Healthy Research Rewards
ResearchHub is incentivizing healthy research behavior. At this time, first authors of open access papers are eligible for rewards. Visit the publications tab to view your eligible publications.
Got it
FL
Fu‐Tong Liu
Author with expertise in Role of Galectins in Immunity and Disease
Achievements
Cited Author
Open Access Advocate
Key Stats
Upvotes received:
0
Publications:
13
(92% Open Access)
Cited by:
4,335
h-index:
87
/
i10-index:
252
Reputation
Biology
< 1%
Chemistry
< 1%
Economics
< 1%
Show more
How is this calculated?
Publications
0

Galectin-3 Expression and Secretion Links Macrophages to the Promotion of Renal Fibrosis

Neil Henderson et al.Jan 18, 2008
Macrophages have been proposed as a key cell type in the pathogenesis of renal fibrosis; however, the mechanism by which macrophages drive fibrosis is still unclear. We show that expression of galectin-3, a β-galactoside-binding lectin, is up-regulated in a mouse model of progressive renal fibrosis (unilateral ureteric obstruction, UUO), and absence of galectin-3 protects against renal myofibroblast accumulation/activation and fibrosis. Furthermore, specific depletion of macrophages using CD11b-DTR mice reduces fibrosis severity after UUO demonstrating that macrophages are key cells in the pathogenesis of renal fibrosis. Disruption of the galectin-3 gene does not affect macrophage recruitment after UUO, or macrophage proinflammatory cytokine profiles in response to interferon-γ/lipopolysaccharide. In addition, absence of galectin-3 does not affect transforming growth factor-β expression or Smad 2/3 phosphorylation in obstructed kidneys. Adoptive transfer of wild-type but not galectin-3−/− macrophages did, however, restore the fibrotic phenotype in galectin-3−/− mice. Cross-over experiments using wild-type and galectin-3−/− macrophage supernatants and renal fibroblasts confirmed that secretion of galectin-3 by macrophages is critical in the activation of renal fibroblasts to a profibrotic phenotype. Therefore, we demonstrate for the first time that galectin-3 expression and secretion by macrophages is a major mechanism linking macrophages to the promotion of renal fibrosis. Macrophages have been proposed as a key cell type in the pathogenesis of renal fibrosis; however, the mechanism by which macrophages drive fibrosis is still unclear. We show that expression of galectin-3, a β-galactoside-binding lectin, is up-regulated in a mouse model of progressive renal fibrosis (unilateral ureteric obstruction, UUO), and absence of galectin-3 protects against renal myofibroblast accumulation/activation and fibrosis. Furthermore, specific depletion of macrophages using CD11b-DTR mice reduces fibrosis severity after UUO demonstrating that macrophages are key cells in the pathogenesis of renal fibrosis. Disruption of the galectin-3 gene does not affect macrophage recruitment after UUO, or macrophage proinflammatory cytokine profiles in response to interferon-γ/lipopolysaccharide. In addition, absence of galectin-3 does not affect transforming growth factor-β expression or Smad 2/3 phosphorylation in obstructed kidneys. Adoptive transfer of wild-type but not galectin-3−/− macrophages did, however, restore the fibrotic phenotype in galectin-3−/− mice. Cross-over experiments using wild-type and galectin-3−/− macrophage supernatants and renal fibroblasts confirmed that secretion of galectin-3 by macrophages is critical in the activation of renal fibroblasts to a profibrotic phenotype. Therefore, we demonstrate for the first time that galectin-3 expression and secretion by macrophages is a major mechanism linking macrophages to the promotion of renal fibrosis. Macrophages are pleiotropic inflammatory cells prominent in both acute and chronic inflammation. In the chronic inflammatory milieu, macrophages interact with other cell types including cells of mesenchymal origin (fibroblasts) that transdifferentiate into matrix-secreting myofibroblasts, with resultant scar formation and disruption of tissue architecture. Advanced renal fibrosis with kidney failure is a major health care burden worldwide,1El Nahas AM Bello AK Chronic kidney disease: the global challenge.Lancet. 2005; 365: 331-340Abstract Full Text Full Text PDF PubMed Scopus (900) Google Scholar and long-term dialysis or transplantation are the only therapeutic options currently available.2Sayegh MH Carpenter CB Transplantation 50 years later—progress, challenges, and promises.N Engl J Med. 2004; 351: 2761-2766Crossref PubMed Scopus (334) Google Scholar Therefore increasing our understanding of the complex interplay between chronic inflammation and progressive fibrosis is a critical step toward the design of rational new treatments. The importance of macrophages in the wound-healing response has been known for some time. In the 1970s studies on skin wound healing by Leibovich and Ross3Leibovich SJ Ross R The role of the macrophage in wound repair. A study with hydrocortisone and antimacrophage serum.Am J Pathol. 1975; 78: 71-100PubMed Google Scholar, 4Leibovich SJ Ross R A macrophage-dependent factor that stimulates the proliferation of fibroblasts in vitro.Am J Pathol. 1976; 84: 501-514PubMed Google Scholar demonstrated that macrophage depletion (with hydrocortisone and anti-macrophage serum) resulted in the delayed appearance of fibroblasts, and their subsequent rate of proliferation was lower than that of controls. More recently, we have shown that selective depletion of macrophages in a model of hepatic inflammation significantly attenuates liver fibrosis.5Duffield JS Forbes SJ Constandinou CM Clay S Partolina M Vuthoori S Wu S Lang R Iredale JP Selective depletion of macrophages reveals distinct, opposing roles during liver injury and repair.J Clin Invest. 2005; 115: 56-65Crossref PubMed Scopus (1282) Google Scholar In the kidney there is a striking correlation between tubulointerstitial macrophage infiltration and the severity of fibrosis in human biopsies and the subsequent development and progression of chronic renal failure to end-stage renal failure requiring dialysis.6Bohle A Muller GA Wehrmann M Mackensen-Haen S Xiao JC Pathogenesis of chronic renal failure in the primary glomerulopathies, renal vasculopathies, and chronic interstitial nephritides.Kidney Int. 1996; 54: S2-S9Google Scholar, 7Becker GJ Hewitson TD The role of tubulointerstitial injury in chronic renal failure.Curr Opin Nephrol Hypertens. 2000; 9: 133-138Crossref PubMed Scopus (134) Google Scholar Experimental hydronephrosis induced by unilateral ureteric obstruction (UUO) is a clinically relevant animal model because it mimics congenital obstructive nephropathy (the major cause of end-stage renal disease in children8Roth KS Koo HP Spottswood SE Chan JC Obstructive uropathy: an important cause of chronic renal failure in children.Clin Pediatr. 2002; 45: 309-314Crossref Scopus (41) Google Scholar), with progression through the different stages of obstructive nephropathy leading to tubulointerstitial fibrosis.9Bascands JL Schanstra JP Obstructive nephropathy: insights from genetically engineered animals.Kidney Int. 2005; 68: 925-937Crossref PubMed Scopus (193) Google Scholar Experimental hydronephrosis secondary to UUO is neutrophil- and lymphocyte-independent and is characterized by a marked tubulointerstitial macrophage infiltrate,10Schreiner GF Harris KP Purkerson ML Klahr S Immunological aspects of acute ureteral obstruction: immune cell infiltrate in the kidney.Kidney Int. 1988; 34: 487-493Crossref PubMed Scopus (228) Google Scholar, 11Diamond JR Macrophages and progressive renal disease in experimental hydronephrosis.Am J Kidney Dis. 1995; 26: 133-140Abstract Full Text PDF PubMed Scopus (86) Google Scholar interstitial myofibroblast and tubular epithelial cell proliferation, and progressive scarring with deposition of extracellular matrix early in the course of the disease.12Hughes J Johnson RJ Role of Fas (CD95) in tubulointerstitial disease induced by unilateral ureteric obstruction.Am J Physiol. 1999; 277: F26-F32PubMed Google Scholar, 13Diamond JR van Goor H Ding G Engelmyer E Myofibroblasts in experimental hydronephrosis.Am J Pathol. 1995; 146: 121-129PubMed Google Scholar Furthermore, the inhibition of tubulointerstitial macrophage recruitment reduces the extent and severity of renal fibrosis14Ophascharoensuk V Giachelli CM Gordon K Hughes J Pichler R Brown P Liaw L Schmidt R Shankland SJ Alpers CE Couser WG Johnson RJ Obstructive uropathy in the mouse: role of osteopontin in interstitial fibrosis and apoptosis.Kidney Int. 1999; 56: 571-580Crossref PubMed Scopus (253) Google Scholar, 15Lange-Sperandio B Cachat F Thornhill BA Chevalier RL Selectins mediate macrophage infiltration in obstructive nephropathy in newborn mice.Kidney Int. 2002; 61: 516-524Crossref PubMed Scopus (86) Google Scholar, 16Anders HJ Vielhauer V Frink M Linde Y Cohen CD Blattner SM Kretzler M Strutz F Mack M Grone HJ Onuffer J Horuk R Nelson PJ Schlondorff D A chemokine receptor CCR-1 antagonist reduces renal fibrosis after unilateral ureter ligation.J Clin Invest. 2002; 109: 251-259Crossref PubMed Scopus (209) Google Scholar, 17Eis V Luckow B Vielhauer V Siveke JT Linde Y Segerer S Perez De Lema G Cohen CD Kretzler M Mack M Horuk R Murphy PM Gao JL Hudkins KL Alpers CE Grone HJ Schlondorff D Anders HJ Chemokine receptor CCR1 but not CCR5 mediates leukocyte recruitment and subsequent renal fibrosis after unilateral ureteral obstruction.J Am Soc Nephrol. 2004; 15: 337-347Crossref PubMed Scopus (113) Google Scholar, 18Kitagawa K Wada T Furuichi K Hashimoto H Ishiwata Y Asano M Takeya M Kuziel WA Matsushima K Mukaida N Yokoyama H Blockade of CCR2 ameliorates progressive fibrosis in kidney.Am J Pathol. 2004; 165: 237-246Abstract Full Text Full Text PDF PubMed Scopus (262) Google Scholar demonstrating that macrophages play a major role in driving fibrosis after UUO. Galectin-3 is a β-galactoside-binding animal lectin of ∼30 kDa19Ho MK Springer TA Mac-2, a novel 32,000 Mr mouse macrophage subpopulation-specific antigen defined by monoclonal antibodies.J Immunol. 1982; 128: 1221-1228PubMed Google Scholar that is highly expressed and secreted by macrophages.20Sato S Hughes RC Regulation of secretion and surface expression of Mac-2, a galactoside-binding protein of macrophages.J Biol Chem. 1994; 269: 4424-4430Abstract Full Text PDF PubMed Google Scholar, 21Liu FT Hsu DK Zuberi RI Kuwabara I Chi EY Henderson Jr, WR Expression and function of galectin-3, a beta-galactoside-binding lectin, in human monocytes and macrophages.Am J Pathol. 1995; 147: 1016-1028PubMed Google Scholar It is up-regulated when monocytes differentiate into macrophages21Liu FT Hsu DK Zuberi RI Kuwabara I Chi EY Henderson Jr, WR Expression and function of galectin-3, a beta-galactoside-binding lectin, in human monocytes and macrophages.Am J Pathol. 1995; 147: 1016-1028PubMed Google Scholar and down-regulated when macrophages differentiate into dendritic cells.22Dietz AB Bulur PA Knutson GJ Matasic R Vuk-Pavlovic S Maturation of human monocyte-derived dendritic cells studied by microarray hybridization.Biochem Biophys Res Commun. 2000; 275: 731-738Crossref PubMed Scopus (127) Google Scholar Furthermore, galectin-3 is a potent mitogen for fibroblasts in vitro,23Moutsatsos IK Wade M Schindler M Wang JL Endogenous lectins from cultured cells: nuclear localization of carbohydrate-binding protein 35 in proliferating 3T3 fibroblasts.Proc Natl Acad Sci USA. 1987; 84: 6452-6456Crossref PubMed Scopus (242) Google Scholar, 24Inohara H Akahani S Raz A Galectin-3 stimulates cell proliferation.Exp Cell Res. 1998; 245: 294-302Crossref PubMed Scopus (171) Google Scholar, 25Sasaki S Bao Q Hughes RC Galectin-3 modulates rat mesangial cell proliferation and matrix synthesis during experimental glomerulonephritis induced by anti-Thy1.1 antibodies.J Pathol. 1999; 187: 481-489Crossref PubMed Scopus (79) Google Scholar, 26Maeda N Kawada N Seki S Arakawa T Ikeda K Iwao H Okuyama H Hirabayashi J Kasai K Yoshizato K Stimulation of proliferation of rat hepatic stellate cells by galectin-1 and galectin-3 through different intracellular signaling pathways.J Biol Chem. 2003; 278: 18938-18944Crossref PubMed Scopus (134) Google Scholar and our previous work has demonstrated that galectin-3 regulates myofibroblast activation and hepatic fibrosis in vivo.27Henderson NC Mackinnon AC Farnworth SL Poirier F Russo FP Iredale JP Haslett C Simpson KJ Sethi T Galectin-3 regulates myofibroblast activation and hepatic fibrosis.Proc Natl Acad Sci USA. 2006; 103: 5060-5065Crossref PubMed Scopus (473) Google Scholar We hypothesized that the major tissue source of galectin-3 driving fibrosis is macrophage-derived, and using a model of hydronephrosis we set out to define whether macrophage-derived galectin-3 is a major mechanism linking macrophages to the promotion of renal myofibroblast activation and fibrosis. Tissue culture reagents were purchased from Life Technologies (Paisley, UK). Tissue culture plastics were obtained from Costar (Loughborough, UK) and Falcon (Runcorn, UK). Cytokines and recombinant mouse galectin-3 were purchased from R&D Systems (Abingdon, UK) and Peprotech EC Ltd. (London, UK). The galectin-3 inhibitor bis- [3-deoxy-3-(3-methoxybenzamido)-β-d-galactopyranosyl]-sulfane was provided by U. Nilsson and H. Leffler, University of Lund, Sweden.28Cumpstey I Sundin A Leffler H Nilsson UJ C2-symmetrical thiodigalactoside bis-benzamido derivatives as high-affinity inhibitors of galectin-3: efficient lectin inhibition through double arginine-arene interactions.Angew Chem Int Ed Engl. 2005; 44: 5110-5112Crossref PubMed Scopus (121) Google Scholar All other reagents were from Sigma-Aldrich Company Ltd. (Poole, UK) unless otherwise stated. Mice were maintained in 12-hour light/12-hour dark cycles with free access to food and water. All procedures were performed in accordance with Home Office guidelines [Animals (Scientific Procedures) Act 1986]. Generation of galectin-3−/− mice by gene-targeting technology has been described previously.29Colnot C Ripoche MA Milon G Montagutelli X Crocker PR Poirier F Maintenance of granulocyte numbers during acute peritonitis is defective in galectin-3-null mutant mice.Immunology. 1998; 94: 290-296Crossref PubMed Scopus (155) Google Scholar As control, age- and sex-matched wild-type littermate mice were used. CD11b-DTR mice were generated and characterized as previously described.5Duffield JS Forbes SJ Constandinou CM Clay S Partolina M Vuthoori S Wu S Lang R Iredale JP Selective depletion of macrophages reveals distinct, opposing roles during liver injury and repair.J Clin Invest. 2005; 115: 56-65Crossref PubMed Scopus (1282) Google Scholar Strain-matched controls (FVB/N) were purchased from B and K Ltd. (Hull, UK). All in vivo studies had six mice in each experimental group. UUO was performed by ligation of the left ureter as described previously.12Hughes J Johnson RJ Role of Fas (CD95) in tubulointerstitial disease induced by unilateral ureteric obstruction.Am J Physiol. 1999; 277: F26-F32PubMed Google Scholar Sham-operated control mice underwent an identical surgical procedure to the UUO mice except ligation of the ureter was not performed. Kidneys were harvested at days 3, 7, and 14 after UUO. For macrophage ablation CD11b-DTR mice and strain-matched control FVB/N mice (six mice per group) received three intravenous injections of either diphtheria toxin (DT) (25 ng/g body weight) or phosphate-buffered saline after UUO on days 4, 5, and 6. Kidneys were harvested at day 7 and quartered, and samples were then fixed in either methyl Carnoy's reagent (60% methanol, 30% chloroform, 10% glacial acetic acid) for assessment of macrophage infiltration or neutral buffered formalin for immunohistochemistry. Samples were also snap-frozen in liquid nitrogen for real-time reverse transcriptase-polymerase chain reaction (RT-PCR) analysis. Bone marrow-derived macrophages (BMDMs) were prepared from wild-type (WT) and galectin-3−/− mice by maturing bone marrow cells in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum and 20% L929 conditioned media for 7 to 9 days as described previously.30Duffield JS Erwig LP Wei X Liew FY Rees AJ Savill JS Activated macrophages direct apoptosis and suppress mitosis of mesangial cells.J Immunol. 2000; 164: 2110-2119PubMed Google Scholar Mature BMDMs (5 × 105/well) were added to the wells of 24-well plates (or plated onto glass coverslips for immunofluorescence). After 3 hours the wells were washed to remove nonadherent cells. Wells were treated with lipopolysaccharide (LPS) (100 ng/ml) and murine interferon-γ (IFN-γ) (100 U/ml) in serum-free media. After 24 hours of incubation, the supernatants were harvested and clarified by centrifugation at 10,000 × g for 5 minutes and frozen at −80°C. In vivo-derived peritoneal macrophages were obtained from peritoneal lavage and separated by adhesion onto tissue culture plastic. Cytokine release in macrophage supernatants was determined by cytometric bead array, mouse inflammation kit (BD Biosciences, Oxford, UK). Paraffin-embedded sections of mouse tissue were processed for immunohistochemistry as described previously,5Duffield JS Forbes SJ Constandinou CM Clay S Partolina M Vuthoori S Wu S Lang R Iredale JP Selective depletion of macrophages reveals distinct, opposing roles during liver injury and repair.J Clin Invest. 2005; 115: 56-65Crossref PubMed Scopus (1282) Google Scholar and the following primary antibodies were used: mouse monoclonal anti-α-SMA clone 1A4 (Sigma), rat monoclonal anti-mouse galectin-3 clone 8942F (Cedarlane, Ontario, Canada), and rat anti-mouse F4/80 clone CI:A3-1 (Serotec, Oxford, UK). Methyl Carnoy's-fixed paraffin-embedded sections (4 μm) were used to assess macrophage infiltration (F4/80-positive cells), and sections were visualized and quantified as previously described.15Lange-Sperandio B Cachat F Thornhill BA Chevalier RL Selectins mediate macrophage infiltration in obstructive nephropathy in newborn mice.Kidney Int. 2002; 61: 516-524Crossref PubMed Scopus (86) Google Scholar The detection of renal fibrocytes was performed by immunofluorescence using specific antibodies against CD34 (Santa Cruz Biotechnology, Santa Cruz, CA), CD45 (Santa Cruz Biotechnology), and type I collagen polyclonal antibody (Chemicon International, Temecula, CA). CD34, CD45, and collagen I immunofluorescence staining of formalin-fixed sections was performed using species-specific Alexa-568- and Alexa-488-conjugated secondary antibodies and fluorescence microscopy (Carl Zeiss Ltd., Welwyn Garden City, UK). BMDMs plated on glass coverslips were washed and fixed in 3% paraformaldehyde and subjected to indirect immunofluorescence with anti-galectin-3-fluorescein isothiocyanate antibody and then anti-fluorescein isothiocyanate-Alexa-488 antibody. Nuclei were labeled with 4,6-diamidino-2-phenylindole. Renal fibrosis was visualized microscopically and quantified with the use of a picrosirius red stain as described previously.5Duffield JS Forbes SJ Constandinou CM Clay S Partolina M Vuthoori S Wu S Lang R Iredale JP Selective depletion of macrophages reveals distinct, opposing roles during liver injury and repair.J Clin Invest. 2005; 115: 56-65Crossref PubMed Scopus (1282) Google Scholar Digital image analysis was used to quantitate the amount of red-stained collagen fibers. Morphometric measurements of 10-μm sections stained with picrosirius red were made using OpenLab software (Improvision, Coventry, UK). Twelve nonoverlapping fields at ×400 magnification from each section (captured with a Leica DMLB microscope, Leica DC300 camera, and Leica image manager; Leica, Milton Keynes, UK) were analyzed in a blinded manner. Each captured field was analyzed by separation into red, green, and blue (RGB) filters, and the red area was mathematically divided by the red, green, and blue (RGB) area and multiplied by 100%. This represents the percentage area staining positively for collagen fibers, providing a quantitative value on a continuous scale. Western blot analysis was undertaken using the following primary antibodies: mouse monoclonal anti-α-smooth muscle actin (SMA) antibody clone 1A4 (Sigma), mouse monoclonal anti-galectin-3 antibody clone A3A12 (Alexis Biochemicals, Nottingham, UK), rabbit polyclonal anti-phospho-Smad2 and anti-phospho-Smad3 (Biosource, Paisley, UK), and goat polyclonal total Smad2/3 antibody (Santa Cruz Biotechnology). Total RNA from whole kidney was reverse-transcribed into cDNA using random hexamers (Applied Biosystems, Warrington, UK). Mouse primers and probes were as follows: galectin-3: forward 5′-TTGAAGCTGACCACTTCAAGGTT-3′, reverse 5′-AGGTTCTTCATCCGATGGTTGT-3′, probe FAM 5′-CGGTCAACGATGCTCACCTACTGCA-3′ TAMRA; α-SMA: forward 5′-TCAGCGCCTCCAGTTCCT-3′, reverse 5′-AAAAAAAACCACGAGTAACAAATCAA-3′; probe FAM 5′-TCCAAA TCATTCCTGCCCA-3′ TAMRA; procollagen(I): forward 5′-TTCACCTACAGCACGCTTGTG-3′, reverse 5′-GATGACTGTCTTGCCCCAAGTT-3′, probe FAM 5′-ATGGCTGCACGAGTCACA-3′ TAMRA; transforming growth factor (TGF)-β: forward 5′-CACCGGAGAGCCCTGGATA-3′, reverse 5′-TGTACAGCTGCCGCACACA-3′, probe FAM 5′-CAACTATTGCTTCAGCTCCACAGAGAAGAACTG-3′ TAMRA. 18S rRNA TaqMan primer probe mix was purchased from Applied Biosystems. Mature BMDMs (day 7) were prelabeled with fluorescent Cell-Tracker Orange as per the manufacturers instructions (Molecular Probes, Eugene, OR). Galectin-3−/− mice underwent UUO surgery at day 0 and received 5 × 106 WT or galectin-3−/− BMDMs at days 1, 3, and 5 intravenously. Tissues were harvested on day 7 including both UUO and contralateral kidney, liver, spleen, and lung. Primary cultures of renal fibroblasts were isolated by trypsin digestion (0.25% for 2 hours at 37°C) of minced normal mouse kidneys, and digests were passed through a 20-μm cell strainer (Becton Dickinson, Oxford, UK) to remove glomeruli. Cells were cultured undisturbed in Dulbecco's modified Eagle's medium containing 15% fetal calf serum for 8 days until cells were confluent. Renal fibroblasts were used at passage 2. Mature BMDMs (1 × 106 cells) were plated in six-well tissue culture dishes and incubated in 1 ml of serum-free media for 48 hours. Conditioned media from BMDMs (0.5 ml) or control media were added to 0.5 ml of renal fibroblasts (4 × 104 cells). Cells were incubated for a further 48 hours (final fetal calf serum concentration, 7.5%) before lysis and Western analysis. Results are presented as means ± SEM. Significance of the differences between means was assessed using one-way analysis of variance or two-tailed Student's t-test. Values of P < 0.05 were considered significant. Unless stated otherwise, studies were performed on three to six independent occasions. Galectin-3 expression was analyzed in a well established experimental model of progressive renal fibrosis (UUO). Galectin-3 expression was markedly increased in the renal interstitium and tubular epithelium after UUO compared with the control sham-operated group (Figure 1, a and b). This increase in galectin-3 expression was confirmed by real-time PCR of whole kidney tissue (Figure 1c, P < 0.001). The significance of the induction of galectin-3 expression in the development of renal fibrosis was examined using the UUO model of progressive renal scarring. Renal collagen deposition was stained with picrosirius red (Figure 2, a and b) and quantified using digital image analysis. Significantly reduced collagen deposition was observed in the galectin-3−/− mice compared with WT (P < 0.05, Figure 2c). Furthermore transcripts for procollagen (I) were also reduced in the galectin-3−/− group compared with WT animals (Figure 2d, P < 0.05). Immunohistochemical examination revealed markedly reduced α-SMA positivity (a marker of activated myofibroblasts, a key cell type involved in extracellular matrix production and scarring in the kidney) in galectin-3−/− compared with WT mice in the UUO model (Figure 2, e and f). α-SMA was quantified using digital image analysis, and significantly less α-SMA staining occurred in the galectin-3−/− mice compared with WT (Figure 2g, P < 0.05). α-SMA mRNA transcripts, as assessed by real-time PCR, were significantly decreased in the galectin-3−/− mice compared with WT animals (Figure 2h, P < 0.01). Therefore absence of the galectin-3 gene protects against renal fibrosis after UUO. Previous studies in which macrophage recruitment to the kidney was inhibited have suggested a role for macrophages in the development of renal fibrosis.14Ophascharoensuk V Giachelli CM Gordon K Hughes J Pichler R Brown P Liaw L Schmidt R Shankland SJ Alpers CE Couser WG Johnson RJ Obstructive uropathy in the mouse: role of osteopontin in interstitial fibrosis and apoptosis.Kidney Int. 1999; 56: 571-580Crossref PubMed Scopus (253) Google Scholar, 15Lange-Sperandio B Cachat F Thornhill BA Chevalier RL Selectins mediate macrophage infiltration in obstructive nephropathy in newborn mice.Kidney Int. 2002; 61: 516-524Crossref PubMed Scopus (86) Google Scholar, 16Anders HJ Vielhauer V Frink M Linde Y Cohen CD Blattner SM Kretzler M Strutz F Mack M Grone HJ Onuffer J Horuk R Nelson PJ Schlondorff D A chemokine receptor CCR-1 antagonist reduces renal fibrosis after unilateral ureter ligation.J Clin Invest. 2002; 109: 251-259Crossref PubMed Scopus (209) Google Scholar, 17Eis V Luckow B Vielhauer V Siveke JT Linde Y Segerer S Perez De Lema G Cohen CD Kretzler M Mack M Horuk R Murphy PM Gao JL Hudkins KL Alpers CE Grone HJ Schlondorff D Anders HJ Chemokine receptor CCR1 but not CCR5 mediates leukocyte recruitment and subsequent renal fibrosis after unilateral ureteral obstruction.J Am Soc Nephrol. 2004; 15: 337-347Crossref PubMed Scopus (113) Google Scholar, 18Kitagawa K Wada T Furuichi K Hashimoto H Ishiwata Y Asano M Takeya M Kuziel WA Matsushima K Mukaida N Yokoyama H Blockade of CCR2 ameliorates progressive fibrosis in kidney.Am J Pathol. 2004; 165: 237-246Abstract Full Text Full Text PDF PubMed Scopus (262) Google Scholar However, a number of the approaches used do not deplete macrophages specifically, and some deplete neutrophils simultaneously, thereby making interpretation of some of the results more difficult.31Ikezumi Y Hurst LA Masaki T Atkins RC Nikolic-Paterson DJ Adoptive transfer studies demonstrate that macrophages can induce proteinuria and mesangial cell proliferation.Kidney Int. 2003; 63: 83-95Crossref PubMed Scopus (130) Google Scholar, 32Diamond JR Pesek-Diamond I Sublethal X-irradiation during acute puromycin nephrosis prevents late renal injury: role of macrophages.Am J Physiol. 1991; 260: F779-F786PubMed Google Scholar We used the CD11b-DTR mouse5Duffield JS Forbes SJ Constandinou CM Clay S Partolina M Vuthoori S Wu S Lang R Iredale JP Selective depletion of macrophages reveals distinct, opposing roles during liver injury and repair.J Clin Invest. 2005; 115: 56-65Crossref PubMed Scopus (1282) Google Scholar, 33Duffield JS Tipping PG Kipari T Cailhier JF Clay S Lang R Bonventre JV Hughes J Conditional ablation of macrophages halts progression of crescentic glomerulonephritis.Am J Pathol. 2005; 167: 1207-1219Abstract Full Text Full Text PDF PubMed Scopus (202) Google Scholar, 34Cailhier JF Partolina M Vuthoori S Wu S Ko K Watson S Savill J Hughes J Lang RA Conditional macrophage ablation demonstrates that resident macrophages initiate acute peritoneal inflammation.J Immunol. 2005; 174: 2336-2342PubMed Google Scholar to investigate further the role of macrophages in the evolution of tubulointerstitial scarring. The administration of DT to CD11b-DTR mice specifically ablates monocytes and macrophages.5Duffield JS Forbes SJ Constandinou CM Clay S Partolina M Vuthoori S Wu S Lang R Iredale JP Selective depletion of macrophages reveals distinct, opposing roles during liver injury and repair.J Clin Invest. 2005; 115: 56-65Crossref PubMed Scopus (1282) Google Scholar, 34Cailhier JF Partolina M Vuthoori S Wu S Ko K Watson S Savill J Hughes J Lang RA Conditional macrophage ablation demonstrates that resident macrophages initiate acute peritoneal inflammation.J Immunol. 2005; 174: 2336-2342PubMed Google Scholar Immunostaining for macrophages confirmed marked depletion of macrophages in DT-treated mice (Figure 3, a–c) compared to vehicle-treated control mice. We confirmed previous data showing that the numbers of circulating neutrophils, eosinophils, and lymphocytes are not affected by DT treatment in the DTR mouse by flow cytometry (data not shown)5Duffield JS Forbes SJ Constandinou CM Clay S Partolina M Vuthoori S Wu S Lang R Iredale JP Selective depletion of macrophages reveals distinct, opposing roles during liver injury and repair.J Clin Invest. 2005; 115: 56-65Crossref PubMed Scopus (1282) Google Scholar, 33Duffield JS Tipping PG Kipari T Cailhier JF Clay S Lang R Bonventre JV Hughes J Conditional ablation of macrophages halts progression of crescentic glomerulonephritis.Am J Pathol. 2005; 167: 1207-1219Abstract Full Text Full Text PDF PubMed Scopus (202) Google Scholar Macrophage ablation significantly reduced myofibroblast activation and decreased fibrosis as characterized by reduced α-SMA (Figure 3, d–f) and collagen expression (Figure 3, g–i), confirming an important mechanistic role for macrophages in tubulointerstitial scarring after UUO. Circulating fibrocytes derived from the bone marrow have also been shown to contribute to renal fibrosis.35Sakai N Wada T Yokoyama H Lipp M Ueha S Matsushima K Kaneko S Secondary lymphoid tissue chemokine (SLC/CCL21)/CCR7 signaling regulates fibrocytes in renal fibrosis.Proc Natl Acad Sci USA. 2006; 103: 14098-14103Crossref PubMed Scopus (220) Google Scholar Therefore we assessed whether administration of diphtheria-toxin (DT) in our macrophage depletion model had any effect on fibrocyte recruitment to the kidney after UUO. Figure 4 demonstrates that DT treatment did not significantly deplete kidney fibrocyte recruitment after UUO compared with non-DT controls. These results demonstrate that the development of tubulointerstitial fibrosis after UUO is macrophage-dependent.Figure 4Macrophage ablation by DT does not affect recruitment of CD34+/ColI+ and CD45+/ColI+ fibrocytes after UUO. a: CD34; b: ColI; c: merged. Arrows indicate CD34+/ColI+ fibrocytes. d: The number of infiltrating fibrocytes (CD34+/ColI+ and CD45+/ColI+) increased at day 7 after UUO, with no significant difference in infiltrating fibrocyte numbers between non-DT-treated (white bars) and DT-treated (black bars) mice. Day 7 UUO DT−CD34+/ColI+, 26.33 ± 2.8 per mm2; day 7 UUO DT+CD34+/ColI+, 29.83 ± 4.7 per mm2 (n
0

Human Galectin-3 Is a Novel Chemoattractant for Monocytes and Macrophages

Hideki Sano et al.Aug 15, 2000
Abstract Galectin-3 is a β-galactoside-binding protein implicated in diverse biological processes. We found that galectin-3 induced human monocyte migration in vitro in a dose-dependent manner, and it was chemotactic at high concentrations (1.0 μM) but chemokinetic at low concentrations (10–100 nM). Galectin-3-induced monocyte migration was inhibited by its specific mAb and was blocked by lactose and a C-terminal domain fragment of the protein, indicating that both the N-terminal and C-terminal domains of galectin-3 are involved in this activity. Pertussis toxin (PTX) almost completely blocked monocyte migration induced by high concentrations of galectin-3. Galectin-3 caused a Ca2+ influx in monocytes at high, but not low, concentrations, and both lactose and PTX inhibited this response. There was no cross-desensitization between galectin-3 and any of the monocyte-reactive chemokines examined, including monocyte chemotactic protein-1, macrophage inflammatory protein-1α, and stromal cell-derived factor-1α. Cultured human macrophages and alveolar macrophages also migrated toward galectin-3, but not monocyte chemotactic protein-1. Finally, galectin-3 was found to cause monocyte accumulation in vivo in mouse air pouches. These results indicate that galectin-3 is a novel chemoattractant for monocytes and macrophages and suggest that the effect is mediated at least in part through a PTX-sensitive (G protein-coupled) pathway.
0

Galectin-3 and Galectin-1 Bind Distinct Cell Surface Glycoprotein Receptors to Induce T Cell Death

Brianna Stillman et al.Jan 15, 2006
Abstract Galectins are a family of mammalian β-galactoside-binding proteins that positively and negatively regulate T cell death. Extracellular galectin-1 directly induces death of T cells and thymocytes, while intracellular galectin-3 blocks T cell death. In contrast to the antiapoptotic function of intracellular galectin-3, we demonstrate that extracellular galectin-3 directly induces death of human thymocytes and T cells. However, events in galectin-3- and galectin-1-induced cell death differ in a number of ways. Thymocyte subsets demonstrate different susceptibility to the two galectins: whereas galectin-1 kills double-negative and double-positive human thymocytes with equal efficiency, galectin-3 preferentially kills double-negative thymocytes. Galectin-3 binds to a complement of T cell surface glycoprotein receptors distinct from that recognized by galectin-1. Of these glycoprotein receptors, CD45 and CD71, but not CD29 and CD43, appear to be involved in galectin-3-induced T cell death. In addition, CD7 that is required for galectin-1-induced death is not required for death triggered by galectin-3. Following galectin-3 binding, CD45 remains uniformly distributed on the cell surface, in contrast to the CD45 clustering induced by galectin-1. Thus, extracellular galectin-3 and galectin-1 induce death of T cells through distinct cell surface events. However, as galectin-3 and galectin-1 cell death are neither additive nor synergistic, the two death pathways may converge inside the cell.
0

Targeted Disruption of the Galectin-3 Gene Results in Attenuated Peritoneal Inflammatory Responses

Daniel Hsu et al.Mar 1, 2000
Galectin-3 is a member of a growing family of β-galactoside-binding animal lectins. Previous studies have demonstrated a variety of biological activities for this protein in vitro, including activation of cells, modulation of cell adhesion, induction of pre-mRNA splicing, and regulation of apoptosis. To assist in fully elucidating the physiological and pathological functions of this protein, we have generated galectin-3-deficient (gal3−/−) mice by targeted interruption of the galectin-3 gene. Gal3−/− mice consistently developed fewer inflammatory cell infiltrations in the peritoneal cavities than the wild-type (gal3+/+) mice in response to thioglycollate broth treatment, mainly due to lower numbers of macrophages. Also, when compared to cells from gal3+/+ mice, thioglycollate-elicited inflammatory cells from gal3−/− mice exhibited significantly lower levels of NF-κB response. In addition, dramatically different cell-spreading phenotypes were observed in cultured macrophages from the two genotypes. Whereas macrophages from gal3+/+ mice exhibited well spread out morphology, those from gal3−/− mice were often spindle-shaped. Finally, we found that peritoneal macrophages from gal3−/− mice were more prone to undergo apoptosis than those from gal3+/+ mice when treated with apoptotic stimuli, suggesting that expression of galectin-3 in inflammatory cells may lead to longer cell survival, thus prolonging inflammation. These results strongly support galectin-3 as a positive regulator of inflammatory responses in the peritoneal cavity. Galectin-3 is a member of a growing family of β-galactoside-binding animal lectins. Previous studies have demonstrated a variety of biological activities for this protein in vitro, including activation of cells, modulation of cell adhesion, induction of pre-mRNA splicing, and regulation of apoptosis. To assist in fully elucidating the physiological and pathological functions of this protein, we have generated galectin-3-deficient (gal3−/−) mice by targeted interruption of the galectin-3 gene. Gal3−/− mice consistently developed fewer inflammatory cell infiltrations in the peritoneal cavities than the wild-type (gal3+/+) mice in response to thioglycollate broth treatment, mainly due to lower numbers of macrophages. Also, when compared to cells from gal3+/+ mice, thioglycollate-elicited inflammatory cells from gal3−/− mice exhibited significantly lower levels of NF-κB response. In addition, dramatically different cell-spreading phenotypes were observed in cultured macrophages from the two genotypes. Whereas macrophages from gal3+/+ mice exhibited well spread out morphology, those from gal3−/− mice were often spindle-shaped. Finally, we found that peritoneal macrophages from gal3−/− mice were more prone to undergo apoptosis than those from gal3+/+ mice when treated with apoptotic stimuli, suggesting that expression of galectin-3 in inflammatory cells may lead to longer cell survival, thus prolonging inflammation. These results strongly support galectin-3 as a positive regulator of inflammatory responses in the peritoneal cavity. Galectins are members of a recently identified family of β-galactoside-binding animal lectins.1Barondes SH Castronovo V Cooper DNW Cummings RD Drickamer K Feizi T Gitt MA Hirabayashi J Hughes C Kasai K Leffler H Liu F-T Lotan R Mercurio AM Monsigny M Pillai S Poirer F Raz A Rigby PWJ Rini JM Wang JL Galectins: A family of animal β-galactoside-binding lectins. (letter to the editor).Cell. 1994; 76: 597-598Abstract Full Text PDF PubMed Scopus (1085) Google Scholar Presently more than ten members have been identified, and additional homologues are likely to be discovered. Existing information suggests that they may be involved in a multitude of biological processes including regulation of cell growth and apoptosis, neoplastic transformation, and inflammatory responses.2Barondes SH Cooper DNW Gitt MA Leffler H Galectins: structure and function of a large family of animal lectins.J Biol Chem. 1994; 269: 20807-20810Abstract Full Text PDF PubMed Google Scholar, 3Kasai K Hirabayashi J Galectins: a family of animal lectins that decipher glycocodes.J Biochem (Tokyo). 1996; 119: 1-8Crossref PubMed Scopus (453) Google Scholar, 4Hughes RC The galectin family of mammalian carbohydrate-binding molecules.Biochem Soc Trans. 1997; 25: 1194-2298Crossref PubMed Scopus (108) Google Scholar, 5Perillo NL Marcus ME Baum LG Galectins: versatile modulators of cell adhesion, cell proliferation, and cell death.J Mol Med. 1998; 76: 402-412Crossref PubMed Scopus (579) Google Scholar One of the best characterized members, galectin-3, is composed of a carboxyl-terminal lectin domain connected to an amino-terminal nonlectin part consisting primarily of short tandem repeats.6Liu F-T Molecular biology of IgE-binding protein, IgE-binding factors and IgE receptors.CRC Crit Rev Immunol. 1990; 10: 289-306Google Scholar, 7Hughes RC Mac-2: a versatile galactose-binding protein of mammalian tissues.Glycobiology. 1994; 4: 5-12Crossref PubMed Scopus (131) Google Scholar, 8Liu F-T Role of galectin-3 in inflammation.in: Caron M Seve D Lectins and Pathology. Harwood Academic Publishers, London2000: 51-65Google Scholar Galectin-3 is widely distributed in tissues and is found in various epithelial cells, dendritic cells, and inflammatory cells.9Flotte TJ Springer TA Thorbecke GJ Dendritic cell and macrophage staining by monoclonal antibodies in tissue sections and epidermal sheets.Am J Pathol. 1983; 111: 112-124PubMed Google Scholar The expression of this lectin is correlated with cell proliferation,10Moutsatsos IK Wade M Schindler M Wang JL Endogenous lectins from cultured cells: nuclear localization of carbohydrate-binding protein 35 in proliferating 3T3 fibroblasts.Proc Natl Acad Sci USA. 1987; 84: 6452-6456Crossref PubMed Scopus (240) Google Scholar, 11Agrwal N Wang JL Voss PG Carbohydrate-binding protein 35: levels of transcription and mRNA accumulation in quiescent and proliferating cells.J Biol Chem. 1989; 264: 17236-17242Abstract Full Text PDF PubMed Google Scholar cell differentiation,12Liu F-T Hsu DK Zuberi RI Kuwabara I Chi EY Henderson Jr, WR Expression and function of galectin-3, a β-galactoside-binding lectin, in human monocytes and macrophages.Am J Pathol. 1995; 147: 1016-1029PubMed Google Scholar, 13Nangia-Makker P Ochieng J Christman JK Raz A Regulation of the expression of galactoside-binding lectin during human monocytic differentiation.Cancer Res. 1993; 53: 1-5PubMed Google Scholar and transactivation by viral proteins.14Hsu DK Hammes SR Kuwabara I Greene WC Liu F-T Human T lymphotropic virus-1 infection of human T lymphocytes induces expression of the β-galactose-binding lectin, galectin-3.Am J Pathol. 1996; 148: 1661-1670PubMed Google ScholarA number of biological activities of galectin-3 have been demonstrated in vitro. This lectin has been shown to activate various cells, including mast cells,15Frigeri LG Zuberi RI Liu F-T ɛBP, a β-galactoside-binding animal lectin, recognizes IgE receptor (FcɛRI), and activates mast cells.Biochemistry. 1993; 32: 7644-7649Crossref PubMed Scopus (167) Google Scholar, 16Zuberi RI Frigeri LG Liu F-T Activation of rat basophilic leukemia cells by ɛBP, an IgE-binding endogenous lectin.Cell Immunol. 1994; 156: 1-12Crossref PubMed Scopus (61) Google Scholar neutrophils,17Yamaoka A Kuwabara I Frigeri LG Liu F-T A human lectin, galectin-3 (ɛBP/Mac-2), stimulates superoxide production by neutrophils.J Immunol. 1995; 154: 3479-3487PubMed Google Scholar monocytes/macrophages,12Liu F-T Hsu DK Zuberi RI Kuwabara I Chi EY Henderson Jr, WR Expression and function of galectin-3, a β-galactoside-binding lectin, in human monocytes and macrophages.Am J Pathol. 1995; 147: 1016-1029PubMed Google Scholar and lymphocytes.14Hsu DK Hammes SR Kuwabara I Greene WC Liu F-T Human T lymphotropic virus-1 infection of human T lymphocytes induces expression of the β-galactose-binding lectin, galectin-3.Am J Pathol. 1996; 148: 1661-1670PubMed Google Scholar, 18Dong S Hughes RC Galectin-3 stimulates uptake of extracellular Ca2+ in human Jurkat T-cells.FEBS Lett. 1996; 395: 165-169Abstract Full Text PDF PubMed Scopus (79) Google Scholar The effects of this lectin in cell adhesion have also been demonstrated.19Kuwabara I Liu F-T Galectin-3 promotes adhesion of human neutrophils to laminin.J Immunol. 1996; 156: 3939-3944PubMed Google Scholar, 20Inohara H Akahani S Koths K Raz A Interactions between galectin-3 and Mac-2-binding protein mediate cell-cell adhesion.Cancer Res. 1996; 56: 4530-4534PubMed Google Scholar, 21Sato S Hughes RC Binding specificity of a baby hamster kidney lectin for H type I and II chains, polylactosamine glycans, and appropriately glycosylated forms of laminin and fibronectin.J Biol Chem. 1992; 267: 6983-6990Abstract Full Text PDF PubMed Google Scholar These extracellular functions suggest a possible role for this protein in modulation of immune reactions and inflammatory responses. In addition, evidence for various intracellular activities of galectin-3 are also available. It has been identified as a component of hnRNP,22Laing JG Wang JL Identification of carbohydrate binding protein 35 in heterogeneous nuclear ribonucleoprotein complex.Biochemistry. 1988; 27: 5329-5334Crossref PubMed Scopus (97) Google Scholar as well as a factor in pre-mRNA splicing,23Dagher SF Wang JL Patterson RJ Identification of galectin-3 as a factor in pre-mRNA splicing.Proc Natl Acad Sci USA. 1995; 92: 1213-1217Crossref PubMed Scopus (355) Google Scholar and shown to have anti-apoptotic activity, possibly through a mechanism involving its interactions with Bcl-2 family members, with which this lectin shares sequence similarity.24Yang R-Y Hsu DK Liu F-T Expression of galectin-3 modulates T cell growth and apoptosis.Proc Natl Acad Sci USA. 1996; 93: 6737-6742Crossref PubMed Scopus (666) Google Scholar, 25Akahani S Nangia-Makker P Inohara H Kim HRC Raz A Galectin-3: a novel antiapoptotic molecule with a functional BH1 (NWGR) domain of Bcl-2 family.Cancer Res. 1997; 57: 5272-5276PubMed Google ScholarGalectin-3 has been found to be overexpressed in certain pathological conditions, including human atherosclerotic lesions.26Nachtigal M Al-Assaad Z Mayer EP Kim K Monsigny M Galectin-3 expression in human atherosclerotic lesions.Am J Pathol. 1998; 152: 1199-1208PubMed Google Scholar The association of this lectin with neoplastic transformation has also been extensively documented. It is overexpressed in some types of cancer for which the normal parental cells do not express the protein; the examples include specific types of lymphoma,14Hsu DK Hammes SR Kuwabara I Greene WC Liu F-T Human T lymphotropic virus-1 infection of human T lymphocytes induces expression of the β-galactose-binding lectin, galectin-3.Am J Pathol. 1996; 148: 1661-1670PubMed Google Scholar, 27Konstantinov KN Robbins BA Liu F-T Galectin-3, a β-galactoside-binding animal lectin, is a marker of anaplastic large-cell lymphoma.Am J Pathol. 1996; 148: 25-30PubMed Google Scholar thyroid carcinoma,28Fernádez PL Merino MJ Gómez M Campo E Medina T Castronovo V Cardesa A Liu F-T Sobel ME Galectin-3 and laminin expression in neoplastic and non-neoplastic thyroid tissue.J Pathol. 1997; 181: 80-86Crossref PubMed Scopus (135) Google Scholar, 29Xu XC El-Naggar AK Lotan R Differential expression of galectin-1 and galectin-3 in thyroid tumors: potential diagnostic implications.Am J Pathol. 1995; 147: 815-822PubMed Google Scholar and hepatocellular carcinoma.30Hsu DK Dowling CA Jeng KCG Chen JT Yang RY Liu FT Galectin-3 expression is induced in cirrhotic liver and hepatocellular carcinoma.Int J Cancer. 1999; 81: 519-526Crossref PubMed Scopus (195) Google Scholar However, down-regulation of this lectin has been observed in other kinds of neoplasms, including colon,31Lotz MM Andrews Jr, CW Korzelius CA Lee EC Steele Jr, GD Clarke A Mercurio AM Decreased expression of Mac-2 (carbohydrate binding protein 35) and loss of its nuclear localization are associated with the neoplastic progression of colon carcinoma.Proc Natl Acad Sci USA. 1993; 90: 3466-3472Crossref PubMed Scopus (226) Google Scholar, 32Castronovo V Campo E van den Brûle FA Claysmith AP Cioce V Liu F-T Fernandez PL Sobel ME Inverse modulation of steady state mRNA levels of two non-integrin laminin binding proteins in human colon carcinoma.J Natl Cancer Inst. 1992; 84: 1161-1167Crossref PubMed Scopus (97) Google Scholar breast,33Castronovo V Van den Brule FA Jackers P Clausse N Liu FT Gillet C Sobel ME Decreased expression of galectin-3 is associated with progression of human breast cancer.J Pathol. 1996; 179: 43-48Crossref PubMed Scopus (185) Google Scholar ovarian,34van den Brûle FA Berchuck A Bast RC Liu F-T Pieters C Sobel ME Castronovo V Differential expression of the 67-kD laminin receptor and 31-kD human laminin-binding protein in human ovarian carcinoma.Eur J Cancer. 1994; 30A: 1096-1099Abstract Full Text PDF PubMed Scopus (108) Google Scholar and uterine carcinoma,35van den Brûle FA Buicu C Berchuck A Bast RC Deprez M Liu FT Cooper DNW Pieters C Sobel ME Castronovo V Expression of the 67-kD laminin receptor, galectin-1, and galectin-3 in advanced human uterine adenocarcinoma.Hum Pathol. 1996; 27: 1185-1191Abstract Full Text PDF PubMed Scopus (149) Google Scholar although it has also been observed that galectin-3 expression correlates positively with the progression of colon carcinoma.36Irimura T Matsushita Y Sutton RC Carralero D Ohannesian DW Cleary KR Ota DM Nicolson GL Lotan R Increased content of an endogenous lactose-binding lectin in human colorectal carcinoma progressed to metastatic stages.Cancer Res. 1991; 51: 387-393PubMed Google Scholar, 37Schoeppner HL Raz A Ho SB Bresalier RS Expression of an endogenous galactose-binding lectin correlates with neoplastic progression in the colon.Cancer. 1995; 75: 2818-2826Crossref PubMed Scopus (213) Google Scholar Studies of cells transfected with galectin-3 cDNA have indeed provided evidence for its role in tumor transformation and metastasis.38Raz A Zhu D Hogan V Shah N Raz T Karkash R Pazerini G Carmi P Evidence for the role of 34-kDa galactoside-binding lectin in transformation and metastasis.Int J Cancer. 1990; 46: 871-877Crossref PubMed Scopus (172) Google ScholarTherefore, functions of galectin-3 appear to be multifaceted, extending to both intracellular and extracellular compartments. Because of its wide tissue distribution and pleiotropic effects in many systems, galectin-3 is likely to be involved in a variety of physiological and pathological processes. In an attempt to elucidate more definitively the biological functions of galectin-3, we have generated a mouse model in which this gene is inactivated through homologous recombination. We initially focused on the study of inflammatory responses in these mice and have found that the galectin-3 deficiency results in significantly altered inflammation elicited in the peritoneal cavity by thioglycollate broth.Materials and MethodsGeneration of Galectin-3-Null MiceA vector used for homologous recombination was constructed from the cloned galectin-3 genomic DNA.39Gritzmacher CA Mehl VS Liu F-T Genomic cloning of the gene for an IgE-binding lectin reveals unusual utilization of 5′ untranslated regions.Biochemistry. 1992; 31: 9533-9538Crossref PubMed Scopus (35) Google Scholar As shown in Figure 1, a segment from exon 4 to exon 5 within the mouse galectin-3 gene was inserted into pMC1Neo (Stratagene, La Jolla, CA) upstream of the thymidine kinase promoter-Neo cassette. Another segment from exon 5 to exon 6 followed downstream from the Neo cassette. Thus, exon 5 is interrupted by the Neo gene in this vector construct. Murine stem cells, D3, were electroporated with this vector, using procedures previously described.40Shier P Otulakowski G Ngo K Panakos J Chourmouzis E Christjansen L Lau CY Fung-Leung W-P Impaired immune responses toward alloantigens and tumor cells but normal thymic selection in mice deficient in the beta2 integrin leukocyte function-associated antigen-1.J Immunol. 1996; 157: 5375-5386PubMed Google Scholar G418 resistant cells were screened for homologous recombination by polymerase chain reaction (PCR) and Southern blotting, using procedures described below. Screening of 894 clones resulted in two with successful homologous recombination.One clone was propagated and injected into 3.5-day-old blastocysts from C57BL/6 mice and the injected blastocysts were implanted into pseudo-pregnant CD1 mothers. Male chimeric mice were bred to C57BL/6 females to produce mice hemizygous for the galectin-3 null mutant (gal3+/−). Interbreeding gal3+/− mice resulted in mice homozygous for the galectin-3-null condition (gal3−/−). Gal3−/− mice were viable and fertile. For experimentation, wild-type (gal3+/+) mice produced by gal3+/− interbreeding were carried in a separate lineage as controls. Both gal3+/+ and gal3−/− mice were propagated and maintained in standard specific-pathogen-free environments.Polymerase Chain Reaction and Southern Blot AnalysisPolymerase chain reactions were performed under standard conditions using Taq DNA polymerase (Promega, Madison, WI. on a Perkin-Elmer 9600 or Ericomp EZ cycler. Southern blotting analyses were performed by capillary transfer of DNA to charged nylon membranes (BioRad, Richmond, CA or Amersham, Arlington Heights, IL). Membranes were probed with an intron 3 fragment corresponding to probe 1, or probe 4 corresponding to the Neo cassette, to determine homologous recombination, as shown in Figure 1B, radiolabeled by random priming with [32P]-dATP.Immunoblot AnalysisTissues were extracted with 20 mmol/L Tris-HCl, pH 7.5 containing 10 mmol/L EDTA, 0.15 mol/L NaCl, 1% Triton X-100 (v/v) and protease inhibitors 0.24 u/ml Aprotinin, 1 μg/ml pepstatin, 1 μg/ml leupeptin, 1 mmol/L phenylmethylsulfonyl fluoride, and 100 μg protein from each extract was then adsorbed with lactosyl-Sepharose 4B.41Hsu DK Zuberi R Liu F-T Biochemical and biophysical characterization of human recombinant IgE-binding protein, an S-type animal lectin.J Biol Chem. 1992; 267: 14167-14174Abstract Full Text PDF PubMed Google Scholar The bound proteins were eluted, separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and analyzed by immunoblotting using specific rabbit antibodies as described.12Liu F-T Hsu DK Zuberi RI Kuwabara I Chi EY Henderson Jr, WR Expression and function of galectin-3, a β-galactoside-binding lectin, in human monocytes and macrophages.Am J Pathol. 1995; 147: 1016-1029PubMed Google ScholarInduction of Peritoneal InflammationMice were injected with 1 ml of autoclaved Brewer's thioglycollate broth as described.42Mishell BB Selected Methods in Cellular Immunology. Freeman, San Francisco1980: 3Google Scholar At various intervals, peritoneal cells were harvested by lavage with minimum essential medium containing Earle's salts (MEM). Cells were washed once with culture medium before subsequent manipulations. For quantitation of leukocyte subpopulations, the cells in the lavage fluid were allowed to attach to glass slides by cytospin, treated with soluble Wright's stain (Leukostat, Fisher Scientific, Pittsburgh, PA) and identified as macrophages, eosinophils, neutrophils, and lymphocytes by standard morphology. Cell counts were obtained in triplicate for each sample from 100–200 cells using a 100× oil immersion objective.Electrophoretic Mobility Shift Assay (EMSA)Nuclear extracts were prepared by a previously described method43Dignam JD Lebovitz RM Roeder RG Accurate transcription initiation by RNA polymerase II in a soluble extract from isolated mammalian nuclei.Nucleic Acids Res. 1983; 11: 1475-1489Crossref PubMed Scopus (9142) Google Scholar with slight modifications44Kravchenko VV Pan Z Han J Herbert JM Ulevitch RJ Ye RD Platelet-activating factor induces NF-kappa B activation through a G protein-coupled pathway.J Biol Chem. 1995; 270: 14928-14934Abstract Full Text Full Text PDF PubMed Scopus (157) Google Scholar and EMSA was performed as described.45Browning DD Pan ZK Prossnitz ER Ye RD Cell type- and developmental stage-specific activation of NF-kappaB by fMet-Leu-Phe in myeloid cells.J Biol Chem. 1997; 272: 7995-8001Crossref PubMed Scopus (59) Google Scholar Briefly, the nuclear extracts (2.5 μg protein) in 12 μl of binding buffer (5 mmol/L HEPES, pH 7.8, 5 mmol/L MgCl2, 50 mmol/L KCl, 0.5 mmol/L dithiothreitol, 0.4 mg/ml poly(dI-dC) (Pharmacia, Piscataway, NJ), 0.1 mg/ml sonicated double-stranded salmon sperm DNA, and 10% glycerol) were incubated for 10 minutes at room temperature. Subsequently, approximately 20 to 50 fmoles of 32P-labeled NF-κB-specific oligonucleotide probe (30,000–50,000 cpm) were added and the reaction mixture was incubated for 10 minutes at room temperature. The samples were analyzed on 6% polyacrylamide gels in 50 mmol/L Tris-borate buffer containing 1 mmol/L EDTA or 50 mmol/L Tris/380 mmol/L glycine/2 mmol/L EDTA, at 12 V/cm for 2 to 2.5 hours. Radioactivities were detected by exposure to X-ray film or phosphorimager plates.Culture and Measurement of Adhesion Areas of Peritoneal MacrophagesPeritoneal macrophages were enriched by adherence onto tissue culture-treated plates in Dulbecco's modified Eagle's medium (DMEM), supplemented with 2 mmol/L glutamine and 10% (v/v) fetal bovine serum, at 37°C in an atmosphere of 7.5% CO2. After two hours of incubation, wells were gently pipetted to remove nonadherent cells. More than 90% of the adherent cells were macrophages as evaluated morphologically after processing with Wright's stain. Initially nonadherent cells were obtained after two serial adherence procedures in tissue culture flasks. Culturing of these cells resulted in additional adherent macrophages. The morphology of adherent macrophages cultured for various periods were imaged from three fields of each well using a Hamamatsu XC-77 CCD camera attached to a Nikon microscope with an inverted stage. Image-1 from Universal Imaging Corporation was used to acquire and enhance image contrast, and NIH Image software was used to measure cell attachment areas from three random fields.Induction of Apoptosis of Peritoneal Macrophages in VitroInflammatory peritoneal macrophages were obtained from mice treated with 1 ml thioglycollate broth for 3 days by lavage, and cultured in RPMI/10% fetal bovine serum (RP10F) for 1 hour. Adherent cells were cultured in the presence of 10 μg/ml lipopolysaccharide (LPS, Escherichia coli 0111:B4 List Biologicals, Campbell, CA) and 10 U/ml interferon-γ (IFN-γ, Boehringer/Roche, Indianapolis, IN), and cell viability was measured by the MTT (3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide) assay46Hansen MB Nielsen SE Berg K Re-examination and further development of a precise and rapid dye method for measuring cell growth/cell kill.J Immunol Methods. 1989; 119: 203-210Crossref PubMed Scopus (3316) Google Scholar at regular intervals. Alternatively, inflammatory macrophages obtained from peritoneal cavities of mice 4 days after 2 ml thioglycollate treatment, were cultured for 30 minutes in RPMI 1640. The adherent cells were exposed to 0 to 60 U/ml IFN-γ in RP10F for 6 hours washed with RP10F, and then 1 μg/ml LPS in RP10F for 24 hours.47Shnyra A Brewington R Alipio A Amura C Morrison DC Reprogramming of lipopolysaccharide-primed macrophages is controlled by a counterbalanced production of IL-10 and IL-12.J Immunol. 1998; 160: 3729-3736PubMed Google Scholar The cells were washed with phosphate-buffered saline, pH 7.2, fixed in 5. formaldehyde/phosphate-buffered saline, and stained with 0.05% crystal violet in 20% EtOH. Absorbance at 550 nm in methanol due to the stained adherent cells was measured after the wells were extensively washed with deionized water.Detection of Apoptotic Macrophages by Annexin-V BindingInflammatory peritoneal exudate cells (5 × 106/ml) were cultured in Teflon beakers in the presence of IFN-γ for 6 hours, washed, and then incubated with 1 μg/ml LPS as described above. At various time points, cells were removed and stained with annexin-V-fluorescein isothiocyanate (PharMingen, La Jolla, CA) according to manufacturer's directions. Processed cells were analyzed by flow cytometry on a Becton Dickinson (San Jose, CA) Facscan.Statistical AnalysisComparison of data from gal3+/+ and gal3−/− mice were performed with the statistical software Statview. Data were subjected to the Mann-Whitney U test or analysis of variance with Bonferroni-Dunn post hoc analysis.ResultsHomologous Recombination and Production of gal3−/− MiceThe genomic structure of galectin-3 contains six exons39Gritzmacher CA Mehl VS Liu F-T Genomic cloning of the gene for an IgE-binding lectin reveals unusual utilization of 5′ untranslated regions.Biochemistry. 1992; 31: 9533-9538Crossref PubMed Scopus (35) Google Scholar (Figure 1A), with exons 2 and 3 coding for the amino-terminal region and exons 4–6 coding for the carboxyl-terminal carbohydrate-binding domain. Our strategy for inactivating galectin-3 in mice was to interrupt the region coding for the carbohydrate-binding domain with a neomycin resistance gene. Specifically, a short intron 4-exon 5 segment (0.5 kb) was substituted with the antibiotic-resistance gene (Neo). Figure 1B depicts the structure of the homologous recombinant galectin-3 gene and Figure 1C predicts the restriction fragment profiles of both wild type and homologous recombinants using the specific probes shown in Figure 1B. The genomic Southern blot of two homologous recombinant gal3+/− mouse embryonic stem cell clones is shown in Figure 1D, visualized with probe 1. The upper band in each lane for clones 4A2 and 9A4 were detected with probe 4, but no bands were observed in parent D3 (data not shown).Figure 2A shows the Southern blot analysis of galectin-3 gene from gal3−/−, gal3+/+, and gal3+/− mice. The 2.9-kb intron 3-exon 5 fragment of the galectin-3+/+ gene and the corresponding 4-kb homologous recombinant of the galectin-3−/− are evident. Hemizygous mice are characterized by the presence of both DNA fragments. To demonstrate that the galectin-3 gene was indeed inactivated in gal3−/− mice, several organ extracts were prepared and adsorbed with lactosyl-Sepharose 4B, and the adsorbed proteins were analyzed by immunoblotting. As shown in Figure 2B, although gal3+/+ mice expressed large amounts of galectin-3 in lung, spleen, and thymus tissues, gal3−/− mice were deficient in this protein in these various organs, as expected. Gal3−/− mice are viable and fertile and do not exhibit any overt defects. Various organs from gal3−/− mice were further examined histologically and no apparent alterations were detected. Organs examined included adrenal gland, brain, colon, duodenum, esophagus, gall bladder, heart, hypophysis, kidney, knee joint, liver, lung, lymph nodes, mesentery, ovary, pancreas, salivary glands, skeletal muscle, skin, small intestine, stomach, spleen, testis, thymus, thyroid, and urinary bladder. No significant differences were observed in body weights and weights of major organs, and no differences were found in blood cell counts and chemistry profiles between gal3−/− and gal3+/+ mice. Lymphocyte subpopulations of thymus, spleen, and lymph node were also examined and total numbers of lymphocytes, ratios of CD4+/CD8+ cells, and numbers of CD3+ cells, were comparable between gal3−/− and gal3+/+ mice.Figure 2Confirmation of the inactivated galectin-3 gene in homologous recombinant mice. A: Southern blot analysis of DNA fragments from homozygous (gal3−/− and gal3+/+) and hemizygous (gal3+/−. mice digested with HindIII and XbaI, and analyzed using probe 1, as described in Figure 1D. The positions of DNA markers (kb) are shown at the left margin. B: Immunoblot of organ extracts from gal3+/+ and gal3−/− mice adsorbed with lactosyl-Sepharose and probed with a specific rabbit anti-galectin-3 antiserum. Each lane represents galectin-3 from 100 μg total protein. Lu, lung; Li, liver; S, spleen; T, thymus. The positions of protein Mr standards (× 10−3) are shown at the left margin.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Gal3−/− Mice Exhibit Decreased Levels of Peritoneal InflammationAs an initial approach to assess the effect of galectin-3 deficiency on inflammatory responses, we examined the peritoneal inflammation induced by thioglycollate broth. Untreated (Day 0. gal3−/− mice consistently contained fewer leukocytes in the peritoneal cavities, although the differences were not statistically significant (P = 0.1879). Although both types of mice mounted inflammatory responses to thioglycollate broth treatment, gal3−/− mice clearly exhibited reduced inflammation (Figure 3, Table 1). One day after thioglycollate broth stimulation, the yields of inflammatory leukocytes in the peritoneal cavities of gal3−/− mice were significantly lower (P < 0.05) than that of gal3+/+ mice (Figure 3). Similar trends were observed 3 and 6 days after thioglycollate stimulation (Table 1).Figure 3Reduced peritoneal inflammatory responses are observed in gal3−/− mice when induced with thioglycollate broth. A: Mice matched by sex and age were injected intraperitoneally with 1 ml of thioglycollate broth, and inflammatory cells were harvested at various intervals. Viable leukocytes determined by trypan blue exclusion were enumerated in duplicate from individual mice. Means for each time point are shown with SE bars from 9 and 5 separate experiments, with a total of 22 and 14 mice for each genotype for days 0 (untreated. and 1, respectively. The results from gal3+/+ mice are shown in closed circles and those from gal3−/− mice are shown in open circles. B: Inflammatory peritoneal cells obtained 1 day after
0
Citation424
0
Save
0

Antibody-dependent SARS coronavirus infection is mediated by antibodies against spike proteins

Sheng‐Fan Wang et al.Jul 26, 2014
The severe acute respiratory syndrome coronavirus (SARS-CoV) still carries the potential for reemergence, therefore efforts are being made to create a vaccine as a prophylactic strategy for control and prevention. Antibody-dependent enhancement (ADE) is a mechanism through which dengue viruses, feline coronaviruses, and HIV viruses take advantage of anti-viral humoral immune responses to infect host target cells. Here we describe our observations of SARS-CoV using ADE to enhance the infectivity of a HL-CZ human promonocyte cell line. Quantitative-PCR and immunofluorescence staining results indicate that SARS-CoV is capable of replication in HL-CZ cells, and of displaying virus-induced cytopathic effects and increased levels of TNF-α, IL-4 and IL-6 two days post-infection. According to flow cytometry data, the HL-CZ cells also expressed angiotensin converting enzyme 2 (ACE2, a SARS-CoV receptor) and higher levels of the FcγRII receptor. We found that higher concentrations of anti-sera against SARS-CoV neutralized SARS-CoV infection, while highly diluted anti-sera significantly increased SARS-CoV infection and induced higher levels of apoptosis. Results from infectivity assays indicate that SARS-CoV ADE is primarily mediated by diluted antibodies against envelope spike proteins rather than nucleocapsid proteins. We also generated monoclonal antibodies against SARS-CoV spike proteins and observed that most of them promoted SARS-CoV infection. Combined, our results suggest that antibodies against SARS-CoV spike proteins may trigger ADE effects. The data raise new questions regarding a potential SARS-CoV vaccine, while shedding light on mechanisms involved in SARS pathogenesis.
0
Citation417
0
Save
0

Identification of Galectin-3 As a High-Affinity Binding Protein for Advanced Glycation End Products (AGE): A New Member of the AGE-Receptor Complex

Helen Vlassara et al.Sep 1, 1995
Advanced glycation end products (AGE), the reactive derivatives of nonenzymatic glucose-protein condensation reactions, are implicated in the multiorgan complications of diabetes and aging. An AGE-specific cellular receptor complex (AGE-R) mediating AGE removal as well as multiple biological responses has been identified. By screening an expression library using antibody against a previously identified component of the AGE-R complex p90, a known partial cDNA clone was isolated with homology to galectin-3, a protein of diverse identity, and member of the galectin family.To explore this unexpected finding, the nature of the interactions between galectin-3 and AGE was studied using intact macrophage-like RAW 264.7 cells, membrane-associated and recombinant galectin-1 through -4, and model AGE-ligands (AGE-BSA, FFI-BSA).Among the members of this family (galectin-1 through 4), recombinant rat galectin-3 was found to exhibit high-affinity 125I-AGE-BSA binding with saturable kinetics (kD 3.5 x 10(7) M-1) that was fully blocked by excess unlabeled naturally formed AGE-BSA or synthetic FFI-BSA, but only weakly inhibited by several known galectin-3 ligands, such as lactose. In addition to the p90, immunoprecipitation with anti-galectin-3, followed by 125I-AGE-BSA ligand blot analysis of RAW 264.7 cell extracts, revealed galectin-3 (28 and 32 kD), as well as galectin-3-associated proteins (40 and 50 kD) with AGE-binding activity. Interaction of galectin-3 with AGE-BSA or FFI-BSA resulted in formation of SDS-, and beta-mercaptoethanol-insoluble, but hydroxylamine-sensitive high-molecular weight complexes between AGE-ligand, galectin-3, and other membrane components.The findings point toward a mechanism by which galectin-3 may serve in the assembly of AGE-R components and in the efficient cell surface attachment and endocytosis by macrophages of a heterogenous pool of AGE moieties with diverse affinities, thus contributing to the elimination of these pathogenic substances.
0

Galectin-3 is an important mediator of VEGF- and bFGF-mediated angiogenic response

Anna Markowska et al.Aug 16, 2010
Recent studies have shown that a carbohydrate-binding protein, galectin-3, is a novel pro-angiogenic molecule. The mechanism by which galectin-3 promotes angiogenesis remains unknown. We demonstrate here that galectin-3 is a mediator of vascular endothelial growth factor (VEGF)- and basic fibroblast growth factor (bFGF)-mediated angiogenic response. Angiogenesis assays revealed that galectin-3 inhibitors, β-lactose and dominant-negative galectin-3, reduce VEGF- and bFGF-mediated angiogenesis in vitro and that VEGF- and bFGF-mediated angiogenic response is reduced in galectin-3 knockdown cells and Gal3−/− animals. Integrin αvβ3 was identified as the major galectin-3–binding protein and anti-αv, -β3, and -αvβ3 integrin function-blocking antibodies significantly inhibited the galectin-3–induced angiogenesis. Furthermore, galectin-3 promoted the clustering of integrin αvβ3 and activated focal adhesion kinase. Knockdown of GnTV, an enzyme that synthesizes high-affinity glycan ligands for galectin-3, substantially reduced: (a) complex N-glycans on αvβ3 integrins and (b) VEGF- and bFGF-mediated angiogenesis. Collectively, these data suggest that galectin-3 modulates VEGF- and bFGF-mediated angiogenesis by binding via its carbohydrate recognition domain, to the GnTV synthesized N-glycans of integrin αvβ3, and subsequently activating the signaling pathways that promote the growth of new blood vessels. These findings have broad implications for developing novel, carbohydrate-based therapeutic agents for inhibition of angiogenesis.
0

Population structure of Han Chinese in the modern Taiwanese population based on 10,000 participants in the Taiwan Biobank project

Chien-Hsiun Chen et al.Oct 11, 2016
The Taiwan Biobank (TWB) aims to build a nationwide research database that integrates genomic/epigenomic profiles, lifestyle patterns, dietary habits, environmental exposure history and long-term health outcomes of 300,000 residents of Taiwan. We describe here an investigation of the population structure of Han Chinese on this Pacific island using genotype data of 591,048 SNPs in an initial freeze of 10,801 unrelated TWB participants. In addition to the North-South cline reported in other Han Chinese populations, we find the Taiwanese Han Chinese clustered into three cline groups: 5% were of northern Han Chinese ancestry, 79.9% were of southern Han Chinese ancestry, and 14.5% belonged to a third (T) group. We also find that this T group is genetically distinct from neighbouring Southeast Asians and Austronesian tribes but similar to other southern Han Chinese. Interestingly, high degree of LD between HLA haplotype A*33:03-B*58:01, an MHC allele being of pathological relevance, and SNPs across the MHC region was observed in subjects with T origin, but not in other Han Chinese. This suggested the T group individuals may have experienced evolutionary events independent from the other southern Han Chinese. Based on the newly-discovered population structure, we detect different loci susceptible to type II diabetes in individuals with southern and northern Han Chinese ancestries. Finally, as one of the largest dataset currently available for the Chinese population, genome-wide statistics for the 10,810 subjects are made publicly accessible through Taiwan View (https://taiwanview.twbiobank.org.tw/index; date last accessed October 14, 2016) to encourage future genetic research and collaborations with the island Taiwan.
0
Citation247
0
Save
Load More