Fungal liver infection in marrow transplant recipients: prevalence at autopsy, predisposing factors, and clinical features.

To determine the prevalence of fungal liver infection at autopsy in marrow transplant recipients, we reviewed autopsy results for the period 1980-1989. Cases were compared to randomly chosen autopsied controls without fungal infection. Fungal liver infection was found in 67 (9%) of 731 patients. Fungal cultures of liver lesions were positive for 34 of 67 patients, most of whom had been culture-positive for the same fungal species (largely Candida) during life. Multivariate analysis revealed that independent predictors of fungal liver infection were deep fungal infection after transplantation (RR, 35), colonization or superficial infection after transplantation (RR, 13), and severe liver dysfunction caused by veno-occlusive disease of the liver and/or graft-versus-host disease (RR, 7). Clinical and laboratory findings during the last month of life revealed no differences between cases and controls. Liver imaging studies performed during the last 15 days of life had a sensitivity of only 18% for detecting fungal liver lesions.

Adherence of Candida albicans to tissues from mice with drug- or radiation-induced immunodeficiencies.

Host factors that influence binding of Candida albicans to murine spleen, lymph node, and kidney were studied. Organs were harvested from BALB/cByJ and AKR/J mice immunocompromised by irradiation, cyclophosphamide, and cortisone acetate alone or in combination. Tissues from treated mice and untreated littermates were compared for their ability to bind C. albicans in ex vivo assays. Immunosuppressive regimens decreased yeast binding to splenic marginal zones, but when mice recovered for 5 days after treatment, adherence to spleen was similar to adherence in untreated littermates. Adherence to lymph node and kidney in treated mice was not different from binding to these tissues in untreated mice. Total serum immunoglobulin titers correlated with binding of yeast cells to mouse spleen. Blocking studies ruled out a mannosyl-fucosyl receptor-mediated binding. These results suggest that ex vivo adherence of C. albicans represents a host immune defense mechanism by which the immunocompetent host binds blood-borne yeast cells to host immune cells in reticuloendothelial organs to prevent dissemination to other organs.

Serotype prevalence of Candida albicans from blood culture isolates.

Blood culture isolates of Candida albicans were collected from 102 patients in Seattle, Wash., hospitals (n = 77) and Hong Kong (n = 25). The patients were classified by immune status into two groups. Group I patients were severely immunosppression, and group II patients had underlying risk factors for candidemia but no underlying immunosuppression. Serotyping by Hasenclever tube agglutination was done. In the Seattle area, the odds of fungemia with type B C. albicans were 3.62 times greater than the odds of type B fungemia in group II patients. Although the odds ratio could not be computed for Hong Kong patients, the direction of the relationship in this population was consistent with the data on Seattle patients. Despite the magnitude of the odds ratios, the relative prevalence of type B over type A in group I compared with group II was not significant when analyzed separately by region, probably because of relatively low numbers of isolates in group II. Accepting that the effect of immune status on serotype is equivalent across regions but presupposing that a regional effect on type B prevalence exists, the pooled odds for fungemia with serotype B in group I patients are increased 5.4-fold over those of group II patients. Logistic regression analysis controlling for region gave similar results.

Adherence of Candida albicans to tissues from mice with genetic immunodeficiencies.

Ex vivo adherence comparisons were made between immunocompetent and immunocompromised mouse tissues, and the roles of serum immunoglobulin and macrophages in the adherence of Candida albicans were investigated. Spleen, lymph node, and kidney tissues were harvested from congenitally immunodeficient mice, including AKR/scid, C.B-17, C3Hscid, and N:NIH nu/bg/xid mice, and their normal counterparts into which the defects were bred (AKR/J, C3H/HeSnJ, and BALB/c-ByJ). Tissues were compared for the ability to bind C. albicans 219 in an ex vivo assay. In general, immunodeficiencies significantly decreased binding of C. albicans to spleen but not to lymph node or kidney tissue compared with immunocompetent mice. In C3Hscid and AKRscid mice, spleen tissues from "nonleaky" mice bound significantly fewer yeast cells (P = 0.0005 and 0.0009, respectively) than did those from C3H/HeSnJ or AKR/J mice. Numbers of adherent yeast cells were similar in "leaky" AKRscid and AKR/J mice. Yeast adherence to spleen tissue from N:NIH nu/bg/xid mice correlated with mouse age (P = 0.01). Measurements of total serum immunoglobulin indicated that the scid defect was most complete in C3Hscid mice and that yeast adherence in spleen tissue correlated with immunoglobulin titers. Results of adherence assays and macrophage-specific immunostains suggested that factors determining adherence differ among reticuloendothelial organs.

Comparison between methods for serotyping of Candida albicans produces discrepancies in results.

Serotyping of 101 clinical isolates of Candida albicans was done with two sets of Hasenclever original anti-Candida typing sera (HSN 1 and 2) and Iatron Candida Check factor 6 typing serum (IF6). The results of these two methods were compared with slide agglutination reactions of yeast with monoclonal antibody H9. Agglutination reactions with this antibody have been previously shown to correlate with serotype. Results indicate the following correlations: between HSN 1 and HSN 2 serotyping, 93% (kappa = 0.85; 95% confidence interval [CI], 0.70 to 0.99); between IF6 and HSN 1, 60% (kappa = 0.39, 95% CI, 0.19 to 0.58); and between IF6 and HSN 2, 74% (kappa = 0.77; 95% CI, 0.64 to 0.90). Results with HSN 1 and 2 antisera correlated with H9 reactivity at 85 and 89% (kappa = 0.88; 95% CI, 0.75 to 1.00; and kappa = 0.85; CI, 0.70 to 0.99, respectively), while agreement between IF6 and H9 reactivities was less than or equal to 64% (kappa less than or equal to 0.43; 95% CI, 0.14 to 0.60). Autoagglutination of yeast during IF6 serotyping occurred with 21 of the 101 (20.8%) yeast strains. In every case, these yeast strains were serotyped by the HSN methods without autoagglutination and were uniformly type B. This study implies that it may not be possible to make valid comparisons between studies which compare serotype prevalence unless the same methods are used to serotype the yeast. The practicality and utility of serotyping in epidemiological studies are discussed, as are some of the problems associated with the available methods.

Differential adherence of hydrophobic and hydrophilic Candida albicans yeast cells to mouse tissues.

Using an ex vivo binding assay, we previously demonstrated that yeast cells grown at 37 degrees C display binding specificity in mouse spleen, lymph node, and kidney tissues. In spleen and lymph node tissues, binding was predominantly in regions rich in macrophages. Here, we tested the possibility that hydrophobic and hydrophilic cells bind differentially to host tissues. Hydrophobic and hydrophilic yeast cells of four Candida albicans strains were incubated for 15 min at 4 degrees C with cryostat sections of organs that had been rapidly frozen after removal from BALB/cByJ mice. Unattached cells were removed by washing, and the sections were examined. Hydrophobic cells bound diffusely and abundantly to all tissues, while hydrophilic cell attachment was restricted to specific sites. For example, hydrophobic cells bound to the white and red pulp and the marginal zones in spleens, whereas hydrophilic cells attached primarily to the marginal zones. Hydrophobic yeast cells attached throughout lymph node tissue including paracortical areas, but hydrophilic cell attachment occurred primarily at the subcapsular and trabecular sinuses, EDTA inhibited the adherence of hydrophilic cells but not hydrophobic cells to mouse tissues. Hydrophobic C. albicans strains displaying similar levels of hydrophobicity differed quantitatively in their levels of attachment to kidney and spleen tissues, confirming our earlier observation that surface hydrophobicity is not the sole determinant in adherence to host cells. Other studies have shown that hydrophobic and hydrophilic cells display different virulence characteristics related to their surface properties and that hydrophobic cells are more virulent than hydrophilic cells. Taken together, the present results suggest that the enhanced virulence of hydrophobic cells over hydrophilic cells may be due, in part, to the potential of hydrophobic cells to bind throughout various organs following clearance from the bloodstream.

Characteristics of Candida albicans adherence to mouse tissues.

An ex vivo binding assay originally described for determining lymphocyte homing receptors was adapted for studying Candida albicans-host cell interactions in unfixed tissue sections. BALB/cByJ mice were sacrificed, and various organs were removed, rapidly frozen on dry ice, and sectioned. C. albicans yeast cells were suspended to 1.5 x 10(8) cells per ml in Dulbecco modified Eagle medium supplemented with 5% newborn calf serum, and 100 microliters of the suspension was added to tissue sections for 15 min with rotation at 4 degrees C or at 22 to 24 degrees C. The sections were then fixed in glutaraldehyde, washed, and examined. Stationary-phase yeast cells adhered better than log-phase cells, and adherence characteristics were similar at 4 degrees C and 22 to 24 degrees C. Yeast cells from nine strains of C. albicans showed similar tissue specificity. Adherence to lymph node tissue was confined to subcapsular spaces and trabecular sinuses. In the spleen, yeast cells bound to the marginal zones. In both tissues, an association of yeast cells with tissue macrophages was suggested by results with macrophage-specific monoclonal antibodies and fluorescent or immunoperoxidase staining techniques. C. albicans adhered to convoluted tubules, glomeruli, and the tunica media of arterioles in the kidney. During experimentally induced fungemia in mice, C. albicans yeast cells associated with the same tissue sites as in the ex vivo assay, except that binding to renal arterioles was not seen in the in vivo test. A strain of Saccharomyces cerevisiae showed some adherence patterns in common with C. albicans, which indicates that tissue adherence is not sufficient for virulence. Mechanisms of attachment were not determined, but strains of C. albicans varied quantitatively in their ability to attach, and binding was inhibited by chelators of divalent cations.

Caveats in the investigation of form-specific molecules of Candida albicans.

Numerous reports purporting the existence of form-specific antigens of Candida albicans have been published, but it is generally unclear whether antigenic variability is an acceptable alternative interpretation. In this study, we used indirect immunofluorescence and immunogold electron microscopy to determine the distribution and form specificities of two antigens during yeast and hyphal growth in several defined and complex media. The results confirmed that antigen expression varies with length of incubation, nutrition, and serotype and indicate that the form specificities of antigens may be misinterpreted when conclusions are based exclusively on indirect immunofluorescence and extraction procedures. We therefore suggest that investigations be designed to include serotype A and B isolates grown in both complex and chemically defined media and that agglutination, immunofluorescence, or enzyme-linked immunosorbent assays on whole cells or cell extracts be used as presumptive tests. Confirmation of form-specific antigens should be done by appropriate immunoelectron microscopic evaluation.

Immunochemical localization of nitrogenase in marine trichodesmium aggregates: relationship to n(2) fixation potential.

Colonial aggregation among nonheterocystous filaments of the planktonic marine cyanobacterium Trichodesmium is known to enhance N(2) fixation, mediated by the O(2)-sensitive enzyme complex nitrogenase. Expression of nitrogenase appears linked to the formation of O(2)-depleted microzones within aggregated bacterium-associated colonies. While this implies a mechanism by which nonheterocystous N(2) fixation can take place in an oxygenated water column, both the location and regulation of the N(2)-fixing apparatus remain unknown. We used an antinitrogenase polyclonal antibody together with postsection immunocolloidal gold staining and transmission electron microscopy to show that (i) virtually all Trichodesmium cells within a colony possessed nitrogenase, (ii) nitrogenase showed no clear intracellular localization, and (iii) certain associated bacteria contained nitrogenase. Our findings emphasize the critical role coloniality plays in regulating nitrogenase expression in nature. We interpret the potential for a large share of Trichodesmium cells to fix N(2) as an opportunistic response to the dynamic nature of the sea state; during quiescent conditions, aggregation and consequent expression of nitrogenase can proceed rapidly.

Oral Candida albicans isolates from nonhospitalized normal carriers, immunocompetent hospitalized patients, and immunocompromised patients with or without acquired immunodeficiency syndrome.

A total of 128 human oral isolates of Candida albicans were collected from asymptomatic healthy carriers (64 isolates); asymptomatic, nonimmunosuppressed, hospitalized patients (25 isolates); immunosuppressed transplant patients (19 isolates); and human immunodeficiency virus-infected patients with symptoms of acquired immunodeficiency syndrome and oral candidiasis (20 isolates). Isolates were serotyped as A or B and tested for reactivity with an agglutinating immunoglobulin M monoclonal antibody (H9). Immunocompetent individuals colonized by oral C. albicans were almost equally likely to carry serotype A as serotype B cells, while immunocompromised individuals were at least twice as likely to be infected by serotype B than serotype A strains. The reactivity of isolates with H9 antibody followed a similar but more distinctive pattern. Approximately half of the strains from immunocompetent individuals reacted strongly with H9, and the remainder reacted weakly. However, up to 75% of the isolates from immunocompromised patients reacted weakly with H9, while the remainder reacted strongly. A correlation between H9 reactivity and the serotypes of these isolates existed (P = 0.16). The correlation between H9 reactivity and immune status was even stronger (P = 0.025). The monoclonal antibody activities described above were determined by agglutination tests during defined phases of C. albicans growth. Expression of antigen at various times during growth of several isolates was confirmed at the cellular level by analysis using fluorescence-activated cell sorting. Despite the correlation between serotype A and H9 reactivity, H9 antigen was not identical to the serotype A antigen because four serotype A strains reacted only weakly with H9 antibody, and one strain reacted strongly with H9 but was serotype B. These data indicate that oral strains of C. albicans from immunocompetent individuals differ as a group from C. albicans isolated from those who are immunosuppressed.

Cell surface and intracellular expression of two Candida albicans antigens during in vitro and in vivo growth.

Intracellular and cell surface expression of two antigenic determinants of Candida albicans was studied by reacting ultrathin sections of the fungus with monoclonal antibodies and locating the sites of antibody-antigen reactions by use of colloidal gold-coupled secondary antibody and electron microscopic techniques. Comparisons were made between fungal cells allowed to develop in vitro and cells (in vivo) recovered from infected mice. Differences between in vitro antigen expression and expression in vivo were particularly noted during early germination events. Antigen expression in vivo appears to be concentrated in deeper cell wall layers and beneath the plasmalemma especially on germ tubes. Cells which develop in vivo and in vitro have antigen associated with an outer flocculent layer and on the innermost cell wall layer of mother cells and hyphae. A middle electron transparent cell wall layer is usually devoid of antigen. Comparison of antigen expression during in vivo growth in early (intraperitoneal) infection versus late disseminated (kidney) disease suggests that both of the antigens are expressed in greater quantity on the surface of germ tubes than on mother cells in kidney tissue as compared to intraperitoneally-grown cells. Results also suggest that at least one of the antigens, which is a substituted mannan, can be found on a lamello-tubular intracellular structure.

Ultrastructural and biochemical studies of two dynamically expressed cell surface determinants on Candida albicans.

Variability in the expression of two different cell surface carbohydrate determinants was examined with two agglutinating immunoglobulin M monoclonal antibodies (H9 and C6) and immunoelectron microscopy during growth of three strains of Candida albicans. A single strain of Candida parapsilosis did not express either antigen at any time during growth. Antigens were detected on the surface of C. albicans by agglutination tests with either H9 or C6 over a 48-h growth period. The difference in specificities of the monoclonal antibodies was demonstrated by Ouchterlony double-diffusion tests with solubilized antigens and by variabilities in the reactivity of the agglutinins among yeast strains. The antigenic determinants were isolated by specific immunoprecipitation and protease digestion and characterized by methods including high-pressure liquid chromatography, gas-liquid chromatography, and mass spectroscopy with both chemical and electron ionization. These determinants both contain mannose and glucose. In the case of antigen H9, an additional carbohydrate was detected with gas chromatography and mass spectroscopy. The location of antigens on individual cells was determined by indirect labeling of the determinants, first reacting cells with H9 or C6 followed by goat anti-mouse antibody conjugated with 20-nm colloidal gold particles. Transmission electron microscopy was used to examine cells. The antigens that were reactive with the monoclonal antibodies were associated with a flocculent surface layer. Expression of this layer and expression of the antigens is a dynamic process which is growth phase and strain dependent. The antigens were not expressed on very young cells and disappeared from the cell surface of most C. albicans strains with age. The use of monoclonal antibody to cell surface determinants may allow characterization of cell surface antigens of C. albicans and be helpful in establishing receptors which mediate adherence.

Variability in expression of cell surface antigens of Candida albicans during morphogenesis.

The location and expression of two different cell surface antigens on germinating and nongerminating Candida albicans cells was examined by using transmission electron microscopy after labeling with monoclonal antibodies (H9 or C6) and immunocolloidal gold. Immunodeterminant expression of the two carbohydrate antigens was followed from early germination events through 20 h of development. The determinant detected by H9 antibody, which was initially lost from the mother cell surface and preferentially expressed only on hyphae during the first 4 h of germination, reappeared on the mother cell by 20 h, whereas the antigen detected by antibody C6 was continually expressed on mother cells and germ tubes throughout germination.

Changes in surface topography of Candida albicans during morphogenesis.

Temporal studies of germ-tube forming yeast cells of Candida albicans by scanning and transmission electron microscopy indicate that extensive vacuolation and possibly also cell wall changes may cause walls of the parent yeast cell to collapse during specimen preparation. This collapse does not occur in cells which have been grown in conditions that suppress germ tube formation and have undergone the same preparative treatment.

Variability in expression of a cell surface determinant on Candida albicans as evidenced by an agglutinating monoclonal antibody.

The expression of a surface immunodeterminant of Candida albicans was investigated with an agglutinating immunoglobulin M monoclonal antibody. The 96 strains of C. albicans tested, of which 76% were recent clinical isolates, were capable of expressing the antigen. The antigen was also produced by strains of Candida tropicalis and Torulopsis glabrata, but not by other yeast species. Expression of the surface immunodeterminant in C. albicans varied as a function of growth as indicated by agglutinin reactions and indirect immunofluorescence tests. When yeast cells were tested with the agglutinin, three patterns of reactivity were observed. In general, cells in the early logarithmic phase were less reactive than cells in the mid-logarithmic phase. Antigen expression, as determined by agglutinin reactivity, was also influenced by the nutritional composition of the growth medium, and in general, cells from broth cultures were usually more reactive than cells grown on solid media. The antigen was solubilized from the cell surface of C. albicans by hot 1 M NaCl. These water-soluble extracts were capable of binding antibody, and a single precipitin band formed when soluble antigen was reacted with the monoclonal antibody in an Ouchterlony double-diffusion test. Whole cell preparations and hot NaCl extracts from yeast strains which did not agglutinate when mixed with the antibody also did not absorb the agglutinin from solution. These data indicate that the expression of surface antigens on C. albicans is a dynamic process which may be influenced by a number of environmental factors. The use of monoclonal antibodies may allow characterization of surface antigens presented to the host during candidiasis.