly LAMB2) 2E-02 0.50 LAMA4 3910 laminin, alpha 4 5E-03 0.40 LAMB1 3912 laminin, beta 1 5E-03 0.17

FSD1L 83856 fibronectin type III and SPRY domain containing 1-like 6E-03 8.91 FNDC3B 64778 fibronectin type III domain containing 3B 6E-04 2.84

DCN 1634 decorin 2E-02 0.21

BMP2 650 bone morphogenetic protein 2 2E-03 5.72 BMP6 654 bone morphogenetic protein 6 2E-03 4.32 BMPR1B 658 bone morphogenetic protein receptor, type IB 4E-04 3.40 BMP5 653 bone morphogenetic protein 5 1E-02 2.90 BMP1 649 bone morphogenetic protein 1 8E-03 1.79 BMP8A 353500 bone morphogenetic protein 8a 2E-02 1.66 BMP7 655 bone morphogenetic protein 7 2E-02 1.57 BMPR1A 657 bone morphogenetic protein receptor, type IA 2E-02 0.24 BMP4 652 bone morphogenetic protein 4 5E-03 0.16

TGFA 7039 transforming growth factor, alpha 3E-02 1.64 TGFB3 7043 transforming growth factor, beta 3 3E-02 0.53 TGFB2 7042 transforming growth factor, beta 2 1E-02 0.46

TNF 7124 tumor necrosis factor 7E-03 23.56

ADAMTS4 9507 ADAM metallopeptidase with thrombospondin type 1 motif, 4 4E-03 6.37 ADAMTS5 11096 ADAM metallopeptidase with thrombospondin type 1 motif, 5 1E-02 0.19

GNL3L 54552 guanine nucleotide binding protein-like 3 (nucleolar)-like 2E-02 0.62 GDF5 8200 growth differentiation factor 5 1E-04 0.16

TLR2 7097 toll-like receptor 2 3E-02 5.94

IGFBP6 3489 insulin-like growth factor binding protein 6 3E-02 0.61 IGF1R 3480 insulin-like growth factor 1 receptor 3E-02 0.37 IGFBP5 3488 insulin-like growth factor binding protein 5 1E-02 0.29 IGF2;INS-IGF2 3481;723961 insulin-like growth factor 2 (somatomedin A) /// INS-IGF2 readthrough transcript 4E-02 0.28

ITGAM 3684 integrin, alpha M (complement component 3 receptor 3 subunit) 1E-02 8.4 ILKAP 80895 integrin-linked kinase-associated serine/threonine phosphatase 2E-02 2.6 ITGA1 3672 integrin, alpha 1 1E-02 2.0 ITGA5 3678 integrin, alpha 5 (fibronectin receptor, alpha polypeptide) 1E-02 1.8 CIB2 10518 calcium and integrin binding family member 2 2E-02 0.6 ITGA11 22801 integrin, alpha 11 2E-02 0.5 integrin, beta 1 (fibronectin receptor, beta polypeptide, antigen CD29 includes ITGB1 3688 5E-02 0.5 MDF2, MSK12) ITGA3 3675 integrin, alpha 3 (antigen CD49C, alpha 3 subunit of VLA-3 receptor) 9E-03 0.5 ITGB5 3693 integrin, beta 5 5E-03 0.4 ITFG1 81533 integrin alpha FG-GAP repeat containing 1 4E-02 0.4 integrin, alpha E (antigen CD103, human mucosal lymphocyte antigen 1; alpha ITGAE 3682 2E-02 0.4 polypeptide) ITGB1BP1 9270 integrin beta 1 binding protein 1 3E-02 0.4 ITGB3BP 23421 integrin beta 3 binding protein (beta3-endonexin) 8E-03 0.3 ITGA10 8515 integrin, alpha 10 1E-04 0.1 ITGBL1 9358 integrin, beta-like 1 (with EGF-like repeat domains) 1E-03 0.1

Supplemental Table II. Regulators of Galectin-1 expression in cells.

Effector Location Factor of Determination of activity Reference (bp)a regulation

AP-1 +1596 to 2-fold decrease deletion of AP-1-specific binding 1 +1602 site, (2nd intron) luciferase measurement in classical Hodgkin lymphoma cell line [cHL HD-MY-Z]

n. d.b correlation between expression of 2 AP-1 and galectin-1 in anaplastic large cell lymphoma cells [ALCL], immunohistochemistry

C/EBPα -48 to -42 up to 5-fold overexpression of C/EBPα, 3 increase luciferase measurement in human embryonic kidney cells [293T]

3-fold decrease mutation of the C/EBPα-specific 3 sequence TTGCAATàTTAAGGT, luciferase measurement in human embryonic kidney cells [293T]

HIF-1α -441 to 2-fold decrease mutation of HIF-1α binding sites, 4 -437/ luciferase measurement in human -427 to embryonic kidney cells [293T] -423

n. d. up to 5-fold detection of mRNA expression 4 increase after incubation of different colorectal adenocarcinoma cell lines [SW620, LS174T, SW1116 and RKO] in hypoxia (1% O2)

n. d. up to 8-fold overexpression of HIF-1α, 4 increase luciferase measurement in human embryonic kidney cells [293T]

n. d. up to 14-fold induction of HIF-1α in renal cell 5 increase carcinoma cell line [CAK-1] by addition of CoCl2, densitometry

n. d. about 2-fold human squamous cell carcinoma 6 increase cells [FaDu] stably transfected with β-hCG gene under the control of a hypoxia-inducible promoter were implanted in mice, which were kept under hypoxia (10% O2) for 2-3 days, plasma samples of mice were analyzed for galectin-1 by ELISA

NF-κB +1185 to n. d. binding of p50 to potential binding 7 +1197 site, immunoprecipitation (1st intron)

n. d. about 5-fold Incubation of peripheral blood 7 increase mononuclear cells (PBMC) with TNF (10 ng/ml), Western blot

n. d. about 10-fold overexpression of p65 in human 7 increase embryonic kidney cells [HEK293], Western blot

n. d. up to 90-fold inhibition of NF-κB by 7 decrease sulfasalazine in monocyte-depleted PBMCs, Western blot

n. d. up to 4-fold expression of superrepressor IκB-α 8 decrease in Kaposi’s sarcoma cells, luciferase measurement, qRT-PCR retinoic acid - 1578 to 3-fold increase cloning of a galectin-1 promoter 9 -1448 fragment upstream of a thymidine (mouse) kinase minimal promoter and incubation with retinoic acid (1 µM), CAT-assay in murine testicular embryonal carcinoma cells [F9]

n. d. n. d. incubation of murine testicular 9 embryonal carcinoma cells [F9] with retinoic acid (1 µM), Northern and Western blots

n. d. about 2-fold treatment of human breast 10 decrease carcinoma cells [MDA-MB 175], human promyelocytic leukemia cells [HL-60] and acute monocytic leukemia cells [THP-1] with retinoic acid (1 µM), Western blot

3-fold increase treatment of murine testicular 10 embryonal carcinoma cells [F9] with retinoic acid (1 µM), Western blot sodium - 62 to - 41 up to 40-fold incubation of PCC4.aza1R cells 11 butyrate/ (mouse) increase (embryonic carcinoma), transfected Sp1 with different promoter variants of the murine galectin-1 gene, with sodium butyrate (2 mM), CAT-assay

- 62 to - 41 n. d. gel shift assay (EMSA) with the 11 (mouse) promoter fragment -67 to -37 and the nuclear extract from PCC4.aza1R cells (embryonic carcinoma), supershift with anti-SP1 antibody

- 62 to - 41 n. d. gene interval -63 to +45 cloned 12 (mouse and upstream of the CAT gene was human) analysed in mouse fibroblasts [NIH3T3] by primer extension

-25 to -17 n. d. sequence analysis 13 (human)

-25 to -17 up to 8-fold deletion analysis of promoter 14 (mouse) decrease region in mouse fibroblast cells [NIH3T3], CAT-assay

n. d. up to 10-fold incubation of five human colorectal 15 increase adenocarcinoma cell lines [DLD1, MIP101, HT-29, LoVo, LS174T] with sodium butyrate (2 mM), Western blot

n. d. up to 6-fold treatment of nine human colorectal 16 increase adenocarcinoma cell lines [HCT15, KM12, LS180, TC7, CX1, HT-29, LS513, SW480, SW707] with sodium butyrate (1-2 mM), RT-PCR

n. d. up to 5-fold treatment of head and neck 17 increase squamous cell carcinoma cell lines [1483, 886LN, PCI4B, UMSCC1, TU138] with sodium butyrate (3 mM), Western blot

n. d. up to 19-fold human intestinal epithelial goblet 18 increase cells [HT29-Cl.16E] treated with sodium butyrate (2-5 mM), microarray

n. d. upregulation treatment of human colon 10 carcinoma [KM12P] with sodium butyrate (2 mM), Western blot

n. d. upregulation Treatment of human prostate 19 adenocarcinoma cells [LNCaP] with sodium butyrate (2-5 mM), densitometry

Anti-CD3 n. d. upregulation stimulation of peripheral blood 20 mononuclear cells (PBMC) with anti-CD3, Western blot, densitometry

Anti-CD3/anti- n. d. about 3-fold activation of murine regulatory T 21, 22 CD28 or anti- increase cells leads to increased galectin-1 CD3/IL-2 expression, Western blot

Anti- n. d. upregulation stimulation of peripheral blood 20 CD28/phorbol mononuclear cells (PBMC) with 12-myristate anti-CD28/PMA, 13-acetate Western blot, densitometry (PMA)

Dibutyryl n. d. 2-fold decrease treatment of murine neuroblastoma 10 cAMP cells [S20] with dibutyryl cAMP (1 mM), Western blot

ERBB2 n. d. 12-fold increase ERBB2 overexpression in HB4a 23 human mammary epithelial cell lines [C3.6, C5.2], microarray

Estrogen n. d. 10-fold increase rat embryo cells [Rat1a(ER- 24 ΔFosB)] treatment with estrogen (1 µM), Northern and Western blots

Glucocorticoids n. d. 10-fold increase treatment of human leukaemia 25 cells [CEM C7] with dexamethasone (1 µM), Northern blot

Minimal n. d. 2-fold increase MM-LDL stimulated human aortic 26 oxidized low- endothelial cells (HAECs), density Northern blot lipoprotein (MM-LDL) p16INK4a n. d. about 3-fold reconstitution of p16INK4a in human 27 increase pancreatic carcinoma cells [Capan- 1], quantitative determination of galectin-1 mRNA, Western blot

Thyroid n. d. up to 8-fold stimulation of rat thyroid cells 28 stimulationg increase [FRTL-5] with (TSH), increase of hormone (TSH) expression 36 h after treatment, gradual decline to basal levels by 72 h, Northern Blot

Tax n. d. up to 4-fold transfection of human embryonic 29 increase kidney cells [293T] cells with vector encoding for Tax, RT-PCR

TGF-ß1 n. d. up to 3-fold treatment of human IMR90 30 increase fibroblasts and murine NIH3T3 with TGF-ß1 (5 ng/ml), Western blot

n. d. up to 17-fold decrease by treatment of bone 31 decrease marrow-derived mast cells (BMMC) from BALB/c mice, with TGF-ß1 (1 ng/ml), 2-DE, FACS, Western blot, RT- PCR

n. d. up to 150-fold treatment of mammary 32 increase adenocarcinoma cells [LM3] with TGF-ß1 (0.4 - 40 ng/ml), Western blot

Valproic acid n. d. 2-fold increase treatment of mouse embryonic 33 carcinoma cells [P19] with valproic acid (1 µM), microarray

a +1 is the transcription start site, based on the NCBI mRNA sequence (NM_002305.3) (3, 4) b not determined

1 Juszczynski, P., Ouyang, J., Monti, S., Rodig, S. J., Takeyama, K., Abramson, J., Chen, W., Kutok, J. L., Rabinovich, G. A., and Shipp, M. A. (2007). The AP1-dependent secretion of galectin-1 by Reed-Sternberg cells fosters immune privilege in classical Hodgkin lymphoma. Proc. Natl. Acad. Sci. USA 104, 13134–9.

2 Rodig, S. J., Ouyang, J., Juszczynski, P., Currie, T., Law, K., Neuberg, D. S., Rabinovich, G. A., Shipp, M. A., and Kutok, J. L. (2008). AP1-dependent galectin-1 expression delineates classical Hodgkin and anaplastic large cell lymphomas from other lymphoid malignancies with shared molecular features. Clin. Cancer Res. 14, 3338-44.

3 Zhao, X. Y., Zhao, K. W., Jiang, Y., Zhao, M., and Chen, G. Q. (2011). Synergistic induction of galectin-1 by CCAAT/enhancer binding protein α and hypoxia-inducible factor 1α and its role in differentiation of acute myeloid leukemic cells. J. Biol. Chem. 286, 36808- 19.

4 Zhao, X. Y., Chen, T. T., Xia, L., Guo, M., Xu, Y., Yue, F., Jiang, Y., Chen, G. Q., and Zhao, K. W. (2010). Hypoxia inducible factor-1 mediates expression of galectin-1: the potential role in migration/invasion of colorectal cancer cells. Carcinogenesis 31, 1367-75.

5 White, N. M., Masui, O., Newsted, D., Scorilas, A., Romaschin, A. D., Bjarnason, G. A., Siu, K. W., and Yousef, G. M. (2014). Galectin-1 has potential prognostic significance and is implicated in clear cell renal cell carcinoma progression through the HIF/mTOR signaling axis. Br. J. Cancer 110, 1250-9.

6 Le, Q. T., Shi, G., Cao, H., Nelson, D. W., Wang, Y., Chen, E. Y., Zhao, S., Kong, C., Richardson, D., O'Byrne, K. J., Giaccia, A. J., and Koong, A. C. (2005). Galectin-1: a link between tumor hypoxia and tumor immune privilege. J. Clin. Oncol. 23, 8932-41.

7 Toscano, M. A., Campagna, L., Molinero, L. L., Cerliani, J. P., Croci, D. O., Ilarregui, J. M., Fuertes, M. B., Nojek, I. M., Fededa, J. P., Zwirner, N. W., Costas, M. A., and Rabinovich, G. A. (2011). Nuclear factor NF-κB controls expression of the immunoregulatory glycan-binding protein galectin-1. Mol. Immunol. 48, 1940-9.

8 Croci, D. O., Salatino, M., Rubinstein, N., Cerliani, J. P., Cavallin, L. E., Leung, H. J., Ouyang, J., Ilarregui, J. M., Toscano, M. A., Domaica, C. I., Croci, M. C., Shipp, M. A., Mesri, E. A., Albini, A., and Rabinovich, G. A. (2012). Disrupting galectin-1 interactions with N-glycans suppresses hypoxia-driven angiogenesis and tumorigenesis in Kaposi's sarcoma. J. Exp. Med. 209, 1985-2000.

9 Lu, Y., Lotan, D., and Lotan, R. (2000). Differential regulation of constitutive and retinoic acid-induced galectin-1 gene transcription in murine embryonal carcinoma and myoblastic cells. Biochim. Biophys. Acta 1491, 13-9.

10 Lotan, R., Lotan, D., and Carralero, D. M. (1989). Modulation of galactoside-binding lectins in tumor cells by differentiation-inducing agents. Cancer Lett. 48, 115-22.

11 Lu, Y., and Lotan, R. (1999). Transcriptional regulation by butyrate of mouse galectin-1 gene in embryonal carcinoma cells. Biochim. Biophys. Acta 1444, 85-91.

12 De Gregorio, E., Chiariotti, L., and Di Nocera, P.P. (2001). The overlap of Inr and TATA elements sets the use of alternative transcriptional start sites in the mouse galectin-1 gene promoter. Gene 268, 215-23.

13 Gitt, M. A., and Barondes, S. H. (1991). Genomic sequence and organization of two members of a human lectin gene family. Biochemistry 30, 82-9.

14 Salvatore, P., Contursi, C., Benvenuto, G., Bruni, C. B., and Chiariotti, L. (1995). Characterization and functional dissection of the galectin-1 gene promoter. FEBS Lett. 373, 159-63.

15 Ohannesian, D. W., Lotan, D., Lotan, R. (1994). Concomitant increases in galectin-1 and its glycoconjugate ligands (carcinoembryonic antigen, lamp-1, and lamp-2) in cultured human colon carcinoma cells by sodium butyrate. Cancer Res. 54, 5992-6000.

16 Katzenmaier, E. M., André, S., Kopitz, J., Gabius H.-J. (2014). Impact of sodium butyrate on the network of adhesion/growth-regulatory galectins in human colon cancer in vitro. Anticancer Res. 34, 5429-38.

17 Gillenwater, A., Xu, X. C., Estrov, Y., Sacks, P. G., Lotan, D., Lotan, R. (1998). Modulation of galectin-1 content in human head and neck squamous carcinoma cells by sodium butyrate. Int. J. Cancer 75, 217-24.

18 Gaudier, E., Forestier, L., Gouyer, V., Huet, G., Julien, R., Hoebler, C. (2004). Butyrate regulation of glycosylation-related gene expression: evidence for galectin-1 upregulation in human intestinal epithelial goblet cells. Biochem. Biophys. Res. Commun. 325, 1044-51.

19 Ellerhorst, J., Nguyen, T., Cooper, D. N., Lotan, D., Lotan, R. (1999). Differential expression of endogenous galectin-1 and galectin-3 in human prostate cancer cell lines and effects of overexpressing galectin-1 on cell phenotype. Int. J. Oncol. 14, 17-24.

20 Fuertes, M. B., Molinero, L. L., Toscano, M. A., Ilarregui, J. M., Rubinstein, N., Fainboim, L., Zwirner, N. W., Rabinovich, G. A. (2004). Regulated expression of galectin-1 during T-cell activation involves Lck and Fyn kinases and signaling through MEK1/ERK, p38 MAP kinase and p70S6 kinase. Mol. Cell. Biochem. 267, 177-85.

21 Garín, M. I., Chu, C. C., Golshayan, D., Cernuda-Morollón, E., Wait, R., Lechler, R.I. (2007). Galectin-1: a key effector of regulation mediated by CD4+CD25+ T cells. Blood 109, 2058-65.

22 Sugimoto, N., Oida, T., Hirota, K., Nakamura, K., Nomura, T., Uchiyama, T., Sakaguchi, S. (2006). Foxp3-dependent and -independent molecules specific for CD25+CD4+ natural regulatory T cells revealed by DNA microarray analysis. Int. Immunol. 18: 1197-209.

23 Mackay, A., Jones, C., Dexter, T., Silva, R. L., Bulmer, K., Jones, A., Simpson, P., Harris, R. A., Jat, P. S., Neville, A. M., Reis, L. F., Lakhani, S. R., and O'Hare, M. J. (2003). cDNA microarray analysis of genes associated with ERBB2 (HER2/neu) overexpression in human mammary luminal epithelial cells. Oncogene 22, 2680-8.

24 Tahara, K., Tsuchimoto, D., Tominaga, Y., Asoh, S., Ohta, S., Kitagawa, M., Horie, H., Kadoya, T., and Nakabeppu, Y. (2003). ΔFosB, but not FosB, induces delayed apoptosis independent of cell proliferation in the Rat1a embryo cell line. Cell Death Differ. 10, 496- 507.

25 Goldstone, S. D., and Lavin, M.F. (1991). Isolation of a cDNA clone, encoding a human beta-galactoside binding protein, overexpressed during glucocorticoid-induced cell death. Biochem. Biophys. Res. Commun. 178, 746-50.

26 Baum L. G., Seilhamer, J, J., Pang, M., Levinem W, B., Beynon, D., and Berliner, J. A. (1995). Synthesis of an endogeneous lectin, galectin-1, by human endothelial cells is up- regulated by endothelial cell activation. Glycoconj. J. 12, 63-8.

27 André, S., Sanchez-Ruderisch, H., Nakagawa, H., Buchholz, M., Kopitz, J., Forberich, P., Kemmner, W., Böck, C., Deguchi, K., Detjen, K. M., Wiedenmann, B., von Knebel Doeberitz, M., Gress, T. M., Nishimura, S., Rosewicz, S., and Gabius, H.-J. (2007). Tumor suppressor p16INK4a: modulator of glycomic profile and galectin-1 expression to increase susceptibility to carbohydrate-dependent induction of anoikis in pancreatic carcinoma cells. FEBS J. 274, 3233-56.

28 Chiariotti, L., Benvenuto, G., Salvatore, P., Veneziani, B. M., Villone, G., Fusco, A., Russo, T., Bruni, C. B. (1994). Expression of the soluble lectin L-14 gene is induced by TSH in thyroid cells and suppressed by retinoic acid in transformed neural cells. Biochem. Biophys. Res. Commun. 199, 540-6.

29 Gauthier, S., Pelletier, I., Ouellet, M., Vargas, A., Tremblay, M. J., Sato, S., Barbeau, B. (2008). Induction of galectin-1 expression by HTLV-I Tax and its impact on HTLV-I infectivity. Retrovirology 105. doi: 10.1186/1742-4690-5-105.

30 Lim, M. J., Ahn, J., Yi, J. Y., Kim, M. H., Son, A. R., Lee, S.L., Lim, D. S., Kim, S. S., Kang, M. A., Han, Y., and Song, J. Y. (2014). Induction of galectin-1 by TGF-β1 accelerates fibrosis through enhancing nuclear retention of Smad2. Exp. Cell Res. 326, 125-35.

31 Pemberton, A. D., Brown, J. K., Wright, S. H., Knight, P. A., and Miller, H. R. (2006). The proteome of mouse mucosal mast cell homologues: the role of transforming growth factor- β1. Proteomics 6, 623-31.

32 Daroqui, C. M., Ilarregui, J. M., Rubinstein, N., Salatino, M., Toscano, M. A., Vazquez, P., Bakin, A., Puricelli, L., Bal, de Kier Joffé E., and Rabinovich, G. A. (2007). Regulation of galectin-1 expression by transforming growth factor beta1 in metastatic mammary adenocarcinoma cells: implications for tumor-immune escape. Cancer Immunol. Immunother. 56, 491-9.

33 Kultima, K., Nyström, A. M., Scholz, B., Gustafson, A. L., Dencker, L., and Stigson, M. (2004). Valproic acid teratogenicity: a toxicogenomics approach. Environ. Health Perspect. 112, 1225-35.