Candida albicans stress mannoproteins expression in superficial and systemic candidiasis

Mycopathologia 133: 89-94, 1996. 9 1996KluwerAcademic Publishers. Printed in the Netherlands.

89

Candida albicans stress m a n n o p r o t e i n s expression in superficial and systemic candidiasis Stress m a n n o p r o t e i n s in Candida albicans

Jose Pont6n 1, Femando L. Hernando2, Maria Dolores Moragues 3, Pedro L. Barea2, Mara Gerloni4, Stefania Conti 4, Paola Fisicaro 4, Cristina Cantelli 4 & Luciano Polonelli4 1Departamento de Inmunologia, Microbiologia y Parasitologia, Facultad de Medicina y Odontologia; 2Facultad de Ciencias; 3Departamento de Enfermeria I, Universidad del Pais Vasco, Apartado 699, E-48080 Bilbao, Spain, and 4Istituto di Microbiologia, Universitd degli Studi di Parma, Viale A. Gramsci 14, 1-43100 Parma, Italy Received 16 May 1995;acceptedin revisedform 18 September1995

Abstract The presence of heat shock mannoproteins (HSMPs) reactive with sIgA was demonstrated in several C. albicans strains. The subculture of the C albicans isolated from mucosal surfaces on Sabouraud's dextrose agar at 25 ~ switched off the HSMP expression. A re-expression of the HSMPs was obtained in the same medium by shifting the temperature of incubation to 37 ~ However, expression of HSMPs in two strains isolated from deep infections was maintained during several subcultures on Sabouraud's dextrose agar at 25 ~ A glycoprotein of 200 kDa seemed to be the main HSMP reacting with vaginal sIgA. The data presented in this study suggest that factors other than temperature can influence the expression of C albicans HSMPs and therefore these antigens should be referred as stress mannoproteins. Key words: Candida albicans, heat shock mannoproteins, secretory IgA, stress mannoproteins Abbreviations: HSMPs - heat shock malmoproteins; MAb - monoclonal antibody; s I g A - secretory IgA Introduction Oral and vaginal candidiasis are the most prevalent clinical presentations of Candida infections [ 1]. Secretory IgA (sIgA) is thought to play an important role in their control by inhibiting Candida adherence to host cells [2] and sIgA coated Candida cells are present in clinical smears form patients with oral and vaginal candidiasis, a feature which may help in the diagnosis of these infections [3]. In a recent paper, Polonelli et al. [4] have identified the C. albicans antigens reacting with salivary sIgA as heat shock mannoproteins (HSMPs) of 180-200, 130-150, 90--110 and 67-70 kDa, which appeared or were expressed at greater levels in cells grown at 37 ~ than in cells grown at 25 ~ Although heat shock glycoproteins have been described in Sacchalvmyces cerevisiae [5] and Histoplasma capsulatum [6], no other HSMPs have been

described in C albicans. However, several sets of proteins have been previously described in C albicans cells under different types of stress [7-12] but only in few cases was there evidence of their reactivity with antibodies [10, 13]. In this report, we have studied the conditions influencing in vitro and in vivo the expression of C. albicans HSMPs which react with vaginal sIgA antibodies.

Methods

Fungal strains and culture conditions. For our studies, three C. aIbicans strains were used. C. albicans UPV- 1057 and UPV- 1360 were isolated at the University of Pais Vasco from a blood culture and from a patient with vulvovaginal candidiasis respectively. For the production of monoclonal antibodies, C. albicans

90 NCPF 3153, obtained from the National Collection of Pathogenic Fungi (London) was used. In some experiments, 37 strains from 37 patients with vulvovaginal C. albicans infection were also studied. They were isolated on Sabouraud's dextrose agar plates containhag chloranphenicol (0.05 mg m1-1) and identified by the API 20C system (API System S.A.), as well as by germ tube and chlamydospore production. The first subculture of C. albicans UPV-1057 was recovered from the plate in which it had been isolated by gently washing the surface of the plates with sterile PBS. The blastoconidium suspensions were centrifuged at 2000 rpm for 10 rain and washed three times with PBS. The pellet of the last centrifugation was stored at - 8 0 ~ until used. C. albicans UPV-1360 was maintained at room temperature with minimal subculturing in Sabouraud's dextrose agar slants. To study the expression of antigens during subculture, C. albicans UPV- 1360 re-isolated from mouse kidneys and C. albicans UPV-1057 were subcultured every 72 h at 25 ~ on Sabouraud's dextrose agar plates. Organisms from each subculture were harvested by washing the surface of the plates with sterile PBS. Blastoconidium suspensions were processed as described above. Antibodies. Vaginal washes from 37 patients with vulvovaginal C. albicans infection were obtained by instillation and subsequent aspiration of 10 ml of sterile isotonic saline into the posterior vaginal fomix and they were processed as described previously [14]. Briefly, each wash was centrifuged at 400 9 for 10 min and the supernatants were concentrated 10 times by lyophilization. In some experiments, vaginal washes were adsorbed with C. albicans UPV 1360 blastoconidia grown at 25 ~ Adsorption was carried out by incubating the vaginal washes with the same volume of heat killed cells at a concentration of 1010 cells m1-1 . The suspension was incubated for 2 h at room temperature and centrifuged at 2500 9 for 10 rain. The supematant was adsorbed again twice at room temperature for 2 h and once more at 4 ~ for 18 h. MAbs B9E and G3B were produced following standard methods. Briefly, BALB/c mice were immunized by subcutaneous injections of a partially purified antigen of 260 kDa from a germ tube cell wall extract eluted from SDS-PAGE gels [15]. Antibodies used in this study were contained in cell supematant fluid harvested from 10-15 day old cultures of the respective hybridoma cells grown in Iscove's medium (Flow Laboratories) supplemented with 10% fetal bovine serum (Flow).

Immunofluorescence. An indirect immunofluorescence assay was used to detect the expression of HSMPs in C. albicans cells grown in vitro and it was carried out as described previously [4]. The direct immunofluorescence assay was used to detect the presence in vivo oflgA antibodies coating C. albicans cells and it was carried out as described by Polonelli et al.

[3]. Animal infection. A systemic infection was established in six male F2 mice infected intravenously with 1 x 105 blastoconidia of C. albicans UPV- 1360. When the animals showed signs of acute infection, they were sacrificed and their kidneys were removed. The kidneys were disaggregated using a stainless steel mesh and 10/zl of the cellular suspension were placed in the wells of immunofluorescence slides, incubated with vaginal washes and studied by indirect immunofluorescence Antigenic extraction, SDS-PAGE and Western blotting. C. albieans UPV-1360 re-isolated from mouse kidneys was grown on Sabouraud's dextrose agar at both 25 and 37 ~ After 48 h of incubation, an alkaline extract of the cells was obtained according to a previously described method [4]. Sodium dodecyl sulfatepolyacrylarnide gel electrophoresis (SDS-PAGE) was performed by the method ofLaemmli [ 16] in a minigel system (Bio-Rad Laboratories). The total amount of protein loaded per lane was 5 /zg for each extract. Electrophoresis was carried out in 12% (w/v) gels at 200 V for 1 h. Standard molecular weights were: rabbit muscle myosin (205,000), Escherichia coli/3galactosidase (116,000), rabbit muscle phosphorylase b (97,400), and bovine serum albumin (66,000). Subsequently, the gels were electrophoretically transferred to a nitrocellulose membrane (Bio-Rad) in a SartoblotII semi dry system (Sartorius) for 30 min at 4 rnA/cm 2 according to Towbin et al. [17]. After the transfer, the nitrocellulose membranes were either stained with Dig Glycan/Protein Double-labeling Kit (Boehringer Manheim) to detect proteins and glycosilated components or blocked in 5% (w/v) nonfat dry milk in PBS for 1 h at 37 ~ The blocked membranes were washed in PBS and incubated with a 1 : 50 dilution of the vaginal washes in 50 mM TBS [0.05(v/v) Tween 20 in PBS] for 1 h at 37 ~ washed, and incubated with peroxidaselabeled, goat anti-human IgA diluted 1 : 100 (Sigma). Immunoreactive bands were visualized after staining for 30 min with diaminobenzidine (1 : 1000). In some experiments, nitrocellulose strips containing the trans-

91 Table 1. Expressionof HSMPsreactivewithvaginalslgAby C. albicans strains grownin vivo and in vitro

slgA reactivity

C. albicans growth In vivo In vitro

Positive Negative

25 12

37 0

ferred antigens were treated before serum incubation by mild periodate oxidation as described previously [4].

Results

All the C. albicans strains isolated from the 37 vaginal specimens studied expressed in vitro HSMPs reactive with slgA (Table 1). In 25 cases the HSMPs were also detectable in vivo since the Candida cells in vaginal smears were coated with slgA (Figure la). However, in 12 specimens the expression of HSMPs was only proved after the incubation of the C. albicans strains grown in vitro with vaginal washes from other patients containing sIgA, suggesting that the conditions in the vagina were not appropriate tbr reactivity with slgA. This reactivity was specific since yeast cells from a specimen which yielded Rhodotorula rubra by culture remained negative for slgA staining after incubation with vaginal washes from patients who had slgA antibodies reactive in vitro with C. albicans cells. The expression of HSMPs in C. albicans isolated from mucosal surfaces was highly transitory and the subculture of the isolates on Sabouraud's dextrose agar at 25 ~ for 72 h switched offthe expression of the HSMPs reactive with slgA. A temporal re-expression of the HSMPs was obtained by shifting the temperature of incubation to 37 ~ but the C. albicans cells were again negative by slgA staining after growth at 25 ~ In an attempt to investigate if only C. albicans strains isolated from mucosal surfaces expressed antigens reactive with slgA, the C. albicans strain UPV1057, isolated from a blood culture, was also studied. This strain also expressed HSMPs reactive with slgA, since its incubation with vaginal washes from patients who had slgA antibodies resulted in slgA coating of the cells. In this case, expression of the HSMPs was maintained during several subcultures on Sabouraud's dextrose agar at 25 ~ and slgA antibodies present in

Figure 1. lmmunofluorescenceof C. albicans cells coatedby slgA

in a specimenfroma patient with vulvovaginalcandidiasis (a) and in a subcultureon Sabouraud'sdextroseagar at 25 ~C froma blood culture (b). Bar, I0 ,urn.

vaginal washes reacted with C albicans cells up to the 7th subculture (Figure lb). To reproduce in a C. albicans strain isolated from a mucosal surface the antigenic expression showed by the C. albicans strain isolated from the blood culture, we developed an in vivo model of C. albicans infection. Originally, the strain of C. albicans infected the mice that expressed HSMPs reactive with slgA antibodies. However, routine subculture in the laboratory at 25 ~ resulted in a loss of reactivity. Infection of mice with this strain resulted in the development of multiple abscesses in the kidneys and the C. albicans cells isolated from the kidneys re-expressed the HSMPs reactive with slgA since they showed slgA coating when incubated with vaginal washes from patients who had slgA antibodies. Interestingly, antigenic re-expression was maintained in a pattern reminiscent of that seen in C. albicans strains isolated from blood cultures. Again, antigens were expressed during at least seven subcultures on Sabouraud's dextrose agar at 25 ~ To prove the specificity of the changes observed in the expression of the HSMPs two monoclonal antibodies (B9E and G3B) which react with different C. albicans cell wall antigens [15] were also used. Irre-

92 spective of the C albicans strain, the source or the number of subcultures, no differences in antigenic expression were observed. Monoclonal antibody B9E showed a strong fluorescence on the entire cell wall of C. albicans cells isolated either from blood cultures or mouse kidney, as well as from different subcultures on Sabouraud's dextrose agar. On the contrary, monoclonal antibody G3B was negative in all the experiments. SDS-PAGE analysis of extracts from subcultures at 37 ~ of the strain re-isolated from the mouse kidneys showed a more varied array of proteins than the extracts from the cells grown at 25 ~ (data not shown). Most of the proteins were present in both extracts. However, bands of approximately 200, 130, 90 and 70 kDa could not be seen in the extract for cells grown at 25 ~ or were greatly increased after incubation at 37 ~ Most of the components present in both extracts reacted with sIgA from vaginal washes when studied by Western blotting. Non-adsorbed vaginal washes were slightly more reactive with the extract from cells grown at 37 ~ than at 25 ~ However, the pattem of reactivity in both extracts was very similar. Secretory IgA reacted with polydispersed materials of molecular masses > 80 kDa (Figure 2). Well-defined bands were seen for proteins of molecular masses