1,2-Linked Mannans Β Expressing Specific Species Candida Galectin

Home , Collectin

Galectin-3 Induces Death of Candida Species Expressing Specific β-1,2-Linked Mannans Luciana Kohatsu, Daniel K. Hsu, Armin G. Jegalian, Fu-Tong Liu and Linda G. Baum This information is current as of September 30, 2021. J Immunol 2006; 177:4718-4726; ; doi: 10.4049/jimmunol.177.7.4718 http://www.jimmunol.org/content/177/7/4718 Downloaded from

References This article cites 72 articles, 34 of which you can access for free at: http://www.jimmunol.org/content/177/7/4718.full#ref-list-1

Why The JI? Submit online. http://www.jimmunol.org/

• 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 by guest on September 30, 2021

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 © 2006 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Galectin-3 Induces Death of Candida Species Expressing Speciﬁc ␤-1,2-Linked Mannans1

Luciana Kohatsu,* Daniel K. Hsu,† Armin G. Jegalian,* Fu-Tong Liu,† and Linda G. Baum2*

Lectins play a critical role in host protection against infection. The galectin family of lectins recognizes saccharide ligands on a variety of microbial pathogens, including viruses, bacteria, and parasites. Galectin-3, a galectin expressed by macrophages, dendritic cells, and epithelial cells, binds bacterial and parasitic pathogens including Leishmania major, Trypanosoma cruzi, and Neisseria gonorrhoeae. However, there have been no reports of galectins having direct effects on microbial viability. We found that galectin-3 bound only to Candida albicans species that bear ␤-1,2-linked oligomannans on the cell surface, but did not bind Saccharomyces cerevisiae that lacks ␤-1,2-linked oligomannans. Surprisingly, binding directly induced death of Candida species containing speciﬁc ␤-1,2-linked oligomannosides. Thus, galectin-3 can act as a pattern recognition receptor that recognizes a

unique pathogen-speciﬁc oligosaccharide sequence. This is the ﬁrst description of antimicrobial activity for a member of the Downloaded from galectin family of mammalian lectins; unlike other lectins of the innate immune system that promote opsonization and phagocytosis, galectin-3 has direct fungicidal activity against opportunistic fungal pathogens. The Journal of Immunology, 2006, 177: 4718–4726.

nfections with opportunistic fungi are a critical health con- isms, including worms, sponges, multicellular fungi, and insects

cern; thus, elucidation of different innate immune strategies (13–19). Galectins are PRRs in several types of organisms, and http://www.jimmunol.org/ I to control fungal infections is an important goal. A number of many mammalian pathogens express saccharide structures that are endogenous lectins participate in innate immune responses and are recognized by galectins. Galectin-1 binds saccharide ligands on important for control of microbial infections, especially those envelope glycoproteins of Nipah virus and HIV (20, 21). Galec- caused by fungal pathogens. As pattern-recognition receptors tin-3 and galectin-9 bind Leishmania major and galectin-9 pro- (PRRs),3 lectins recognize unique carbohydrate ligands, or patho- motes L. major-macrophage interactions (22, 23). Galectin-3 binds gen-associated molecular patterns (PAMPs), present on the surface mycolic acids, a major component of the cell envelope of Myco- of the pathogen but absent in the host (1–5). Fungal oligosaccha- bacterium tuberculosis (24), and participates in clearance of late ride PAMPs are recognized by lectin PRRs to induce rapid and mycobacterial infections (25). Galectin-3 also binds Pseudomonas broad host defense responses such as opsonization, activation of aeruginosa, Klebsiella pneumoniae, and Neisseria gonorrhoeae by guest on September 30, 2021 complement, activation of coagulation cascades, phagocytosis, in- (26–28), as well as GalNAc1–4GlcNAc sequences in Schistosoma flammation, and direct microbial killing (3). Engagement of PRRs mansoni soluble egg Ag (29). Galectin-3 mediates adhesion of on macrophages and dendritic cells can also attract, activate, and Trypanosoma cruzi to human vascular smooth muscle cells (30), regulate T cells that are critical for an acquired immune response and expression of human galectin-1 and galectin-3 is up-regulated to fungal pathogens (6). in APCs and gastric epithelial cells infected with T. cruzi and Several types of lectins function as PRRs during the host re- Helicobacter pylori (31–33). Several studies indicate that galec- sponse to fungal infections, including pentraxin-3, dectin-1, and tin-3 specifically participates in innate immunity, as galectin-3 is the collectin family members surfactant proteins A and D (1, expressed in a variety of cell types including dendritic cells, mac- 7–11). Recently, it has become apparent that the galectin family of rophages, and NK cells, as well as activated T and B cells (34–38). lectins can also participate in the innate immune defense against However, while galectin-3 can recognize specific PAMPs, no di- pathogens (12). Galectins are present in all multicellular organ- rect microbicidal function for galectin-3, or any galectin, has been reported. Galectins possess a conserved carbohydrate-recognition domain *Department of Pathology and Laboratory Medicine, Jonsson Comprehensive Cancer (CRD) and typically bind to glycans containing ␤-galactosides Center, School of Medicine, University of California, Los Angeles, CA 90095; and †Department of Dermatology, School of Medicine, University of California Davis, (39). However, subtle structural differences among galectin CRDs Sacramento, CA 95817 result in distinct binding affinities for specific glycan ligands (40). Received for publication February 6, 2006. Accepted for publication July 19, 2006. Galectin-3 has an extended carbohydrate-binding pocket compared The costs of publication of this article were defrayed in part by the payment of page with galectin-1 (41). This structural difference allows galectin-3 to charges. This article must therefore be hereby marked advertisement in accordance bind to a wider range of oligosaccharide structures, including with 18 U.S.C. Section 1734 solely to indicate this fact. structures containing mannose (41). Galectin-3 also differs struc- 1 This work was supported by National Institutes of Health (NIH) Grant AI07323 turally from the other members of the galectin family. The 14 (Microbial Pathogenesis Training Grant) (to L.K.), NIH Grant R01 GM63281 (to L.G.B.), and R01 AI20958 and R01 AI39620 (to F.-T.L.). mammalian galectins are divided into three subgroups, mono- 2 Address correspondence and reprint requests to Dr. Linda G. Baum, Department of meric, tandem repeat, and chimeric, based on domain structure. Pathology and Laboratory Medicine, School of Medicine, University of California Galectin-3 is the only member of the chimera-type galectin sub- Los Angeles School of Medicine, 10833 Le Conte Avenue, Los Angeles, CA 90095. group, with a CRD in the C terminus and a distinct N-terminal E-mail address: [email protected] domain that mediates oligomerization of the lectin into pentamers 3 Abbreviations used in this paper: PRR, pattern recognition receptor; PAMP, pathogen-associated molecular pattern; CRD, carbohydrate recognition domain; SA, upon binding multivalent saccharide ligands (13, 42, 43). As men- streptavidin; DTAF, (4,6-dichlorotriazinyl) aminofluorescein. tioned above, galectin-3 is highly expressed in macrophages and

immature dendritic cells; expression is up-regulated in activated search Laboratories) by flow cytometry on a BD-LSR I Analytic Flow macrophages and is down-regulated during dendritic cell matura- Cytometer and data were analyzed with CellQuest software (BD Bio- ␮ tion (37, 44). Galectin-3 is also expressed in many types of epi- sciences). For inhibition assays, 22.5 g of galectin-3 was preincubated on ice with 0.1 M ␤-lactose (Sigma-Aldrich), 0.1 M sucrose (Fisher Scien- thelial cells and stromal cells, including fibroblasts after activation tific), 1 mg/ml S. cerevisiae cell wall mannan (Sigma-Aldrich), or 1 mg/ml or adhesion (45, 46). Like all galectins, galectin-3 is made as a C. albicans serotype A cell wall mannan Takara MG001 (Takara Mirus monomer in the cytosol and is secreted from the cytosol via a Bio), before binding to live yeast. nonclassical secretion mechanism; secreted galectin-3 binds to Serotyping of cell surface expression of ␤-1,2-oligomannan multivalent saccharide ligands on cells and extracellular matrix in antigenic factors 5 and 6 the immediate milieu (47–50). Candida check rabbit antisera against antigenic factors 5 (F5) and 6 (F6) Surprisingly, Fradin et al. (51) demonstrated that galectin-3 ␤ ␤ (Iatron Laboratories) that specifically identify distinct Candida -1,2- binds to -1,2-linked oligomannosides, an uncommon PAMP linked oligomannans were used. Rabbit antisera against F5 recognizes ␤ present on the surface of the pathogenic fungus Candida albicans (Man- 1,2-Man)nՆ2 sequences on either acid labile or acid stable phos- but absent on Saccharomyces cerevisiae. Three types of ␤-1,2- phodiesterified N-glycans (45). Rabbit antisera against F6 recognizes ␤ ␣ linked oligomannans have been identified in different species of (Man- 1,2-Man)nՆ1- 1,2-Man sequences only on acid labile phosphodi- esterified N-glycans (53, 54). A total of 5 ϫ 105 Candida or S. cerevisiae the genus Candida (52–54). We now demonstrate that galectin-3, cells were incubated on ice for 1 h with 1/25 dilution of primary antiserum but not galectin-1, binds in a carbohydrate-dependent manner to to F5 (IF5, Candida Check Iatron RM302-3), 1/50 dilution of primary Candida species bearing ␤-1,2-oligomannosides in the yeast cell antiserum to F6 (IF6, Candida Check Iatron RM302-4), or 1/25 dilution of wall, and that galectin-3 binding to C. albicans is inhibited by C. preimmune rabbit serum. Bound Ab was detected with FITC-labeled goat anti-rabbit IgG (diluted 1/50) (Jackson ImmunoResearch Laboratories) by albicans mannans but not by S. cerevisiae mannans. Moreover, Downloaded from flow cytometry and data analyzed with CellQuest software as described binding of galectin-3 resulted in death of yeast expressing specific above. ␤-1,2-oligomannoside structures. Thus, we report a novel fungi- Assessment of yeast viability cidal activity for galectin-3, suggesting that galectin-3 participates in host protection against opportunistic fungal infections. For initial assessment of CFU, 0.5 ϫ 105 C. albicans ATCC 10231 were incubated with the indicated concentration of galectin-3 or buffer alone in a final volume of 50 ␮l at 37°C for 2 h. For cell sorting, 5 ϫ 105 C. Materials and Methods albicans ATCC 10231 were incubated with the indicated concentration of http://www.jimmunol.org/ Yeast species and reagents galectin-3 or buffer alone in a final volume of 50 ␮l at 37°C for 2 h. Six C. albicans serotype A (ATCC 10231), C. albicans serotype B (ATCC tubes of treated yeast were pooled for flow cytometric cell sorting using a 60193), Candida glabrata (ATCC 2001), and Candida guilliermondii FACSAria cell sorter (BD Biosciences). The small, granular population of (ATCC 6260) were obtained from the American Type Culture Collection. cells in galectin-3-treated samples were sorted from the cell population that Cells were cultured on Sabouraud dextrose agar plates at 37°C, and sub- demonstrated normal morphology and each population was collected under cultured for three passages in Sabouraud dextrose broth for 24 h before sterile conditions in separate tubes. The number of cells collected in each tube was recorded, and volumes were adjusted to yield 1.8 ϫ 104 cells/ each experiment. S. cerevisiae BY4741a (gift from Dr. G. Payne, Univer- ␮ sity of California, Los Angeles, CA (UCLA)) was passaged three times in ml/tube. Samples were diluted 1/10 and 10- l samples were plated on YPD medium (1% Becto yeast extract, 2% galactose, and 2% dextrose) at Sabouraud dextrose agar and cultured at 37°C for 24 h. Colony images 30°C for 24 h before each experiment. For all experiments, cells were were captured using a model 2.1.1 camera and a Chemilmager 5500 digital by guest on September 30, 2021 ϫ imaging system (Alpha Innotech). washed three times in cold PBS and adjusted to a final OD450 of 1.0 (0.5 107 yeast/ml). Recombinant human galectin-3 was made as previously de- Quantitative assessment of yeast viability by Fun-1 scribed (55). Cell viability was analyzed using the dye Fun-1 (Molecular Probes). The Tissue immunostaining Fun-1 dye initially produces a diffuse green cytoplasmic staining of yeast. To confirm that flow cytometric analysis of Fun-1 staining could discrim- Human tissue was obtained from the Division of Anatomic Pathology, inate live (FL-1bright) and dead (FL1dim) cells, 20 ␮M Fun-1 was added to UCLA Department of Pathology, with appropriate institutional review and live or killed yeast for 30 min at 37°C, to avoid the conversion of the dye approval for use. Five-micron sections of paraffin-embedded tissue were in live cells to yellow-orange that occurs by 1 h, according the manufac- heated at 60°C for 30 min, deparaffinized with xylene and alcohol, and turer’s insert. Analysis of cells in the FL1 and FL2 channels at various time fixed with 3% H O for 5 min. Slides were blocked with 1% BSA in PBS 2 2 points after addition of dye demonstrated greater discrimination in FL1 for 20 min before addition of anti-galectin-3 rabbit serum or nonimmune than in FL2, consistent with the emission data in the manufacturer’s insert, rabbit serum diluted 1/1000 in 1% BSA/PBS. Bound primary Ab was de- so that all samples were analyzed in FL1 to detect green fluorescence. As tected with HRP-conjugated goat anti-rabbit IgG 1/1000 in PBS, and de- controls for discrimination of dead and live cells, cells were killed with veloped with chromogen reagent (Peroxidase Chromogen kit; Biomeda) heat (80°C for 5 min and stored on ice) or Fungizone (4 ␮g/ml; Invitrogen for 10 min at room temperature. Slides were counterstained with hema- Life Technologies). Triplicate samples were analyzed by flow cytometry toxylin for 1–2 min at room temperature and washed with dH O. Parallel 2 using a BD-LSR I Analytic Flow Cytometer equipped with an argon-ion sections were stained with periodic acid-Schiff reagent. Images were ana- laser. The excitation laser was set at 488 nm and Fun-1 green fluorescence lyzed on a BX41 light microscope (Olympus). was collected with a 530/28-nm bandpass filter. A total of 10,000 events Detection of galectin-3 binding to yeast were acquired per analysis. To assess the effects of galectin-3 on yeast cell viability, 5 ϫ 105 cells of each Candida strain were incubated with 7.5–105 Galectin-1, -3, and control BSA were biotinylated using the manufacturer’s ␮g (5–70 ␮M) of galectin-3 in a final volume of 50 ␮l; for inhibition recommended protocol (EZ-Link Sulfo-NHS-Biotin kit; Pierce). A total of studies, 105 ␮g of galectin-3 was preincubated on ice with 0.1 M ␤-lactose ϫ 5 ␮ 5 10 C. albicans ATCC 60193 in 50 lofdH2O were heat fixed on (Sigma-Aldrich), or 0.1 M sucrose (Fisher Scientific) before addition to the slides, blocked with 1% BSA/0.1 M sucrose, and blocked with blocking cells. Samples were treated at 37°C for 2 h and cell viability analyzed with reagent (Avidin/Biotin Blocking kit; Vector Laboratories). Cells were in- Fun-1 as described above. cubated with 30 ␮gof(20␮M) biotinylated galectin-3, 30 ␮g of biotinylated galectin-3 plus 0.1 M lactose, or buffer alone. Bound protein was Results detected using streptavidin (SA)-HRP (Zymed Laboratories) and devel- Galectin-3 is present in fungal granulomata oped as above for 10 min at room temperature (RT). Images were analyzed on a Zeiss Axioskop2 plus microscope and photographed with a Zeiss Galectin-3 has been reported to bind to ␤-1,2-mannans isolated Axiocam digital camera. from C. albicans serotype A (51). Because galectin-3 has been 5 Alternatively, triplicate samples of 5 ϫ 10 Candida or S. cerevisiae found in liver granulomata from hamsters infected with Schisto- were incubated for 45 min at 4°C with 15 ␮g (20 ␮M) biotinylated galectin-1, 18–30 ␮g (12.5 to 20 ␮M) biotinylated galectin-3, or 50 ␮gof soma mansoni (29), we asked whether galectin-3 was present in biotinylated BSA. Bound protein was detected with 1.8 ␮g of SA-(4,6- human tissues infected with C. albicans. As shown in Fig. 1, ga- dichlorotriazinyl) aminofluorescein (SA-DTAF) (Jackson ImmunoRe- lectin-3 was detected in lung tissue from a patient with systemic 4720 GALECTIN-3 INDUCES DEATH OF PATHOGENIC FUNGI

FIGURE 1. Galectin-3 expression in lung tissue from a patient with disseminated C. albicans infection. Serial sections were stained with Pe- riodic acid Schiff (A) to detect C. albicans, or anti-galectin-3 antiserum (B) and counterstained with hematoxylin. A, Periodic acid-Schiff staining dem-

onstrates fungal cells in the center of granulomatous inﬂammation in lung Downloaded from (left, ϫ100; right, ϫ400). B, Galectin-3 (red) was detected in the stroma surrounding the fungi (arrowhead, right panel), as well as in individual cells (arrows) in the tissue surrounding the granuloma (left, ϫ100; right, ϫ400). Inset, Galectin-3 expression in a macrophage (ϫ1000).

candidiasis. Galectin-3 was detected throughout granulomata in http://www.jimmunol.org/ the infected lung tissue, especially in the walls of fungal abscesses, as well as in macrophages and stromal cells in the granulomata and surrounding tissue, suggesting that galectin-3 could interact with C. albicans in infected tissues. In noninfected tissues, galectin-3 expression was only observed in macrophages, consistent with pre- vious work demonstrating expression of galectin-3 in the cytoplasm and on the surface of monocytes and macrophages (38). We observed no reactivity in tissues stained with control nonimmune serum (data not shown). by guest on September 30, 2021

Galectin-3 binds to C. albicans in a carbohydrate-specific manner As mentioned above, galectin-3 binds to ␤-1,2-oligomannans iso- FIGURE 2. Galectin-3 directly binds to C. albicans, and not to S. cer- lated from C. albicans (51). However, direct binding of galectin-3 evisiae in a carbohydrate-dependent manner. A, Biotinylated galectin-3 to intact yeast has not been examined. As a semiquantitative as- (center, right) or biotinylated BSA as a control (left) was added to heat sessment of galectin-3 binding, we examined galectin-3 binding to killed C. albicans and bound protein detected with streptavidin-HRP (red). fixed yeast by immunohistochemistry. A suspension of C. albicans Galectin-3 bound to C. albicans (center) and binding was inhibited by was air-dried and heat-fixed onto glass slides. Biotinylated BSA or lactose (right), demonstrating that binding was carbohydrate dependent. B, biotinylated galectin-3 was added to the cells, and bound protein Biotinylated galectin-3 (red) or BSA (gray) was added to C. albicans (left) was detected with streptavidin-HRP. As shown in Fig. 2A,we or S. cerevisiae (right) and bound protein detected by flow cytometry using detected significant galectin-3 binding to C. albicans, and galec- SA-DTAF. One representative from one of four independent experiments tin-3 binding was reduced in the presence of lactose, demonstrat- is shown. C, Biotinylated galectin-3 was bound to C. albicans in the absence (red) or presence (dotted line) of saccharide competitors and binding ing that galectin-3 binding to yeast was carbohydrate dependent. detected as above. Lactose (left), but not sucrose (right), inhibited galec- Direct binding of galectin-3 to live C. albicans was measured by tin-3 binding) to the level of background binding seen with biotinylated ␤ flow cytometric analysis. -1,2-oligomannans that have been de- BSA. Gray, Biotinylated BSA. One representative from three independent scribed as ligands for galectin-3 are present on C. albicans but experiments is shown. D, Biotinylated galectin-3 was added to C. albicans absent on S. cerevisiae (51). To assess specific binding of galec- in the absence (red) or presence (dotted line) of yeast cell wall mannans tin-3 to yeast containing ␤-1,2-oligomannans, we compared bind- and binding detected as above. C. albicans mannans that contain ␤-1,2- ing of biotinylated galectin-3 to C. albicans and S. cerevisiae. Ga- linked oligomannans inhibited binding (left), while S. cerevisiae mannans lectin-3 bound to C. albicans, but we detected no binding of that lack ␤-1,2-linked oligomannans had minimal inhibitory effect (right). galectin-3 to S. cerevisiae above the level of background binding Gray, Biotinylated BSA. Mean fluorescence intensity of histograms are of biotinylated BSA (Fig. 2B). Galectin-3 binding to C. albicans depicted below. One representative of two independent experiments is shown. was carbohydrate dependent, as binding could be competed to the level of background BSA binding with lactose, a disaccharide that binds galectin-3, but not with the nonspecific disaccharide inhib- itor sucrose (Fig. 2C). Moreover, preincubation of galectin-3 with C. albicans serotype A cell wall mannans significantly reduced (Fig. 2D). These observations indicate that galectin-3 preferen- galectin-3 binding to C. albicans cells, while S. cerevisiae cell wall tially binds to oligosaccharide ligands present in C. albicans man- mannans only minimally reduced galectin-3 binding to C. albicans nans that are not present in the cell wall mannans of S. cerevisiae. The Journal of Immunology 4721

Galectin-3 binds to Candida species expressing ␤-1,2-mannans Table 1. Surface expression of ␤-1,2-oligomannans F5, F6, and F9 on selected yeast There are three known types of ␤-1,2-linked oligomannans in the cell walls of different species of the genus Candida. These struc- Factor Factor Factor tures have been termed F5, F6, and F9 (Fig. 3A), and different Yeast Species 5a 6 9 Candida species express one or more of these structures (Table I). As mentioned above, galectin-3 bound to ␤-1,2-linked oligoman- C. albicans serotype A Yes Yes No C. albicans serotype B Yes No No nans isolated from C. albicans serotype A strain VW32 (51), that C. glabrata No Yes No contains both F5 and F6 types of ␤-1,2- linked oligomannans. To C. guilliermondii No No Yes investigate whether galectin-3 binding to cells required the pres- S. cerevisiae No No No ␤ ence of a specific type of -1,2-linked oligomannans, we assayed a Phenotypic analysis of yeast species with antisera specific for different ␤-1,2- galectin-3 binding to four different Candida species that display linked mannans designated factors 5, 6, and 9 (from Refs. 52–54). specific ␤-1,2-linked oligomannans on the cell surface. We first confirmed expression of different ␤-1,2-linked oligomannans on C. albicans serotypes A and B, C. guilliermondii, and liermondii is F5ϪF6Ϫ, and S. cerevisiae is F5ϪF6Ϫ, corresponding S. cerevisiae by flow cytometry using specific antisera to the F5 to the previously reported patterns of expression shown in Table I. and F6 mannan structures. As shown in Fig. 3B, C. albicans se- We examined binding of biotinylated galectin-3 or galectin-1 to rotype A is F5ϩF6ϩ, C. albicans serotype B is F5ϩF6Ϫ, C. guil- four different Candida species (Fig. 4). Galectin-3 bound to C. albicans serotype A and serotype B, C. glabrata and C. guilliermondii. Thus, galectin-3 bound to Candida species expressing sur- Downloaded from face F5, F6, or F9 ␤-1,2-linked oligomannans (Table I). In contrast, galectin-1 did not bind to any of the four Candida species, demonstrating that recognition of saccharides on the yeast cell wall is a specific function of galectin-3.

Galectin-3 induces C. albicans death http://www.jimmunol.org/ During binding analyses, we observed that binding of puriﬁed, recombinant galectin-3 resulted in distinct morphological changes in the cells. We observed a population of yeast in galectin-3- treated samples with reduced forward and increased side scatter by ﬂow cytometric analysis (Fig. 5A). The morphological changes resulting from addition of galectin-3 were inhibited in the presence of 0.1 M lactose, indicating that the effect was mediated by car-

bohydrate-dependent binding of galectin-3. We quantiﬁed viability by guest on September 30, 2021 in galectin-3 and control-treated cells, and found that there was a 38% reduction in CFU for galectin-3 vs control-treated cells (Fig. 5B). Moreover, the decrease in CFU for cells treated with galectin-3 was comparable to the fraction of cells that demonstrated morphologic changes after galectin-3 binding, ϳ30% in Fig. 5A. This morphologic change was not observed when cells were incubated with recombinant galectin-1, consistent with the lack of galectin-1 binding shown in Fig. 4 (data not shown). As the altered morphology of cells in galectin-3-treated samples was similar to the ﬂow cytometry characteristics of dying cells (decreased cell

FIGURE 3. Cell surface expression of ␤-1,2-linked oligomannans in FIGURE 4. Galectin-3, but not galectin-1, binds to Candida species Candida species. A, Schematic of ␤-1,2-linked oligomannans F5, F6, and bearing a variety of ␤-1,2-linked oligomannans. Biotinylated galectin-3 but F9 present in the cell wall of various Candida species. B, Different fungal not biotinylated galectin-1 bound to C. albicans serotype A, C. albicans species express different combinations of ␤-1,2-linked oligomannans. serotype B, C. glabrata, and C. guilliermondii. Biotinylated BSA was used Binding of speciﬁc rabbit antisera to F5 and F6 or preimmune rabbit serum as a negative control. Bound proteins were detected with streptavidin FITC (control) was detected with FITC-labeled goat anti-rabbit IgG and analyzed and experiments analyzed as indicated in Materials and Methods. One as in Materials and Methods. representative of three independent experiments is shown. 4722 GALECTIN-3 INDUCES DEATH OF PATHOGENIC FUNGI

We developed a high throughput method to measure galectin-3 induced fungal death using the dye Fun-1. Developed for fluores- cent microscopic analysis of yeast viability, Fun-1 initially produces green fluorescence in cells that take up the dye. We reasoned that we could discriminate live and dead cells by the intensity of Fun-1 fluorescence. To establish the assay, we analyzed Fun-1 staining of live yeast vs yeast killed by heat or Fungizone. As shown in Fig. 6A (left), Fungizone- or heat-treated dead cells were Fun-1dim, while live cells were Fun-1bright. As a control, a mixture of 50% live and 50% dead yeast demonstrated two peaks, a Fun- 1bright and a Fun-1dim, demonstrating that this Fun-1 assay could discriminate live and dead yeast populations by flow cytometry (Fig. 6A, right). We assessed yeast viability after galectin-3 binding using the Fun-1 assay. As shown in Fig. 6B, galectin-3 treatment of C. albicans resulted in a distinct Fun-1dim population that corresponded to the small, granular population detected by forward vs side scatter analysis of the same cells. Thus, as shown in Fig. 5, galectin-3

directly reduced the viability of the cells. Furthermore, we per- Downloaded from formed a dose-response analysis to determine the minimum concentration of galectin-3 required for the fungicidal effect. We observed minimal galectin-3 fungicidal activity at concentrations below 5 ␮M, while the maximal effect was observed at Ͼ20 ␮M (Fig. 7A). To conﬁrm that the fungicidal effect detected by Fun-1

uptake was carbohydrate dependent, we performed inhibition as- http://www.jimmunol.org/ says using lactose and sucrose, as in Fig. 2. In the presence of lactose, we observed Ͼ65% decrease in the Fun-1dim population compared with the samples treated with galectin-3 alone, while sucrose had no inhibitory effect (Fig. 7B).

Galectin-3-mediated Candida species death requires specific ␤- 1,2-linked oligomannans on the yeast cell surface As we observed galectin-3 binding to four different species of Can- dida expressing one or more of the three types of ␤-1,2-linked by guest on September 30, 2021 FIGURE 5. Galectin-3 induces C. albicans cell death. A, Galectin-3 oligomannans (Table I), we asked whether galectin-3 binding to binding to C. albicans results in distinct morphological changes. Flow cytometric analysis of C. albicans cells after galectin-3 binding revealed a population of cells (middle) with decreased forward scatter and increased side scatter, compared with buffer control (left). Galectin-3 induced morphologic changes were inhibited in the presence of 100 mM lactose (right), indicating a carbohydrate-dependent effect. B, Colony-forming assay for cells treated in the absence and presence of galectin-3. Cells were treated as described in Materials and Methods, and 10 ␮l plated at the indicated dilutions. Cells treated with galectin-3 (f) demonstrated decreased CFU compared with control cells (Ⅺ). One of seven replicate experiments is shown (range: 16–45% inhibition). C, Decreased viability of the small granular cells. C. albicans were treated with 30 ␮g of galectin-3 and the large and small cell populations were isolated by FACS. The cell number in each sample was adjusted to 1.8 ϫ 104 cells/ml and 10 ␮l of 1/10 serial dilutions were plated as indicated (bottom). The small, granular cells (right) yielded significantly reduced colony formation compared with the larger cells (left). FIGURE 6. The small C. albicans cells resulting from galectin-3 treatment are Fun-1dim. A, Left, Live cells are Fun-1bright and dead cells are size and increased cell granularity), we directly investigated the Fun-1dim. Cells were killed with heat (dotted line) or Fungizone (gray line) viability of the two populations of cells. We sorted the two distinct and the ability of live and dead cells to retain the vital dye Fun-1 was cell populations generated by galectin-3 treatment by flow cytom- analyzed by flow cytometry. Cells treated with buffer alone remained Fun- etry, and analyzed each population for the ability to grow on agar 1bright (black line), while dead cells were Fun-1dim. Right, A mixture of bright plates (Fig. 5C). The small, granular population of cells that ap- 50% live and 50% heat-killed yeast yielded discreet Fun-1 and Fun- dim peared after galectin-3 treatment demonstrated almost no growth, 1 peaks, demonstrating that the assay could discriminate live and dead cells in the same sample. One representative from 1 of 4 independent while the larger cells that were morphologically similar to un- experiments is shown. B, Left, Treatment of C. albicans with galectin-3 treated yeast demonstrated normal colony growth. Both sorted (black line), but not with buffer control (gray line), yielded a Fun-1 dim populations were plated at serial dilutions to rule out a fungistatic, population. (center). When galectin-3 treated cells were analyzed by for- rather than fungicidal, effect. As we observed minimal colony ward vs side scatter as in Fig. 5, the small nongranular cells (R1 gate) were growth of the small, granular cells at 1/10, 1/100, and 1/1000 di- Fun-1dim (right). One representative from 1 of 14 independent experiments lutions (Fig. 5C), the galectin-3 effect appeared to be fungicidal. is shown. The Journal of Immunology 4723

FIGURE 8. Galectin-3-mediated Candida cell death requires specific ␤-1,2-oligomannosides on the yeast cell surface. C. albicans serotype A (F5ϩF6ϩ), C. albicans serotype B (F5ϩF6Ϫ), C. glabrata (F5ϪF6ϩ), C. guilliermondii (F5ϪF6Ϫ)orS. cerevisiae were treated with galectin-3 (open histogram) or buffer control (filled histogram) at 37°C and stained Downloaded from with Fun-1. A significant fraction of C. albicans serotype A, C. albicans serotype B and C. glabrata cells became Fun-1dim, while galectin-3 treatment of C. guilliermondii and S. cerevisiae did not demonstrate an increase in the percent Fun-1dim cells. One representative from two independent experiments is shown.

FIGURE 7. Specificity of galectin-3 induced C. albicans cell death. A, http://www.jimmunol.org/ Galectin-3 induced C. albicans cell death is dose dependent. Cells were ␤ treated with the indicated concentration of galectin-3 at 37°C and stained more, galectin-3 binds to -1,2-oligomannosides isolated from the with Fun-1. The mean percent Ϯ SD. Fun-1dim cells from triplicate samples cell wall of C. albicans serotype A (51), and synthetic analogs of is shown on the y-axis. A small increase in the Fun-1dim population was ␤-1,2-oligomannosides administered orally to mice prevented col- observed at 5 ␮M galectin-3, while maximal cell death was observed at onization of the intestine by C. albicans (56). However, the effect Ͼ20 ␮M. B, Galectin-3-mediated cell death of C. albicans is carbohydrate of galectin-3 binding to intact yeast was not explored, and no direct dependent. Cells treated with the highest concentration of galectin-3 in A microbicidal activity for galectin-3 has been described. dim dim (80 mM), and percent Fun-1 cells determined. The percent Fun-1 We found that galectin-3 bound to four Candida species ex- cells was reduced in the presence of lactose, but not sucrose. The mean pressing different combinations of ␤-1,2-oligomannans, but did not by guest on September 30, 2021 percent Ϯ SD Fun-1dim cells from triplicate samples is shown on the y-axis. bind S. cerevisiae that lacks ␤-1,2-oligomannans. Consistent with the binding specificity, binding of galectin-3 to C. albicans was inhibited by C. albicans cell wall mannans but not by S. cerevisiae the yeast cell surface was sufficient to induce death, or whether a cell wall mannans. Thus, galectin-3 is a PRR that recognizes spe- specific type of ␤-1,2-linked oligomannan was required for galec- cific oligosaccharides on various Candida species. tin-3 to kill the cells. C. albicans serotypes A and B, C. glabrata, Surprisingly, we found that galectin-3 binding to C. albicans and C. guilliermondii were treated with galectin-3 and cell viabil- was directly fungicidal. We observed a unique population of cells ity assessed by Fun-1 uptake. As shown in Fig. 8, flow cytometric with morphologic characteristics of dead or dying cells after treat- analysis revealed a Fun-1dim population in galectin-3-treated C. ment with galectin-3, and the direct fungicidal activity of galec- ϩ ϩ Ϫ albicans serotype A (F5 F6 F9 ), C. albicans serotype B tin-3 was confirmed by demonstrating markedly reduced colony ϩ Ϫ Ϫ Ϫ ϩ Ϫ (F5 F6 F9 ), and C. glabrata (F5 F6 F9 ). However, like S. forming capability of this population. Although lectins can partic- Ϫ Ϫ ϩ cerevisiae, galectin-3-treated C. guilliermondii (F5 F6 F9 ) did ipate in the host defense against fungal pathogens by promoting not have a significant Fun-1dim population (Fig. 8). Thus, the fun- opsonization by complement and/or fungal clearance by host mac- gicidal effect of galectin-3 appeared to be specific for Candida rophages, the fungicidal activity of galectin-3 is complement in- species bearing cell surface F5 or F6 ␤-1,2-linked oligomannans, dependent, as we observed the effect with purified recombinant while expression of F9 ␤-1,2-linked oligomannans in the absence human galectin-3 in the absence of serum. And, although galec- of F5 or F6, although sufficient for galectin-3 binding, was not tin-3 is known to participate in phagocytosis of microbial PAMP- sufficient to render the cells susceptible to galectin-3-mediated coated beads (29), we observed the fungicidal effect of galectin-3 death. in the absence of macrophages or other cells of the mammalian immune system. The mechanism of galectin-3 fungal death is not Discussion clear at this point; however, recent studies of oligosaccharide- Infections with opportunistic fungi are a significant concern, es- based vaccines for C. albicans demonstrated a direct fungicidal pecially in immunocompromised patients. Thus, identification of activity for Abs that recognize ␤-glucans in the yeast cell wall novel antifungal activities against C. albicans will allow develop- (57). These observations suggest that cross-linking of oligosaccha- ment of new therapeutic strategies to combat fungal infections. ride components in the fungal cell wall, either by multivalent ga- Galectin-3 has been suggested to act as a PRR for several micro- lectins such as galectin-3 or by Abs, can directly trigger death. bial pathogens (26, 27), including the yeast C. albicans (51). Al- Although we consistently observed galectin-3-induced death of though galectins were originally defined by the ability to bind to C. albicans, we never observed morphologic changes or death of ␤-galactosides, some members of the galectin family, e.g., galec- the entire population of cells, although all the cells bound galectin-10 (42), can recognize mannose with high affinity. Further- tin-3. In all assays we performed, we detected death of ϳ30–70% 4724 GALECTIN-3 INDUCES DEATH OF PATHOGENIC FUNGI of cells (Figs. 7 and 8). Thus, the reduction in CFU was most flammatory cells in human tissues (58), and galectin binding to apparent when we sorted the two morphologically distinct popu- extracellular matrix proteins has been shown to concentrate galec- lations (Fig. 5). We determined that incomplete susceptibility was tin (59). Moreover, total galectin-3 concentration in homogenized not due to gross variation in cell cycle in an asynchronous popu- splenic tissue from various mammals can reach 80 mg/kg tissue, or lation, as all assays were performed on cells 24 h after dilution roughly 2 ␮M (66). Thus, in infected tissues, the local concentra- from stationary cultures (see Materials and Methods). We also tion of galectin-3 surrounding macrophages and stromal cells asked whether longer exposure to galectin-3 would increase the could be quite high. Our observation of galectin-3 expression in fraction of susceptible cells. In time course experiments, we ob- granulomata in lung tissue from a patient with systemic candidiasis served negligible death after galectin-3 binding at 30 min, with (Fig. 1) supports a role for galectin-3 in host protection against dead cells becoming apparent at 60–90 min and maximal death Candida infection. Candida species are the fourth leading cause of observed after 120 min binding of galectin-3 (data not shown). At nosocomial bloodstream infections, with a mortality rate of 40% in longer time points (4 and 6 h) the effects of galectin-3 were con- the United States (67), and vulvovaginal candidiasis affects 75% of founded by proliferation of the surviving cells (data not shown), so women of childbearing age (68). Thus, understanding the contri- we limited the assays to 2 h. Intriguingly, in studies of galectin-3- bution of human lectins such as galectin-3 to resistance to Candida and galectin-1-induced death of mammalian lymphocytes and lym- may contribute to control of infections in both immunocompetent phoid cell lines, 100% cell death is also not observed, although all and immunocompromised patients. Intriguingly, a recent report cells in the populations examined bind the galectins (13, 58, 59). demonstrated that C. albicans infection of a macrophage cell line As glycosylation is not templated, and factors such as residence in vitro reduced expression of galectin-3 ϳ3-fold, suggesting a time in the Golgi apparatus will affect the type and abundance of mechanism for the pathogen to evade the fungicidal effect of ga- Downloaded from glycans added to a glycoprotein (60), subtle differences in glyco- lectin-3 (69). However, increased galectin-3 production by acti- sylation or density of critical ligands on the surface of cells may vated but noninfected macrophages and stromal cells may contrib- regulate the susceptibility of the cells to galectin-3. The mecha- ute to control of Candida infections in vivo. nism of galectin-3-mediated fungal death is currently being inves- Antimicrobial activity may be a common feature of galectins in tigated in our laboratories. different species. AJL-1 is a member of the galectin family of

Although galectin-3, but not galectin-1, bound to all Candida lectins expressed in the skin mucus of the Japanese eel Anguilla http://www.jimmunol.org/ species that contained one or more types of ␤-1,2-linked oligo- japonica. High expression of AJL-1 correlates with resistance to mannans, galectin-3 killing of Candida appeared to require the infection in the eel. Furthermore, AJL-1 has a carbohydrate- and presence of the specific ␤-1,2-linked oligomannans F5 or F6 (Figs. species-specific agglutinating activity for the pathogenic Gram- 3A and 8). In contrast, the ␤-1,2-linked oligomannan F9 did not positive bacteria Streptococcus difficile (70), suggesting that ag- appear to be sufficient to mediate galectin-3 induced death, as C. glutination of invading bacteria on the eel skin surface traps the guilliermondii (F5ϪF6ϪF9ϩ) bound galectin-3 but did not die pathogen. As mentioned above, mammalian galectin-3 also inter- (Figs. 4 and 8). Thus, direct fungicidal activity of galectin-3 may acts with L. major (22, 23) and T. cruzi (30, 71), as well as bac- be restricted to specific Candida species. It is of interest to note terial and mycobacterial pathogens (24–28). Thus, galectin-3 may that the F5 and F6 ␤-1,2-linked oligomannan epitopes are both on have multiple functions in human host protection against microbial by guest on September 30, 2021 the same phosphomannan-containing branch structure, while the pathogens. Importantly, recent evidence demonstrates that a robust F9 ␤-1,2-linked oligomannan epitope is on a different N-glycan response to fungal pathogens involves several host lectin PRRs, structure (Fig. 3A); this suggests that the phosphomannan-contain- including the macrophage mannose receptor and dectin-1, recog- ing oligosaccharides on specific Candida species may be important nizing different glycans on the fungal cell surface (72). Galectin-3, for galectin-3 death. Gow and colleagues (61) recently reported specifically recognizing ␤-1,2-linked mannans, may be an addi- that outer chain N-glycans, such as those to which the F5 and F6 tional lectin PRR that contributes a unique, direct fungicidal ac- epitopes can be attached, are essential for cell wall integrity of C. tivity to the innate immune defense against Candida infection. albicans. Induction of Candida death is a unique function of galectin-3, Acknowledgments because galectin-1 did not bind to any of the Candida species We thank Dr. Elizabeth Wagar, Dr. Joseph Hernandez, and Esteban Fer- examined (Fig. 4), nor did galectin-1 have a fungicidal effect on C. nandez for helpful discussions, and Mabel Pang for technical assistance. albicans serotype B at concentrations up to 35 ␮M (data not We also thank Michael Gulrajani and the staff of the UCLA Flow Cytom- shown). Galectin-3 binding and induction of Candida death are etry Core Facility at the Jonsson Comprehensive Cancer Center (UCLA carbohydrate specific, as both functions were blocked by the dis- CFAR CA-16042). accharide ligand lactose, but not by the disaccharide sucrose that is not recognized by galectin-3 (Figs. 2A,2C,5A, and 7B). Further- Disclosures more, galectin-3 binding to C. albicans was markedly inhibited by The authors have no financial conflict of interest. C. albicans mannans, with little inhibition by S. cerevisiae mannans (Fig. 2D), demonstrating that galectin-3 specifically binds to References ligands present in the cell wall of Candida species but absent on S. 1. Gantner, B. N., R. M. Simmons, S. J. Canavera, S. Akira, and D. M. Underhill. 2003. Collaborative induction of inflammatory responses by dectin-1 and Toll- cerevisiae. Our results indicate that galectin-3 binding and induc- like receptor 2. J. Exp. Med. 197: 1107–1117. tion of death of Candida species is a specific, lectin-mediated ac- 2. Herre, J., J. A. Willment, S. Gordon, and G. D. Brown. 2004. The role of Dectin-1 tivity of galectin-3. in antifungal immunity. Crit. Rev. Immunol. 24: 193–203. 3. Janeway, C. A., Jr., and R. Medzhitov. 2002. Innate immune recognition. Annu. Galectin-3 is expressed by epithelia lining the body surfaces that Rev. Immunol. 20: 197–206. are exposed to Candida colonization, such as oral mucosa, corneal, 4. Medzhitov, R., and C. Janeway, Jr. 2000. Innate immunity. N. Engl. J. Med. 343: 338–344. and conjunctival epithelia, and intestinal epithelia (62–65), that 5. Medzhitov, R., and C. A. Janeway, Jr. 2002. Decoding the patterns of self and constitute the first line of host defense against Candida invasion. nonself by the innate immune system. Science 296: 298–300. Although relatively high concentrations of soluble galectin-3 (20 6. Linehan, S. A., L. Martinez-Pomares, and S. Gordon. 2000. Macrophage lectins ␮ in host defence. Microbes Infect. 2: 279–288. M) were required to trigger Candida death in vitro, abundant 7. al-Ramadi, B. K., M. Ellis, F. Pasqualini, and A. Mantovani. 2004. Selective galectin-3 has been detected in the local milieu surrounding in- induction of pentraxin 3, a soluble innate immune pattern recognition receptor, in The Journal of Immunology 4725

infectious episodes in patients with haematological malignancy. Clin. Immunol. modulation of adhesion and migration of murine dendritic cells. Glycobiology 14: 112: 221–224. 647–657. 8. Garlanda, C., E. Hirsch, S. Bozza, A. Salustri, M. De Acetis, R. Nota, 33. Lim, J. W., H. Kim, and K. H. Kim. 2003. Cell adhesion-related gene expression A. Maccagno, F. Riva, B. Bottazzi, G. Peri, et al. 2002. Non-redundant role of the by Helicobacter pylori in gastric epithelial AGS cells. Int. J. Biochem. Cell Biol. long pentraxin PTX3 in anti-fungal innate immune response. Nature 420: 35: 1284–1296. 182–186. 34. Joo, H. G., P. S. Goedegebuure, N. Sadanaga, M. Nagoshi, W. von Bernstorff, 9. Willment, J. A., S. Gordon, and G. D. Brown. 2001. Characterization of the and T. J. Eberlein. 2001. Expression and function of galectin-3, a ␤-galactoside- human ␤-glucan receptor and its alternatively spliced isoforms. J. Biol. Chem. binding protein in activated T lymphocytes. J. Leukocyte Biol. 69: 555–564. 276: 43818–43823. 35. Acosta-Rodriguez, E. V., C. L. Montes, C. C. Motran, E. I. Zuniga, F. T. Liu, 10. Brown, G. D., and S. Gordon. 2001. Immune recognition: a new receptor for G. A. Rabinovich, and A. Gruppi. 2004. Galectin-3 mediates IL-4-induced sur- ␤-glucans. Nature 413: 36–37. vival and differentiation of B cells: functional cross-talk and implications during 11. McCormack, F. X., R. Gibbons, S. R. Ward, A. Kuzmenko, H. Wu, and Trypanosoma cruzi infection. J. Immunol. 172: 493–502. G. S. Deepe, Jr. 2003. Macrophage-independent fungicidal action of the pulmo- 36. Lee, V. H., A. B. Lee, E. B. Phillips, J. K. Roberts, and H. M. Weitlauf. 1998. nary collectins. J. Biol. Chem. 278: 36250–36256. Spatio-temporal pattern for expression of galectin-3 in the murine utero-placental 12. Rabinovich, G. A., and A. Gruppi. 2005. Galectins as immunoregulators during complex: evidence for differential regulation. Biol. Reprod. 58: 1277–1282. infectious processes: from microbial invasion to the resolution of the disease. 37. Dietz, A. B., P. A. Bulur, G. J. Knutson, R. Matasic, and S. Vuk-Pavlovic. 2000. Parasite Immunol. 27: 103–114. Maturation of human monocyte-derived dendritic cells studied by microarray 13. Hernandez, J. D., and L. G. Baum. 2002. Ah, sweet mystery of death: galectins hybridization. Biochem. Biophys. Res. Commun. 275: 731–738. and control of cell fate. Glycobiology 12: 127R–136R. 38. Liu, F. T., D. K. Hsu, R. I. Zuberi, I. Kuwabara, E. Y. Chi, and W. R. Henderson, 14. Hirabayashi, J., S. K. Dutta, and K. Kasai. 1998. Novel galactose-binding pro- Jr. 1995. Expression and function of galectin-3, a ␤-galactoside-binding lectin, in teins in Annelida: characterization of 29-kDa tandem repeat-type lectins from the human monocytes and macrophages. Am. J. Pathol. 147: 1016–1028. earthworm Lumbricus terrestris. J. Biol. Chem. 273: 14450–14460. 39. Cooper, D. N., and S. H. Barondes. 1999. God must love galectins; he made so 15. Ahmed, H., M. A. Bianchet, L. M. Amzel, J. Hirabayashi, K. Kasai, many of them. Glycobiology 9: 979–984. Y. Giga-Hama, H. Tohda, and G. R. Vasta. 2002. Novel carbohydrate specificity 40. Rini, J. M., and Y. D. Lobsanov. 1999. New animal lectin structures. Curr. Opin. of the 16-kDa galectin from Caenorhabditis elegans: binding to blood group Struct. Biol. 9: 578–584. Downloaded from ␣ ␤ precursor oligosaccharides (type 1, type 2, T , and T ) and gangliosides. Gly- 41. Krzeslak, A., and A. Lipinska. 2004. Galectin-3 as a multifunctional protein. Cell. cobiology 12: 451–461. Mol. Biol. Lett. 9: 305–328. 16. Buck, F., C. Schulze, M. Breloer, K. Strupat, and H. Bretting. 1998. Amino acid 42. Leffler, H., S. Carlsson, M. Hedlund, Y. Qian, and F. Poirier. 2004. Introduction sequence of the D-galactose binding lectin II from the sponge Axinella polypoides to galectins. Glycoconj. J. 19: 433–440. (Schmidt) and identification of the carbohydrate binding site in lectin II and 43. Ahmad, N., H. J. Gabius, S. Andre, H. Kaltner, S. Sabesan, R. Roy, B. Liu, related lectin I. Comp. Biochem. Physiol. Mol. Biol. 121: 153–160. F. Macaluso, and C. F. Brewer. 2004. Galectin-3 precipitates as a pentamer with 17. Cooper, D. N., R. P. Boulianne, S. Charlton, E. M. Farrell, A. Sucher, and synthetic multivalent carbohydrates and forms heterogeneous cross-linked com- B. C. Lu. 1997. Fungal galectins, sequence and specificity of two isolectins from plexes. J. Biol. Chem. 279: 10841–10847. Coprinus cinereus. J. Biol. Chem. 272: 1514–1521. http://www.jimmunol.org/ 44. Kim, K., E. P. Mayer, and M. Nachtigal. 2003. Galectin-3 expression in macro- 18. Pace, K. E., T. Lebestky, T. Hummel, P. Arnoux, K. Kwan, and L. G. Baum. phages is signaled by Ras/MAP kinase pathway and up-regulated by modified 2002. Characterization of a novel Drosophila melanogaster galectin: expression lipoproteins. Biochim. Biophys. Acta 1641: 13–23. in developing immune, neural, and muscle tissues. J. Biol. Chem. 277: 45. Henderson, N. C., A. C. Mackinnon, S. L. Farnworth, F. Poirier, F. P. Russo, 13091–13098. J. P. Iredale, C. Haslett, K. J. Simpson, and T. Sethi. 2006. Galectin-3 regulates 19. Pace, K. E., and L. G. Baum. 2004. Insect galectins: roles in immunity and myofibroblast activation and hepatic fibrosis. Proc. Natl. Acad. Sci. USA 103: development. Glycoconj. J. 19: 607–614. 5060–5065. 20. Levroney, E. L., H. C. Aguilar, J. A. Fulcher, L. Kohatsu, K. E. Pace, M. Pang, 46. Neidhart, M., F. Zaucke, R. von Knoch, A. Jungel, B. A. Michel, R. E. Gay, and K. B. Gurney, L. G. Baum, and B. Lee. 2005. Novel innate immune functions for S. Gay. 2005. Galectin-3 is induced in rheumatoid arthritis synovial fibroblasts galectin-1: galectin-1 inhibits cell fusion by Nipah virus envelope glycoproteins after adhesion to cartilage oligomeric matrix. Ann. Rheum. Dis. 64: 419–424. and augments dendritic cell secretion of proinflammatory cytokines. J. Immunol. 47. Davidson, P. J., M. J. Davis, R. J. Patterson, M. A. Ripoche, F. Poirier, and 175: 413–420.

J. L. Wang. 2002. Shuttling of galectin-3 between the nucleus and cytoplasm. by guest on September 30, 2021 21. Ouellet, M., S. Mercier, I. Pelletier, S. Bounou, J. Roy, J. Hirabayashi, S. Sato, Glycobiology 12: 329–337. and M. J. Tremblay. 2005. Galectin-1 acts as a soluble host factor that promotes HIV-1 infectivity through stabilization of virus attachment to host cells. J. Im- 48. Yu, F., R. L. Finley Jr., A. Raz, and H. R. Kim. 2002. Galectin-3 translocates to munol. 174: 4120–4126. the perinuclear membranes and inhibits cytochrome c release from the mitochon- dria: a role for synexin in galectin-3 translocation. J. Biol. Chem. 277: 22. Pelletier, I., T. Hashidate, T. Urashima, N. Nishi, T. Nakamura, M. Futai, 15819–15827. Y. Arata, K. Kasai, M. Hirashima, J. Hirabayashi, and S. Sato. 2003. Specific recognition of Leishmania major poly-␤-galactosyl epitopes by galectin-9: pos- 49. Yang, R. Y., D. K. Hsu, and F. T. Liu. 1996. Expression of galectin-3 modulates sible implication of galectin-9 in interaction between L. major and host cells. T-cell growth and apoptosis. Proc. Natl. Acad. Sci. USA 93: 6737–6742. J. Biol. Chem. 278: 22223–22230. 50. Hughes, R. C. 1999. Secretion of the galectin family of mammalian carbohydrate- 23. Pelletier, I., and S. Sato. 2002. Specific recognition and cleavage of galectin-3 by binding proteins. Biochim. Biophys. Acta 1473: 172–185. ␤ Leishmania major through species-specific polygalactose epitope. J. Biol. Chem. 51. Fradin, C., D. Poulain, and T. Jouault. 2000. -1,2-linked oligomannosides from 277: 17663–17670. Candida albicans bind to a 32-kilodalton macrophage membrane protein homol- 24. Barboni, E., S. Coade, and A. Fiori. 2005. The binding of mycolic acids to ogous to the mammalian lectin galectin-3. Infect. Immun. 68: 4391–4398. galectin-3: a novel interaction between a host soluble lectin and trafficking my- 52. Shibata, N., R. Akagi, T. Hosoya, K. Kawahara, A. Suzuki, K. Ikuta, cobacterial lipids? FEBS Lett. 579: 6749–6755. H. Kobayashi, K. Hisamichi, Y. Okawa, and S. Suzuki. 1996. Existence of novel ␤ ␣ 25. Beatty, W. L., E. R. Rhoades, D. K. Hsu, F. T. Liu, and D. G. Russell. 2002. branched side chains containing -1,2 and -1,6 linkages corresponding to an- Association of a macrophage galactoside-binding protein with Mycobacterium- tigenic factor 9 in the mannan of Candida guilliermondii. J. Biol. Chem. 271: containing phagosomes. Cell. Microbiol. 4: 167–176. 9259–9266. 26. Gupta, S. K., S. Masinick, M. Garrett, and L. D. Hazlett. 1997. Pseudomonas 53. Shibata, N., M. Arai, E. Haga, T. Kikuchi, M. Najima, T. Satoh, H. Kobayashi, aeruginosa lipopolysaccharide binds galectin-3 and other human corneal epithe- and S. Suzuki. 1992. Structural identification of an epitope of antigenic factor 5 lial proteins. Infect. Immun. 65: 2747–2753. in mannans of Candida albicans NIH B-792 (serotype B) and J-1012 (serotype ␤ 27. Mey, A., H. Leffler, Z. Hmama, G. Normier, and J. P. Revillard. 1996. The animal A) as -1,2-linked oligomannosyl residues. Infect. Immun. 60: 4100–4110. lectin galectin-3 interacts with bacterial lipopolysaccharides via two independent 54. Mercure, S., S. Senechal, P. Auger, G. Lemay, and S. Montplaisir. 1996. Candida sites. J. Immunol. 156: 1572–1577. albicans serotype analysis by flow cytometry. J. Clin. Microbiol. 34: 2106–2112. 28. John, C. M., G. A. Jarvis, K. V. Swanson, H. Leffler, M. D. Cooper, M. E. Huflejt, 55. Hsu, D. K., R. I. Zuberi, and F. T. Liu. 1992. Biochemical and biophysical and J. M. Griffiss. 2002. Galectin-3 binds lactosaminylated lipooligosaccharides characterization of human recombinant IgE-binding protein, an S-type animal from Neisseria gonorrhoeae and is selectively expressed by mucosal epithelial lectin. J. Biol. Chem. 267: 14167–14174. cells that are infected. Cell. Microbiol. 4: 649–662. 56. Dromer, F., R. Chevalier, B. Sendid, L. Improvisi, T. Jouault, R. Robert, 29. van den Berg, T. K., H. Honing, N. Franke, A. van Remoortere, W. E. Schiphorst, J. M. Mallet, and D. Poulain. 2002. Synthetic analogues of ␤-1,2-oligomanno- F. T. Liu, A. M. Deelder, R. D. Cummings, C. H. Hokke, and I. van Die. 2004. sides prevent intestinal colonization by the pathogenic yeast Candida albicans. LacdiNAc-glycans constitute a parasite pattern for galectin-3-mediated immune Antimicrob. Agents Chemother. 46: 3869–3876. recognition. J. Immunol. 173: 1902–1907. 57. Torosantucci, A., C. Bromuro, P. Chiani, F. De Bernardis, F. Berti, C. Galli, 30. Kleshchenko, Y. Y., T. N. Moody, V. A. Furtak, J. Ochieng, M. F. Lima, and F. Norelli, C. Bellucci, L. Polonelli, P. Costantino, et al. 2005. A novel glyco- F. Villalta. 2004. Human galectin-3 promotes Trypanosoma cruzi adhesion to conjugate vaccine against fungal pathogens. J. Exp. Med. 202: 597–606. human coronary artery smooth muscle cells. Infect. Immun. 72: 6717–6721. 58. Stillman, B. N., D. K. Hsu, M. Pang, C. F. Brewer, P. Johnson, F. T. Liu, and 31. Zuniga, E., A. Gruppi, J. Hirabayashi, K. I. Kasai, and G. A. Rabinovich. 2001. L. G. Baum. 2006. Galectin-3 and galectin-1 bind distinct cell surface glyco- Regulated expression and effect of galectin-1 on Trypanosoma cruzi-infected protein receptors to induce T cell death. J. Immunol. 176: 778–789. macrophages: modulation of microbicidal activity and survival. Infect. Immun. 59. He, J., and L. G. Baum. 2004. Presentation of galectin-1 by extracellular matrix 69: 6804–6812. triggers T cell death. J. Biol. Chem. 279: 4705–4712. 32. Vray, B., I. Camby, V. Vercruysse, T. Mijatovic, N. V. Bovin, 60. Granffenried, C. L., and C. R. Bertozzi. 2004. The roles of enzyme localization P. Ricciardi-Castagnoli, H. Kaltner, I. Salmon, H. J. Gabius, and R. Kiss. 2004. and complex formation in glycan assembly within the Golgi apparatus. Curr. Up-regulation of galectin-3 and its ligands by Trypanosoma cruzi infection with Opin. Cell Biol. 16: 356–363. 4726 GALECTIN-3 INDUCES DEATH OF PATHOGENIC FUNGI

61. Bates, S., H. B. Hughes, C. A. Munro, W. P. Thomas, D. M. MacCallum, 66. Allen, H. K., M. Cywinski, R. Palmberg, and R. A. Di Cioccio. 1987. Compar- G. Bertram, A. Atrih, M. A. Ferguson, A. J. Brown, F. C. Odds, and N. A. Gow. ative analysis of galactoside-binding lectins from mammalian spleens. Arch. Bio- 2006. Outer chain N-glycans are required for cell wall integrity and virulence of chem. Biophys. 256: 523–533. Candida albicans. J. Biol. Chem. 281: 90–98. 67. Wenzel, R. P. 2005. John E. Bennett Forum on Deep Mycoses Study Design 62. Holikova, Z., E. Hrdlickova-Cela, J. Plzak, K. Smetana, Jr., J. Betka, 2004: candidiasis and salvage therapy for aspergillosis. Clin. Infect. Dis. B. Dvorankova, M. Esner, K. Wasano, S. Andre, H. Kaltner, et al. 2002. Defining 41(Suppl. 6): S369–S370. the glycophenotype of squamous epithelia using plant and mammalian lectins: differentiation-dependent expression of ␣2,6- and ␣2,3-linked N-acetylneura- 68. Sobel, J. D. 1992. Pathogenesis and treatment of recurrent vulvovaginal candi- minic acid in squamous epithelia and carcinomas, and its differential effect on diasis. Clin. Infect. Dis. 14(Suppl. 1): S148–S153. binding of the endogenous lectins galectins-1 and -3. APMIS 110: 845–856. 69. Shin, Y. K., H. J. Lee, J. S. Lee, and Y. K. Paik. 2006. Proteomic analysis of 63. Hrdlickova-Cela, E., J. Plzak, K. Smetana, Jr., Z. Melkova, H. Kaltner, mammalian basic proteins by liquid-based two-dimensional column chromatog- M. Filipec, F. T. Liu, and H. J. Gabius. 2001. Detection of galectin-3 in tear fluid raphy. Proteomics 6: 1143–1150. at disease states and immunohistochemical and lectin histochemical analysis in 70. Tasumi, S., W. J. Yang, T. Usami, S. Tsutsui, T. Ohira, I. Kawazoe, human corneal and conjunctival epithelium. Br. J. Ophthalmol. 85: 1336–1340. M. N. Wilder, K. Aida, and Y. Suzuki. 2004. Characteristics and primary struc- 64. Honjo, Y., H. Inohara, S. Akahani, T. Yoshii, Y. Takenaka, J. Yoshida, ture of a galectin in the skin mucus of the Japanese eel, Anguilla japonica. Dev. K. Hattori, Y. Tomiyama, A. Raz, and T. Kubo. 2000. Expression of cytoplasmic Comp. Immunol. 28: 325–335. galectin-3 as a prognostic marker in tongue carcinoma. Clin. Cancer Res. 6: 4635–4640. 71. Moody, T. N., J. Ochieng, and F. Villalta. 2000. Novel mechanism that Trypano- 65. Jensen-Jarolim, E., R. Gscheidlinger, G. Oberhuber, C. Neuchrist, T. Lucas, soma cruzi uses to adhere to the extracellular matrix mediated by human galec- G. Bises, C. Radauer, M. Willheim, O. Scheiner, F. T. Liu, and tin-3. FEBS Lett. 470: 305–308. G. Boltz-Nitulescu. 2002. The constitutive expression of galectin-3 is downregu- 72. Netea, M. G., N. A. R. Gow, C. A. Munro, S. Bates, C. Collins, G. Ferwerda, lated in the intestinal epithelia of Crohn’s disease patients, and tumour necrosis R. P. Hobson, G. Bertram, H. B. Hughes, T. Jansen, et al. 2006. Immune sensing factor ␣ decreases the level of galectin-3-specific mRNA in HCT-8 cells. Eur. of Candida albicans requires cooperative recognition of mannans and glucans by J. Gastroenterol. Hepatol. 14: 145–152. lectin and Toll-like receptors. J. Clin. Invest. 116: 1642–50. Downloaded from http://www.jimmunol.org/ by guest on September 30, 2021