The Library
The Ultimate On-line Candida Resource
Glossary    Contact Us
 
   
Browse by Category
Your Are Here >> Library Home > Candida In The Body > Immune System > Neutrophils > Interaction of the Fungal Pathogen Candida albicans with Integrin CD11b/CD18:

Back

We'd love to hear from you - the Candida Library is for you. Let us know if you have any suggestions, ideas or feedback. Click here to email us.

Interaction of the Fungal Pathogen Candida albicans with Integrin CD11b/CD18:

Interaction of the Fungal Pathogen Candida albicans with Integrin CD11b/CD18:


Christopher B. Forsyth, Edward F. Plow and Li Zhang


"Interactions of microorganisms with integrins are central to the host defense mechanisms
. The leukocyte integrin CD11b/CD18 is the principal adhesion receptor on leukocytes for Candida albicans, a major opportunistic pathogen. In this study we have investigated the roles of three regions within the receptor, the inserted (I) and lectin-like domains within the CD11b subunit, and the CD18 subunit, in CD11b/CD18-C. albicans interactions. We report four major findings. 1) A mutation in CD18 exerts a dominant negative effect on the function of the CD11b/CD18 complex. This interpretation is based on the observation that in the absence of CD18, the CD11b subunit alone binds C. albicans well, but a single point mutation at Ser138 of CD18 abolishes CD11b/CD18 binding of the fungus. 2) The lectin-like domain is not sufficient for CD11b/CD18-C. albicans interactions. Rather, the lectin-like domain appears to influence CD11b/CD18 binding activity by modulating the function of the I domain. 3) The I domain is the primary binding site for C. albicans in the receptor and is sufficient to support an efficient interaction. 4) We have identified specific amino acid sequences within the I domain that engage the microorganism. Compared with other ligands of CD11b/CD18, C. albicans has some unique as well as common contact sites within the I domain of the receptor. Such unique contact sites may underlie the ability of C. albicans to modulate CD11b/CD18 function and raise the possibility for selective interference of the microorganism-host leukocyte interactions."

Pathogenic organisms use a variety of cell surface receptors to gain entry into host cells and to bypass the natural defense mechanisms, such as the oxidative burst (1, 2). One of the most prominent receptors used in this fashion is the leukocyte adhesion receptor CD11b/CD18. This integrin is the primary leukocyte receptor for a number of pathogens, including Histoplasma capsulatum (3), Blastomyces dermatitidis (4), Mycobacterium tuberculosis (5), Leishmania sp. (6), Bordetella pertussis (2), and Candida albicans. CD11b/CD18 is the principal adhesion receptor for C. albicans on neutrophils, macrophages, and lymphocytes (7, 8), and this interaction leads to suppression of the immune response to the microorganism (7). As the major cause of hospital-acquired fungal infections (9), the recognition of C. albicans by CD11b/CD18 has major pathogenetic consequences. Nevertheless, only limited information is available on the molecular mechanisms involved in recognition of this fungus by the receptor. The primary goal of this study was to identify the critical segments within CD11b/CD18 that compose the binding pocket for C. albicans.

Integrins are a large family of {alpha}/ heterodimeric adhesion receptors and can be divided into subfamilies according to their common subunits. CD11b/CD18 ({alpha}M2, Mac-1, CR3) is a member of the CD18 subfamily of integrins, which also contains CD11a/CD18 ({alpha}L2, LFA-1), CD11c/CD18 ({alpha}X2), and a recently discovered fourth member, {alpha}D/CD18({alpha}D2) (10). CD11b/CD18 is found on all major classes of leukocytes, including virtually all neutrophils and macrophages as well as subsets of NK cells, T cells, and B cells (3, 10). CD11b/CD18 has been shown to play key and often the principal role in leukocyte adhesion, extravasation, migration, phagocytosis, and degranulation-cytotoxicity (3). Two domains have been implicated in the binding functions of CD11b/CD18 (3). The integrin recognizes protein ligands via a segment of about 200 amino acids in the CD11b subunit, termed the I domain (10). CD11b/CD18 recognizes mannose and -glucan carbohydrate structures by a lectin-like ligand binding domain also in the CD11b ({alpha}) subunit (11). I domain ligands include ICAM-1, C3bi, fibrinogen (Fg),3 and the helminth ligand neutrophil inhibitory factor (NIF) (12, 13, 14, 15). The lectin domain ligands include zymosan, Saccharomyces cereviseae, and nonspecific activators of cell-mediated immunity such as lentinan (11). For Leishmania sp., B. dermatitidis, and B. pertussis the protein ligands that bind to the I domain and the carbohydrate ligands that bind to the lectin domain have been identified (1, 2). Previous studies have suggested that recognition of C. albicans8), as well as other fungi (4), by CD11b/CD18 depends upon both the I domain and the lectin domains within its CD11b subunit. (

The structural basis for the ligand binding to these two domains of CD11b/CD18 has begun to emerge. The crystal structure of the CD11b I domain was solved and revealed a novel divalent cation coordination site, the MIDAS motif, which is essential for the ligand binding functions of CD11b/CD18 (16). Certain ligands of CD11b/CD18 bind to the expressed I domain in a cation-dependent manner (17, 18, 19, 20), and many CD11b/CD18 blocking Abs map to the I domain (21). The lectin-like adhesion site of CD11b/CD18 was also mapped using carbohydrate ligands (11). This domain resides in an area proximal to the cell membrane and exhibits a broad specificity for -glucan (glucose) and mannan (mannose) polysaccharides. Subsequent studies showed that binding of such carbohydrates to the lectin domain of CD11b resulted in priming of the receptor for phagocytosis and cytotoxicity, which involved a tyrosine kinase, a Mg-dependent conformational change in the I domain, and exposure of the mAb CBRM1/5 activation epitope (22). Consistent with this observation, certain mAbs can activate CD11b/CD18 adhesion by binding to the lectin domain (23). These studies support data showing that the I and the lectin domains are both involved in CD11b/CD18-mediated adhesion to intact microorganisms, including C. albicans (5, 8).

Previously, we have used the homologue-scanning mutagenesis strategy (24) to analyze the binding site for the I domain ligands of CD11b/CD18 (25). This strategy takes advantage of the known crystal structures of the CD11a and CD11b I domains and entails swapping homologous structural motifs, either {alpha}-helical segments, -sheets, or their connecting loops from CD11a into CD11b to identify requisite sites for ligand binding function while conserving the structure of CD11b/CD18. Mutated CD11b/CD18 receptors containing CD11b I domain segments of 610 amino acids "swapped" with their corresponding segments of CD11a I domain have been stably expressed in 293 cells. We have demonstrated previously that these mutant CD11b/CD18 receptors exhibit the conformation of the wild-type receptor by a variety of criteria (25). Using these mutant cell lines, we have shown that the I domain binding pockets for NIF, C3bi, and Fg are overlapping but not identical (26). Recently, these mutants were used to map the CD11b/CD18 binding pocket for NIF (25). The experiments in the present study describe the utilization of these mutants to map the segments of the CD11b/CD18 I domain that mediate adhesion to C. albicans and identify C. albicans as a unique CD11b/CD18 ligand. Additional mutant receptors have been used to examine the roles of the lectin-like domain and the CD18 subunit in the recognition of this fungal ligand.

Fungal culture

C. albicans (58716, American Type Culture Collection, Manassas, VA) was provided by Dr. Herbert L. Mathews, Loyola University of Chicago (Maywood, IL), and used throughout this investigation. Cultures were stored at 25C on Sabourauds dextrose agar (Difco, Detroit, MI). Cells used for experimentation were cultured overnight at 37C on Sabourauds dextrose agar, collected as isolated colonies, washed once in complete HBSS, and counted with a hemocytometer.

NIF protein and mAbs

The NIF protein was provided by Corvas International (La Jolla, CA). The mAbs used were as follows: OKM1 (anti-CD11b, IgG2b), 44a (anti-CD11b, IgG1), LM2/1 (anti-CD11b, IgG1), IB4 (anti-CD18, IgG2a), and W6/32 (anti-MHC class I, IgG1; all from American Type Culture Collection). These Abs were purified from conditioned tissue culture medium using a column of recombinant protein G as described by the manufacturer (Zymed, San Francisco, CA). All anti-human mAbs were of mouse origin, and the secondary Ab used for immunofluorescence analysis was FITC goat anti-mouse IgG (Zymed). Protein concentrations were determined spectrophotometrically.

Cell lines and culture

The human kidney 293 fibroblastoid cell line was maintained as described previously (27) in DMEM-F12 plus 10% FBS, 100 U/ml penicillin, 100 g/ml streptomycin, and 2 mM L-glutamine (all from BioWhittaker, Walkersville, MD). Adherent cells were removed for passage and experiments using enzyme-free cell dissociation buffer (Life Technologies, Grand Island, NY). Construction of 293 cells stably expressing wild-type (WT) or mutant forms of CD11b/CD18 has been previously described (26). Before use in adhesion assays, expression levels of CD11b/CD18 were verified to be similar for all mutant and WT cells using FACS analysis. CD11b/CD18 mutants with low expression levels were enriched for clones with expression levels similar to WT using limiting dilution followed by FACS analysis to select for higher expressing clones. For cases in which this step was required, data were pooled from at least two separate clones. CD11b/CD18 expression also was verified at the time of the experiments to insure similar levels of CD11b/CD18 expression between WT and mutant cells. The mean fluorescent intensities for the WT and mutants used in this study are approximately 300 when stained with an anti-CD11b mAb (OKM1), compared with 5 for mock-transfected cells. All FACS analysis was performed with FACStar (Becton Dickinson, San Jose, CA) in the core flow cytometry facility at The Cleveland Clinic supported by the Keck Foundation. The expression levels of the cell lines used in this study differed by <50%. There was no correlation between the adhesive activity and the CD11b/CD18 expression levels within this range. For example, two subclones of WT-CD11b/CD18 with mean fluorescence intensities of 358 and 697 had 2.5 x 104 and 1.5 x 104 adherent cells, respectively, when a total of 5.0 x 104 cells were added to the assay. The average was 1.99 0.39 x 104 adherent cells over 20 independent experiments.

IL-2-activated human lymphocytes

Human PBMC were obtained by venipuncture from normal donors with informed consent and were isolated using Ficoll-Histopaque (Pharmacia, Piscataway, NJ) by centrifugation at 2000 x g for 25 min. Lymphocytes at the interface were removed with a sterile Pasteur pipette, washed once in HBSS, and cultured in RPMI 1640 supplemented with 10% FBS, 100 U/ml penicillin, 100 g/ml streptomycin, 100 M nonessential amino acids, and 2 mM L-glutamine (all from BioWhittaker, Walkersville, MD) as well as 50 M 2-ME. The cells were distributed at 2.5 x 106 cells/ml into Costar 24-well plates (Becton Dickinson, Lincoln Park, NJ) with 1500 U/ml IL-2 (Sigma, St. Louis, MO), and they were harvested after 7 days at 37C in 5% CO2. These cells were >99% lymphocytes as judged by Wright staining.

Calcein labeling of 293 cells and lymphocytes

Human 293 kidney cells and IL-2-activated lymphocytes were labeled with the fluorochrome calcein-AM (Molecular Probes, Eugene, OR) for use in adhesion assays. The calcein-AM solution was prepared fresh for each experiment using 50-g aliquots provided by the manufacturer by initial solubilization in 10 l of DMSO followed by 950 l of HBSS without Ca2+ or Mg 2+. Cells were suspended at a concentration of 107/ml in DMEM F-12 with 0.5% BSA (Sigma) and 5 g/ml calcein-AM (100 l) for 30 min at 37C in 5% CO2. Labeling was determined to have no effect on cell viability as assessed by trypan blue staining.

Adhesion of lymphocytes to C. albicans

This assay was performed as described previously (28) with some modifications. Briefly, C. albicans hyphae were prepared by growth of 105 yeast in 50 l of RPMI 1640/well (without serum) for 3 h at 37C in flat-bottom 96-well plastic plates (Corning, Corning, NY). After 3 h, 90100% confluence of hyphae was obtained, and 100 l of DMEM/Hams F-12/0.5% BSA was added per well to reduce nonspecific binding. After 30 min, calcein-AM-labeled cells were added to wells of the assay plates, the final volume was adjusted to 250 l/well using DMEM/Hams F-12/0.5% BSA, and the plate was incubated in a 5% CO2 incubator at 37C for 1 h. The assay was terminated by removal of unbound cells from each well by gently washing three times with 200 l of DMEM/Hams F-12/0.5% BSA using a multichannel pipette. For each cell concentration tested and in each experiment, triplicate wells were washed, and another triplicate set remained unwashed for calculation purposes. Fluorescence was then measured using a fluorescence plate reader (Cytofluor, San Diego, CA), and the values obtained were converted to cells bound per well using the Excel software program (Microsoft, Seattle, WA) to construct a standard curve from the fluorescence of the unwashed wells at the various cell concentrations added and to compute mean fluorescence per cell. Adhesion of WT CD11b/CD18 293 cells was assessed in each experiment and was assigned a value of 100% adhesion. Background (nonspecific) adhesion was determined for each cell type for every experiment as the cells remaining bound after washing when the cells were added to wells containing no hyphae. Data for all cell types (experimental) were then expressed as the percentage of WT adhesion after subtraction of background from each value using the formula: % WT adhesion = [(mutant cells bound - background) (WT cells bound - background) x 100]. Background adhesion averaged 710% for all cells tested. Data represent the mean percent WT adhesion for triplicate values of three or more experiments SD.

Competition for binding of lymphocytes to C. albicans

This procedure was performed as described for the adhesion of cells to C. albicans, except that 5 x 104 were preincubated for 1 h with the indicated proteins or mAbs at 37C in 200 l of DMEM/Hams F-12 in polypropylene test tubes. The entire 200-l preincubation mixture was transferred to wells containing C. albicans. Background was determined and subtracted from total counts as described above. For all inhibition experiments, fluorescence per cell was determined as described above and expressed as the percent inhibition of cell adhesion relative to binding of untreated cells using the formula: % inhibition = [1 - [(cells bound treated experimental - background) (cells bound untreated - background)] calcein-AM-labeled cellsx 100]. Data are calculated as the mean percent inhibition for triplicate values of three or more experiments.
CD11b/CD18 transfectants bind C. albicans hyphae in a manner comparable to IL-2-activated human lymphocytes

Numerous studies (10, 12, 29), including those from our laboratory (26), have demonstrated that CD11b/CD18 can be expressed at high levels on cell surfaces as functional heterodimers. We initially assessed the ability of the WT-CD11b/CD18 receptor expressed on 293 cell lines to mediate cell adhesion to C. albicans. The hyphal form of C. albicans was used in these experiments and was obtained by growing the yeast for 3 h at 37C in 96-well plates. The formation of an extensive hyphal network was verified by microscopic examination in each experiment. Transfected 293 cells, labeled with calcein, were added over a range of cell concentrations to the C. albicans hyphae, and the extent of adhesion was quantitated at 1 h. As shown in Fig. 1Go, adhesion increased as a function of the number of cells added. The optimal number of cells added was determined to range from 104 to 5 x 104 cells/well, and individual concentrations of 104, 2 x 104, 3 x 104, and 5 x 104 cells/well were examined in subsequent experiments. As a negative control, the adhesion of mock-transfected cells also was evaluated at these same cell concentrations in adjacent wells. To account for variability among individual experiments and to allow comparison between experiments, data were expressed as binding of cells, as a percentage of WT transfectants. Binding of WT-CD11b/CD18 cells was 5- to 10-fold greater than that of mock-transfected cells at all cell concentrations tested. As shown in Fig. 1Go, the recombinant WT-CD11b/CD18 behaved in a similar manner as the IL-2-activated human lymphocytes. Previous studies have shown that adhesion of such activated lymphocyte preparations to C. albicans is CD11b/CD18 mediated (30). Fig. 1Go shows that the extent of adhesion of both cell types to the hyphae was similar. In contrast, 293 transfectants expressing CD11a/CD18 at a similar level as the WT-CD11b/CD18 transfectants adhered minimally to the hyphae; the extent of adhesion of the CD11a/CD18 transfectants was comparable to that of the mock-transfected cells.

To further verify the role of CD11b/CD18 in cell adhesion to C. albicans, the effects of known inhibitors of CD11b/CD18 function were examined on adhesion of the WT-CD11b/CD18 transfectants and the IL-2-activated lymphocytes to the hyphae in parallel (Fig. 2Go). These inhibitors included several mAbs (OKM1, 44a, and IB4) to different regions of CD11b/CD18. The anti-MHC class I mAb W6/32, which does bind to the 293 transfectants, was used as a negative control. mAb 44a to the I domain of CD11b strongly inhibited the adhesion of WT-CD11b/CD18 transfectants and IL-2-activated lymphocytes to the hyphae; adhesion of both cell types was inhibited by >80%. Treatment with the control mAb (W6/32) had no effect on the adhesion of either cell type to the hyphae. Consistent with a role for the I domain of CD11b in hyphal recognition, NIF, a CD11b/CD18 I domain-specific ligand (25), eliminated adhesion of both the WT-CD11b/CD18 transfectants and IL-2-activated lymphocytes to the hyphae. In addition, an anti-CD18-blocking mAb, IB4, also strongly inhibited (~80% inhibition) WT-CD11b/CD18 and IL-2-activated lymphocyte adhesion to the hyphae. OKM1, a mAb that blocks the function of the lectin domain of CD11b (11), inhibited adhesion of WT-CD11b/CD18 cells and IL-2-activated lymphocytes by 50%, and a further increase in the OKM1 concentration did not substantially increase the extent of inhibition. Taken together, these data demonstrate the role of CD11b/CD18 in mediating cell adhesion to C. albicans hyphae and are consistent with a contribution of multiple domains in recognition of the pathogen by the receptor (8).
To dissect the roles of individual CD11b and CD18 chains in adhesion, we conducted binding experiments with the transfected cells expressing only the CD11b subunit or the heterodimeric WT-CD11b/CD18 (Fig. 3Go). Cell surface expression of CD11b in the absence of CD18 on 293 cells and lack of association with other endogenous subunits have been verified previously using several independent approaches, including FACS analysis with a CD11b-specific mAb (44a) and a CD18-specific mAb (IB4) and surface labeling and immunoprecipitation (25). The cells expressing only CD11b bound hyphae well, although at a reduced level compared with WT-CD11b/CD18 (5476% of WT). To insure that the decreased adhesion was not a result of small differences in receptor expression, several subclones of the CD11b-only cells were developed and tested. These subclones expressed from 50100% of WT levels of CD11b by FACS analyses, and no correlation was found between expression level and the extent of adhesion among these cell lines (not shown). The specificity of CD11b-C. albicans interaction was verified using the specific blocking mAbs noted above. Both OKM1 and 44a inhibited adhesion of CD11b cells to hyphae, while the CD18-specific mAb IB4, which inhibited binding of the cells expressing the heterodimer (see above), had no effect on adhesion of CD11b-only cells. Thus, the CD11b subunit alone is sufficient to mediate hyphal binding and suggests that the CD18 subunit does not provide a major contact site for C. albicans.
We next examined the effects of a CD18 mutation (S138A) on CD11b/CD18 adhesion. Previous studies (26, 29) have shown that this mutation or ones at the immediately adjacent oxygenated residues, D134 and S136, block the capacity of CD11b/CD18 to bind certain ligands, including fibrinogen and C3bi. As shown in Fig. 3Go, mutation of Ser138 to Ala abolished CD11b/CD18 adhesion to C. albicans. Cell surface expression of the WT and mutant CD11b/CD18 was similar as judged by FACS (26). Therefore, although CD18 is not required for adhesion to C. albicans, i.e., CD11b in the absence of CD18 supports cell adhesion, mutation of Ser138 abrogates the adhesive activity of the CD11b/CD18 complex.

Roles of the lectin-like domain in CD11b/CD18-mediated adhesion to C. albicans

To investigate the role of the CD11b lectin-like domain in adhesion to C. albicans, cells were developed expressing a heterodimer in which the I domain of CD11a was substituted with that of CD11b (termed L/M). CD11a/CD18 has been shown to lack the lectin binding activity found in CD11b/CD18 (11). In the absence of the CD11b lectin-like domain, the chimeric CD11a/CD18-expressing cells (L/M), which contains only the I domain of CD11b, bound C. albicans well, ranging from 6082% of the WT-CD11b/CD18 adhesion (Fig. 4Go). Verifying the specificity, the wild-type CD11a/CD18-expressing cells failed to adhere to the fungi. These data demonstrate that the CD11b/CD18 lectin-like domain is not required for CD11b/CD18-mediated adhesion to hyphae.

The absence of the lectin-like domain in CD11a/CD18 was confirmed. Known carbohydrate ligands of this site (11), N-acetyl- D-glucosamine and methyl--glucopyranoside, were assessed for their ability to inhibit adhesion of the WT-CD11b/CD18 and the chimeric receptor (L/M)-expressing cells to the hyphae (Fig. 5Go). Each carbohydrate was used at a concentration of 150 mM, previously shown to maximally inhibit the function of the lectin-like domain (11). -D-glucose was used as a negative control. As shown in Fig. 5Go, N-acetyl-D-glucosamine and methyl--glucopyranoside inhibited adhesion of the WT-CD11b/CD18 cells by 60 and 70%, respectively, demonstrating a contribution of the lectin-like domain in CD11b/CD18 to adhesion. In contrast, these carbohydrates had no effect on L/M adhesion (Fig. 5Go). In separate experiments, we evaluated the effects of these sugars on CD11b/CD18-NIF interaction. At 150 mM, none of the three carbohydrates had any effect on NIF binding to the transfected cells as judged by FACS analysis using biotinylated NIF and avidin-FITC conjugate (25) (data not shown).
To test the hypothesis that modulation of CD11b/CD18 function by the lectin-like domain depends upon its influence on the conformational state of the I domain, we examined the effects of the inhibitory carbohydrates on a constitutively active CD11b/CD18 mutant, CD11b(Q190-S197)/CD18. This mutant contains a substitution within its I domain and is constitutively active with respect to the binding of activation-dependent CD11b/CD18 ligands, such as Fg (31). Consistent with its active state, this mutant exhibited enhanced binding to C. albicans; the adhesion of the CD11b(Q190-S197)/CD18 mutant cells was equal to or greater than that of the WT-CD11b/CD18 at higher cell concentrations (not shown). As shown in Fig. 5Go, the inhibitory carbohydrates, N-acetyl-
D-glucosamine and methyl--glucopyranoside, had no effect on adhesion of this constitutively active CD11b/CD18 receptor, in contrast to their potent inhibition of the WT-CD11b/CD18 receptor.

Mapping the C. albicans binding region within the CD11b I domain

Since mAbs to the CD11b I domain and NIF blocked CD11b/CD18 adhesion to C. albicans (Fig. 2Go), and since the chimeric CD11a/CD18 receptor containing the CD11b I domain adhered well to the fungus (Fig. 4Go), we sought to localize the specific segments within the CD11b/CD18 I domain that mediate this adhesion. The adhesion to C. albicans of 10 homologue-scanning mutants within the I domain was assessed. These previously described mutants contain swaps of homologous small structural elements, {alpha}-helixes, -sheets, and connecting loops, of the CD11a I domain into the CD11b I domain (25). The adhesion of these mutants to C. albicans hyphae was assessed over the range of cell concentrations shown in Fig. 1Go, but only data from a representative concentration of 5 x 104 cells are shown (Fig. 6Go). The 10 mutants assessed were CD11b (P147-R152), CD11b (M153-T159), CD11b (Q190-S197), CD11b (E178-T185), CD11b (R208-K217), CD11b (K245FG), CD11b (D248-Y252), CD11b (E253-R261), CD11b (D273-K279), and CD11b (R281-I287). Adhesion of these mutants to the hyphae fell into two general categories; those that are similar to WT-CD11b/CD18, including CD11b (M153-T159), CD11b (Q190-S197), CD11b (E178-T185), CD11b (K245FG), and CD11b (D273-K279) (Fig. 6Goa), and those that demonstrate virtually no adhesion to C. albicans, including CD11b (P147-R152), CD11b (R208-K217), CD11b (D248-Y252), CD11b (E253-R261), and CD11b (R281-I287) (Fig. 6Gob).
The Center for Disease Control has noted that the incidence of life-threatening opportunistic fungal infections have increased an alarming 500-fold since 1980, and C. albicans is responsible for 7080% of all opportunistic fungal infections (9). A principal reason for the high mortality in C. albicans and other fungal infections is the ability of fungal surface structures (mannoproteins) to interact with leukocytes and modulate the immune response to these organisms, enabling evasion of the immune response in patients who are often already ill from other causes such as premature birth, immunosuppressive therapy, cancer therapy, and AIDS (9). This study has sought to delineate how C. albicans interacts with its major recognition site on leukocytes, CD11b/CD18. Specific regions of the receptor involved in recognition of C. albicans have been identified. These results allow us to compare and contrast C. albicans to other CD11b/CD18 ligands. Moreover, the information obtained also provides some general insights into ligand recognition by integrins. We reached the following conclusions. 1) Even though the CD11b subunit is sufficient for efficient recognition of C. albicans, a single point mutation, S138 to A, within the CD18 subunit abolishes the interaction of the fungus with the CD11b/CD18 complex. Thus, the mutation exerts a dominant negative effect on receptor function. 2) The lectin domain is not sufficient for CD11b/CD18 adhesion. Its influence on CD11b/CD18 activity depends primarily on its modulation of the function of the I domain. 3) The CD11b I domain provides the major contact interface for C. albicans. This contact interface is mapped to a region containing segments of P147-R152, R208-K217, D248-R261, and R281-I287. 4) Compared with NIF, Fg, and C3bi, other mapped CD11b/CD18 ligands, the interactive mechanism of C. albicans with the receptor is distinct.

In undertaking the present analyses, we were greatly concerned that differences in expression levels of the various mutants might influence CD11b/CD18-mediated adhesion to the fungal hyphae. Accordingly, we only compared cells that expressed equivalent levels of WT and mutated forms of CD11b/CD18 as assessed by FACS analyses. These analyses were performed with each cell type before and after adhesion experiments, and low expressing mutants were subcloned to select for high expression equivalent to WT. Thus, the expression level of the transfectants used in our study differed by <50%. In addition, WT-CD11b/CD18 and mock-transfected cells as well as background binding for each cell type to wells containing no fungus were determined as internal controls for each experiment. The physiologic relevance of the CD11b/CD18-expressing 293 cells used in our study was verified by the demonstration that these cells behaved qualitatively and quantitatively similarly to IL-2-activated lymphocytes, which are known to use CD11b/CD18 to bind C. albicans (30). Furthermore, CD11a/CD18-expressing 293 cells did not bind to hyphae specifically, and this integrin does not play a role in C. albicans binding to leukocytes (8, 30). The failure of CD11a/CD18 to recognize the hyphae allowed implementation of the homologue-scanning strategy in which segments of CD11a were exchanged for the corresponding segments of CD11b to map the I domain contact segments for C. albicans.

The role of the CD11b subunit in CD11b/CD18 ligand binding is well documented (17, 20, 21, 32) and is confirmed in this study. Transfected cells expressing only CD11b bound specifically to the hyphae. An mAb to the CD11b subunit that blocked binding of the heterodimeric receptor to the fungus remained inhibitory, whereas a mAb to the CD18 subunit lost inhibitory function. The CD11b cells did not bind to C. albicans as efficiently as the WT cells (65% WT; Fig. 3Go), but the modest reduction in binding precludes the existence of a substantial number of direct contacts between the hyphae and CD18. However, the function of the CD18 subunit in C. albicans binding to the heterodimeric receptor or, for that matter, of other ligands to this or other integrins is significant but remains difficult to define. The D134XSXS sequence of CD18 has been previously demonstrated to be critical for the recognition of several CD11b/CD18 ligands. Mutation of D134, S136, or S138 eliminated adhesion of CD11b/CD18-expressing cells to dOva, C3bi, ICAM-1, and fibrinogen (26, 29), and C. albicans can be added to this list. The homologous region within 3 also is critical to the ligand binding function of this integrin subfamily (33), and a naturally occurring point mutation at one of these residues has been reported to fully inactivate these receptors (34). Recently, it has been proposed that the subunits of integrins, including CD18, contain an I domain-like structure, and the DXSXS sequence is part of a MIDAS motif providing cation coordination sites (35, 36, 37, 38) and also may contain direct ligand contact sites as well (39, 40, 41). However, as demonstrated in this study, cells expressing CD11b alone adhered well to C. albicans, and removal of the entire CD18 subunit had limited effects on the recognition of this ligand, suggesting that, at least for this ligand, the CD18 subunit is not required for ligand engagement. Thus, the consequence of mutation of Ser138 is the creation of a dominant negative effect, which masks the intrinsic activity of the CD11b subunit. The mechanism of this effect is unclear. One possibility is that the absence of this residue changes the cation coordination complex to one that is incompatible with ligand binding. Clearly, the relationship between ligand and cation binding is vital to integrin function (42).

Previous studies with C. albicans (8) and with other ligands (22, 43) have implicated the lectin-like domain in CD11b/CD18 functions. In this study, the lectin domain blocking mAb OKM1 and two carbohydrate ligands of the lectin-like domain significantly reduced the adhesion of both WT transfectants and IL-2-activated lymphocytes. Although a large proportion of the surface of C. albicans is composed of -glucan polymers, this domain is not sufficient for binding of C. albicans to the receptor. This conclusion is supported by the observations that 1) certain receptors with mutations only in the I domain and without alterations in the lectin domain failed to adhere to C. albicans; and 2) the chimeric CD11a/CD18 receptor that contained only the I domain of CD11b and lacked the lectin-like domain (the carbohydrates that inhibited WT adhesion had no effect on the function of the chimeric receptor) was able to bind the fungus to an extent similar (80%) to that of the WT-CD11b/CD18 cells. The mechanism by which the lectin-like domain modulates CD11b/CD18 function appears to reside in its modulation of I domain conformation. The mutant CD11b/CD18 receptor, CD11b(Q190-S197)/CD18, contains an activating mutation within the I domain (31), and the inhibitory carbohydrates did not block C. albicans binding by this activated receptor. These data and interpretation are consistent with the capacity of the lectin domain-specific mAb VIM12 to exert an activating effect on I domain-mediated adhesion of CD11b/CD18 as observed in association with homotypic aggregation (23). Recently, several studies have suggested the existence of a promiscuous lectin binding site with somewhat distinct sugar specificity within CD11a/CD18 (44). In the same study, N-acetyl-D-glucosamine was shown to inhibit the association between CD11a/CD18 and the Fc{gamma}R333 receptor (44). Although this polysaccharide had no effect on L/M chimera adhesion to hyphae (Fig. 5Go), our data cannot rule out the possibility that a lectin site with different carbohydrate specificity may exist in CD11a/CD18 and could participate directly in L/M chimera interactions. Hyphae binding to a low affinity I domain may require the assistance of the lectin domain (in the cases of WT and L/M chimera), whereas a high affinity I domain can overcome the requirement for an additional contact site in the lectin domain (in the case of the CD11b(Q190-S197) mutant). Nevertheless, since cells expressing WT CD11a/CD18 and mutant CD11b/CD18 with intact lectin domain did not bind hyphae, the direct contribution from the lectin domain to hyphae binding, if any, cannot be substantial. The lectin domain also could be involved in directly priming CD11b/CD18 for antifungal cytotoxicity, a possibility that was not addressed in this study (22, 30).

That the I domain provides the major contact interface for C. albicans is supported by the following experiments. First, CD11b I domain-specific mAbs and the high affinity CD11b/CD18 ligand (NIF) that has been shown to interact exclusively with the I domain (25) blocked CD11b/CD18 adhesion to C. albicans. Second, the chimeric CD11a/CD18 receptor containing the CD11b I domain adhered well to the fungi. Accordingly, we sought to localize the functional segments within the I domain using our existing homologue-scanning mutants. Of the 10 representative mutants we tested, five (CD11b (M153-T159), CD11b (Q190-S197), CD11b (E178-T185), CD11b (K245FG), and CD11b (D273-K279)) had similar C. albicans adhesion to WT-CD11b/CD18 (Fig. 6Goa). Of note are mutants CD11b (K245FG) and CD11b (D273-K279). CD11b (K245FG) has approximately 60% the adhesive activity of WT-CD11b/CD18. This mutant binds significantly less well to C3bi (<20% of WT; our manuscript in preparation), thus distinguishing C. albicans from C3bi as a CD11b/CD18 ligand. The CD11b (D273-K279) mutant has been shown previously to exhibit virtually no adhesion to Fg (26), although it bound well to hyphae. This result distinguishes C. albicans from Fg as a CD11b/CD18 ligand. Five other mutants, including CD11b (P147-R152), CD11b (R208-K217), CD11b (D248-Y252), CD11b (E253-R261), and CD11b (R281-I287), demonstrate virtually no adhesion to C. albicans (Fig. 6Gob). The first four of these segments have been shown to constitute the NIF binding site within the I domain (Fig. 7Go) (25). Thus, C. albicans shares an overlapping ligand binding site with NIF, providing the molecular basis for the potent inhibitory activity of NIF toward C. albicans binding. However, the CD11b (R281-I287) and CD18 (S138A) mutations have no effect on NIF recognition (25) while eliminating adhesion to C. albicans, thus setting these hyphae apart from NIF as a microbial ligand for CD11b/CD18. Taken together, our data show that the NIF binding site (P147-R152, R208-K217, D248-R261) represents the core structure of a broad surface that is involved in establishing the oligospecificity of the CD11b/CD18 integrin toward its multiple and unrelated ligands. The C. albicans recognition site encompasses the entire NIF binding site plus an additional segment (R281-I287; Fig. 7Go).

In summary, this study has provided important information regarding the roles of the CD18 subunit and the lectin-like domain of CD11b in the function of the CD11b/CD18 receptor. It also represents the first mapping of a pathogenic human microorganism to an integrin receptor and demonstrates that relative to previously mapped protein ligands for CD11b/CD18, C. albicans is a unique CD11b/CD18 ligand. Such a unique interaction may very well underlie the demonstrated ability of C. albicans to modulate CD11b/CD18 function. Greater understanding of this unique interaction could provide avenues for novel adhesion based antimicrobial or antiinflammatory therapies.


Full Article:

http://www.jimmunol.org/cgi/content/abstract/161/11/6198





Keywords: Fungal Pathogen Candida albicans Integrin CD11b/CD18 I Domain Lectin-Like Domain CD18 Subunit advanced

Rate This Article:

Related Articles

Interaction of Candida albicans with Human Leukocytes and Serum
Interaction of Candida albicans with Human Leukocytes and...

 
Secreted proteinases and Candida albicans virulence

PAP inhibitor with in vivo efficacy identified by Candida albicans
PAP inhibitor with in vivo efficacy identified by Candida...