top of page
_edited.jpg

BLOG

Candida albicans Is an Immunogen for Anti–Saccharomyces cerevisiae Antibody Markers of Crohn’s Disease

Abstract

Background and Aims: Antibodies directed against oligomannose sequences α-1,3 Man (α-1,2 Man α-1,2 Man)n (n = 1 or 2), termed anti–Saccharomyces cerevisiae antibodies (ASCAs) are markers of Crohn’s disease (CD). S cerevisiae mannan, which expresses these haptens, is used to detect ASCA, but the exact immunogen for ASCA is unknown. Structural and genetic studies have shown that Candida albicans produces mannosyltransferase enzymes that can synthesize S cerevisiae oligomannose sequences depending on growth conditions. This study investigated whether C albicans could act as an immunogen for ASCA. Methods: Sequential sera were collected from patients with CD, systemic candidiasis, and rabbits infected with C albicans. Antibodies were purified by using chemically synthesized (Σ) ASCA major epitopes. These affinity-purified antibodies and lectins were then used to analyze the expression of ASCA epitopes on molecular extracts and cell walls of C albicans and S cerevisiae grown in various conditions. Results: In humans and rabbits, generation of ASCA was shown to be associated with the generation of anti–C albicans antibodies resulting specifically from infection. By using affinity-purified antibodies, C albicans was shown to express ASCA epitopes on mannoproteins similar to those of S cerevisiae. By changing the growth conditions, C albicans mannan was also able to mimic S cerevisiae mannan in its ability to detect ASCA associated with CD. This overexpression of ASCA epitopes was achieved when C albicans grew in human tissues. Conclusions: C albicans is one of several immunogens for ASCA and may be at the origin of an aberrant immune response in CD.

Abbreviations used in this paper

The pathogenesis of Crohn’s disease (CD) is based on a genetically determined loss of immune tolerance to luminal antigens initiating an uncontrolled inflammatory process.1 Antibodies to several antigens have been reported to be more or less specific for CD. Among the microbial antigens supporting an abnormal immune response in CD are yeast antigens.2 Main et al3 have shown that antigens of the nonpathogenic yeast Saccharomyces cerevisiae could discriminate between the antibody response of patients with CD and those with ulcerative colitis (UC). We subsequently identified yeast cell wall phosphopeptidomannan (PPM) as a discriminating antigen and showed that, among the complex repertoire of oligomannose structures present in PPM, the sequence α-1,3 Man (α-1,2 Man α-1,2 Man)n (n = 1 or 2) was the major epitope supporting the antibody response.4,5 A standardized immunoassay using PPM of S cerevisiae strain SU1 was developed for the detection of anti-mannan antibodies in sera from patients,4 and these antibodies were named anti–S cerevisiae antibodies (ASCAs). ASCAs were found to be present in 50%–60% of patients with CD, 10%–15% of those with UC,6 20% of healthy first-degree relatives of CD patients,7 and 0%–5% of controls.8 The ASCA test has been used by numerous groups, and these initial data have been confirmed.9,10 Recently, several studies have shown that ASCAs are associated with CD phenotypes corresponding to ileal disease, young age at onset, structuring as well as penetrating behavior, and multiple-bowel surgery.11,12 Although there have been extensive reports on ASCAs as markers of CD, the immunogen inducing the production of these antibodies is unknown. Because the major epitope recognized by ASCAs is a hapten,4,5 the search for the immunogen would involve micro-organisms likely to express these oligomannoses on a carrier molecule.

Although the ASCA epitope is associated with PPM and other mannoproteins (MPs) in S cerevisiae,13 it is unlikely that this exogenous yeast found in foodstuffs3 would be the unique origin of antibodies prevalent in CD patients and their first-degree relatives because these individuals come from different countries and have different ways of life.14 The closely related yeast, Candida albicans, can colonize all segments of the human digestive tract and is also a prominent opportunistic pathogen responsible for mucosal infection in human immunodeficiency virus–infected patients and nosocomial systemic infections,15 among which the majority originates from the gut.16 Recent research on Candida and candidiasis has shown that C albicans–host relationships are under the control of finely tuned mechanisms based on host adaptative and innate immunity on the one hand, and genetic modulation of virulence attributes leading the yeast to adapt to a wide variety of environmental conditions on the other.17 Some of these adaptations involve α-mannosylation of proteins (which contain ASCA epitopes), which is under the control of mannosyltransferases (MNTs)18 homologous to those of S cerevisiae.

This background led us to explore the possibility that C albicans infection could act as an immunogen for ASCA. A step-by-step study was therefore designed based on bioclinical data and basic experience in structure/immunoreactivity relationships of yeast oligomannoses in CD4 and candidiasis.19

Materials and Methods

Patients

Twenty-seven patients with CD (16 females, 11 males; 26–88 years of age), 17 patients with proven systemic candidiasis (7 females, 10 males; 1–79 years of age), 25 patients with UC (12 females, 13 males; 27–76 years of age), and 10 healthy controls (5 females, 5 males; 19–40 years of age) were included in the study. Sera from all subjects were collected from the Clinical Mycology Laboratory of the University Hospital of Lille. Diagnosis of CD or UC was based on clinical, endoscopic, and histologic criteria. The characteristics of the patients with candidiasis are shown in Table 1. These patients were hospitalized in various wards of our hospital in which a large number of predisposing conditions are known to promote development of systemic candidiasis. A diagnosis of candidiasis was established after the isolation and identification of C albicans from at least 1 blood culture or biopsy material. These patients distributed in 3 categories. Category A consisted of 12 patients from which 1 serum drawn during C albicans infection exhibited significant anti–C albicans antibody levels. Category B consisted of 5 patients for which regular serum sampling was made during the time course of C albicans infection. Category C consisted of 3 patients for whom the diagnosis of candidiasis was made by histopathology on biopsy specimens dehydrated and embedded in paraffin that were still available for further immunostaining of C albicans pathogenic phase.

Yeasts and Culture Conditions

S cerevisiae strain SU14 and C albicans (serotype A) strains VW3220 and CAF2-121 were used throughout the study. Two S cerevisiae mutants, mnn1Δ and mnt4Δ, in which α-1,3-MNT had been inactivated, MNN1 and MNT4, respectively, and the corresponding parental wild-type strain, S288C, a gift from Professor Dujon (Institut Pasteur, Paris, France), were used in some experiments. Before the experiments, the yeasts were cultured for 20 hours on Sabouraud dextrose agar (SDA) at either 28°C or 37°C. C albicans hyphal formation was induced for 3 hours at 37°C in RPMI 1640 (Gibco, Germany).

Animals

Six New Zealand white rabbits (2–3 kg) were inoculated intravenously either with suspensions of live yeasts (500 μL of C albicans VW32 2.106 yeasts/mL; 2 rabbits and S cerevisiae SU1 2.108 yeasts/mL; 2 rabbits) or with suspensions of formalin-killed yeasts (C albicans VW32 2.106 yeasts/mL; 2 rabbits). The animals receiving killed C albicans yeasts or live S cerevisiae yeasts were injected weekly to maintain antigenic stimulation in the absence of pathogenic development. Serum samples were drawn every week for 12 weeks and stored at −20°C.

Immunoglobulin G (IgG) fractions from rabbits were purified from sera on G-protein coupled beads according to the manufacturer’s instruction (MabTrap Kit; Amersham Biosciences, Upsala, Sweden). After elution with 5 mL elution buffer (MabTrap Kit) and neutralization with 100 μL neutralizing buffer, IgG were dialyzed by 2 passages with 1 mL phosphate-buffered saline (PBS) through a Centricon YM-30 (Millipore Corporation, Billerica, MA) and finally concentrated to 0.2 mL in PBS. Purified IgG were stored at −20°C.

Natural and synthetic antigens

Cell-wall phosphopeptidomannans from C albicans and S cerevisiae

PPMs were extracted from S cerevisiae or C albicans grown in neutral or acidic medium (the pH was adjusted from 6.0 to 2.0 by adding 6 M HCl) as described previously.22 Briefly, cell pellets were suspended in 0.02 mol/L citrate buffer and autoclaved at 125°C for 90 minutes. Suspensions were harvested, and the PPM was precipitated with Fehling’s solution. The precipitates were then dissolved in methanol/acetic acid and PPM was separated by centrifugation. The sugar concentration was estimated by the sulphuric phenol colorimetric method.23

Synthetic oligomannosides

Oligomannosides previously identified as ASCA major discriminating epitopes,4,5 Man α-1,3 Man α-1,2 Man (Man3), were constructed chemically (JM Mallet, CNRS, Paris, France).24 ΣMan3, corresponding to the structure (Man α-1,3 Man α-1,2 Man)8-carboxyoctyl, was used for inhibition in the test described later. To ensure a selective and easy coupling essential for enzyme-linked immunosorbent assay (ELISA) and specific antibody purification by affinity chromatography, ΣMan3 was coupled through the linker arm to poly-L-lysine (Sigma 2636, mol wt 30,000–70,000). ΣMan3-poly-L-lysine was ascribed to the structure α-1,3 Man α-1,2 Man α-1,2 Man O (CH2)8-CO-(poly-L-lysine).

Antibodies and antibody purification

Detection of anti-yeast cell-wall PPM antibodies

ASCAs were detected by ELISA by using plates coated with PPM from S cerevisiae SU1 as described previously.7 Antibodies to C albicans PPM were detected by using the commercially available Platelia Candida Ab test (BIO-RAD Laboratories, Marnes-la-Coquette, France) according to the manufacturer’s instructions.

Detection of anti-mannoside antibodies

An ELISA to detect antibodies against synthetic α-linked ΣMan3 was performed by coating microtitre plates with 200 μL ΣMan3-poly-L-lysine (10 μg/mL sugar) in 0.15 mol/L PBS for 24 hours at 20°C. After washing in PBS-Tween 0.2%, the plates were saturated with 0.15 mL/L PBS containing 5% dextrose and 0.6% bovine serum albumin (BSA). Serum samples (100 μL) diluted 1:400 were added and incubated for 1 hour at 37°C. After 3 washes in TNT (Tris-HCl 0.05 mol/L, NaCl 0.15 mol/L, Tween 20 0.1%, pH 7.5), peroxidase-conjugated polyvalent anti-human immunoglobulins were added for 1 hour at 37°C. Absorbance was read in a microplate reader after addition of tetramethylbenzidine. The reactivity of sera was expressed as a percentage of the highest reactivity of a pool of sera from CD patients strongly reacting with oligomannoses.

Affinity purification

Two to 5 mg ΣMan3-poly-L-lysine in 1 mL coupling buffer (0.1 mol/L NaHCO3 pH 8, containing 0.5 mol/L NaCl) were incubated for 2 hours at room temperature with 1 mL activated CH Sepharose beads (Amersham Biosciences). After centrifugation, the remaining active groups were blocked with 0.1 mol/L Tris-HCl, pH 8, for 1 hour. ΣMan3-poly-L-lysine-coupled beads were finally washed with 3 cycles of alternating pH, with acetic buffer (pH 4) and 0.1 mol/L Tris-HCl (pH 8), and then equilibrated with coupling buffer. Serum or purified IgG diluted 1:4 in PBS was incubated with ΣMan3-poly-L-lysine-coupled beads overnight at 4°C. After several washes with PBS, bound antibodies were eluted with 0.1 mol/L glycine-HCl, pH 2.4, and neutralized with 0.1 mol/L Tris-HCl, pH 8. Purified material was finally passed through a Centricon (Millipore Corporation) for dialysis and concentration and resuspended in 0.2 mL PBS.

Immunologic and immunochemical methods

Whole yeast extract

Yeast cells were grown on SDA and extracted by alkaline extraction under reducing conditions as described previously.25 Extracts were analyzed by SDS-PAGE on an 8%–20% acrylamide gel slab. Membranes were probed with either human or rabbit sera or purified immunoglobulins (diluted 1:5000) and then incubated with a 1:5000 dilution of peroxydase-conjugated goat anti-human IgGAM or anti-rabbit IgG. After washing, the membrane was incubated with ECL detection reagents (Supersignal; West Pico Luminol, Interchim, France) and exposed to hyperfilm ECL.

Immunofluorescence assay

Fifty microliters containing 106 yeasts of S cerevisiae, C albicans, or C albicans hyphae were dropped onto 10-well microscope slides. The slides were saturated with PBS containing 1% BSA for 30 minutes at room temperature. Yeasts were then incubated for 60 minutes at 37°C with purified anti-ΣMan3 immunoglobulins (dilution 1:50). After 3 washes, a 1:100 dilution of fluorescein isothyiocyanate (FITC)-conjugated goat anti-human IgGAM was added for 1 hour at 37°C. Samples were washed and mounted for microscopic examination.

Tissue sections from human biopsies were cleared in toluene and rehydrated in graded alcohol followed by water. After blocking unspecific sites with PBS containing 1% BSA for 30 minutes at room temperature, sections were incubated with either purified anti-ΣMan3 immunoglobulins (1:50) or biotinylated-Galanthus nivalis lectin (GNL; Vector Laboratories, Burlingame, CA), specific for α-1,3 linked mannose,26 diluted 1:250 for 60 minutes at 37°C. After 3 washes with PBS-BSA 1%, sections were incubated with FITC-conjugated goat anti-human IgGAM or FITC-labelled streptavidin for 60 minutes at 37°C. The sections were then washed and mounted for microscopic examination. Fluorescence per fungal cell was quantified as a mean of optical density with Leica Qwin software (Leica Microsystems, Cambridge, UK).

Statistics

Statistics were performed by using SPSS software (SPSS Inc, Chicago, IL). Pearson correlation coefficients were calculated to determine the relationship between the reactivity of sera against PPM from S cerevisiae and PPM from C albicans.

Results

Bioclinical Evidence for a Relationship Between Generation of ASCA and Anti–C albicans Antibodies in Patients With CD and Candidiasis

Although ASCAs are usually stable serologic markers of CD, it was possible to select 3 patients from our diagnostic serum bank who presented ASCA seroconversion (from negative to positive) during their serologic monitoring. The clinical characteristics of these patients are summarized in Table 2. Sequential analysis of ASCA titers in these 3 CD patients is shown in Figure 1A. During the follow-up period of between 400 and 1000 days, ASCA titers against S cerevisiae PPM became significant for CD according to the test cut-off value. When the same sera were tested against C albicans PPM, all patients had antibody titers significant for C albicans infection. These titers decreased when ASCA levels became significant.

To verify that a C albicans infection could generate ASCAs, sera were selected from 17 patients with documented systemic candidiasis and significant levels of anti–C albicans PPM antibodies (Figure 1B). When tested with S cerevisiae PPM, all sera presented ASCA titers considered to be significant or highly significant for CD. For 5 of them (50 sera), the antibody response against both antigens was followed during the period of C albicans infection. ASCAs and anti–C albicans PPM antibody levels were strongly correlated (r = 0.766, P < .0001). Examples of representative kinetics are shown on Figure 2. By contrast to what was observed in CD patients, ASCA response in patients with candidiasis was only of short duration and restricted to the infection period. Taken together, these results show that C albicans infection could generate ASCAs that disappear after the infection period, whereas they persist in CD patients having exhibited significant anti–C albicans antibody titers.

ASCA Are Generated by C albicans Experimental Infection and Are Specific for the Major Epitope Recognized in CD

To verify the observations in humans that C albicans infection generates ASCAs, antibody responses to C albicans and S cerevisiae PPMs were analyzed in infected rabbits. As shown in Figure 3A, animals infected with live C albicans yeasts developed an antibody response against both antigens. By contrast, only anti–C albicans PPM antibodies were raised in rabbits injected with C albicans yeast killed by formalin (Figure 3B). The controls, consisting in rabbits infected with live S cerevisiae yeasts, exhibited only an ASCA response (Figure 3C). The specificity of the ASCA response was studied by ELISA through inhibition of binding to S cerevisiae PPM by using the ΣASCA major epitope (Figure 4). The serum from rabbit infected with C albicans was absorbed by increasing concentration of the ΣASCA major epitope and the remaining ASCA activity was analysed by ELISA. Inhibition of binding to S cerevisiae PPM was 90% using ΣMan3, showing that C albicans infection generates a strong and specific ASCA response. Taken together, these results suggest that C albicans is a strong immunogen for ASCA and that the ASCA epitope is specifically expressed during C albicans infection.

The CD Epitope That Is Recognized by Antibodies Produced During C albicans Infection and During CD Is Present on Both C albicans and S cerevisiae 120-kDa MPs

To identify the C albicans molecules that contain the Man3 epitope, antigenic extracts of C albicans were probed in Western blots with sera from patients with candidiasis, sera from rabbits infected with C albicans, and sera from CD patients with ASCAs. All sera reacted with C albicans antigens presenting with the typical polydispersion in gels, which is characteristic for MPs. Among the MPs, a 120-kDa antigen was recognized constantly. When the same sera were used to probe a S cerevisiae extract, similar reactivity was observed with a 120-kDa MP.

To determine whether these 120 kDa MPs supported anti-Man3 specificity associated with CD, immunoglobulins specific for the Man3 epitope were affinity purified from total rabbit IgGs or from total CD patients’ sera and tested by Western blotting. These purified immunoglobulins reacted in ELISA with both S cerevisiae PPM and ΣMan3-poly-L-lysine covalently linked to microtitre plates (data not shown). When these specific anti-ΣMan3 Igs were used to probe C albicans (Figure 5Aa) and S cerevisiae (Figure 5Ab), reactivity was restricted to the 120-kDa MP of both species. The presence of epitopes containing terminal α-1,3-linked mannose, at the level of both C albicans and S cerevisiae 120-kDa MPs, was confirmed by their reactivity with GNL (data not shown).

Because the synthesis of this determinant of the ASCA epitope is under the control of specific MNTs in S cerevisiae, the effect of specific gene disruption in S cerevisiae on 120-kDa MP ASCA epitope expression was determined. The reactivity of immunopurified anti-ΣMan3 antibodies from CD patients was then tested by using S cerevisiae mutants that do not express α-1,3 MNT: mnn1Δ and mnt4Δ (Figure 5B). As controls, 2 strains of S cerevisiae were used, the mutant parental strain S288C and the SU1 strain used in the original ASCA test, to reveal and discriminate serologic responses of patients with CD. With both mutants, extracts (Figure 5B, lanes 3–4), lower anti-ΣMan3 antibody reactivity was observed with the 120-kDa antigen compared with the reactivity with wild-type strain S288C extracts (Figure 5B, lane 2). This was more obvious with mnn1Δ (Figure 5B, lane 3) than mnt4Δ (Figure 5B, lane 4). Interestingly, when compared with the parental strain S288C, the reactivity of anti-ΣMan3 antibodies to the 120-kDa MP of the SU1 strain was stronger (Figure 5B, lane 1), confirming that there is a larger amount of CD-specific epitope in this strain.

These results show that C albicans is able to synthesize the ASCA epitope at the level of a 120-kDa carrier molecule analogous to a 120-kDa S cerevisiae MP and provide evidence that C albicans may be an immunogen for ASCAs. However, a basic characteristic of yeast oligomannoses is their possible expression on various carrier molecules and modulation of their expression according to the growth conditions. In S cerevisiae, the CD epitope is expressed on both the 120-kDa MP and PPM. The effect of modulating the growth conditions on CD epitope expression by C albicans PPM was therefore investigated.

The CD Epitope Present in S cerevisiae PPM May Be Overexpressed in C albicans PPM Depending on the Growth Conditions

Structural studies on C albicans PPM have shown that the expression of terminal α-1,3 mannose residues is enhanced under acidic growth conditions.27 The reactivity of sera from CD and UC patients displaying different ASCA titers was therefore determined by ELISA by using PPM from C albicans grown at neutral or acidic pH. As shown in Figure 6A, a strong correlation was observed between reactivity of sera against S cerevisiae PPM and that against C albicans PPM grown at pH 2 (r = 0.699, P < .001). This shows that C albicans can mimic antigenic characteristics of S cerevisiae depending on the growth conditions.

Overexpression of ASCA Major Epitopes in C albicans Occurs When the Yeast Infects Human Tissues

Because the serologic results obtained with sera from patients and rabbits strongly suggested that expression of the ASCA epitope occurs during infection, biopsies from humans with candidiasis were probed with immunopurified ASCA-specific antibodies and GNL, and the results were compared with the cell wall expression of the same oligomannose epitopes when S cerevisiae and C albicans were grown in vitro (Figure 7). For yeasts grown in vitro, more intense and uniform surface binding of the specific probes was observed with S cerevisiae (Figure 7A, panel 1), whereas binding was fainter and patchier with C albicans, suggesting a heterogeneous distribution of the oligomannosyl detected (Figure 7A, panels 2–4). As shown in Figure 7B, when the same ASCA-specific probes were used to label C albicans infecting human tissues, very intense labelling was observed. These results, which were confirmed by quantitative analysis of fluorescence signals (data not shown), show that specific overexpression of the ASCA epitope is triggered by pathogenic development of C albicans in human tissues supporting the intense specific ASCA response observed during human candidiasis.

Discussion

The main result of this study is that C albicans was found to express the major ASCA epitopes on several cell wall molecules. Overexpression of ASCA epitopes by C albicans was shown to be triggered by growth conditions. Among these, the pathogenic phase of C albicans was shown to be a strong immunogen for ASCA.

Initial clinical studies performed by using different yeast species and strains revealed the presence of antibodies to both C albicans and S cerevisiae in sera from patients with inflammatory bowel disease with a better ability of anti–S cerevisiae antibodies to discriminate CD versus UC.4,28 These observations led to the development of the initial ASCA test using the S cerevisiae mannan as an antigen.6 In the present study, a dynamic connection between anti–C albicans response and ASCAs was suggested for the first time by the unique observation of three CD patients who developed an antibody response to C albicans before the appearance of ASCA. This led us to explore if a C albicans infection could generate ASCAs. This was shown in patients with documented candidiasis and presenting high anti–C albicans antibody titers as a consequence of this C albicans infection.29 All of them displayed ASCA titers considered as significant for CD.6 Futhermore, a strong correlation between ASCA and anti–C albicans PPM antibodies was observed during the follow-up of 5 patients with candidiasis. The same kinetics was confirmed in rabbits infected by C albicans live yeasts, whereas rabbits injected by C albicans–killed yeast were unable to mount an ASCA response, strongly suggesting that ASCA synthesis resulted from C albicans pathogenic development.

These observations, made with S cerevisiae and C albicans PPMs, which are a mixture of oligomannoses, were then verified by using an ASCA major epitope synthesized chemically, Man α-1,3 Man α-1,2 Man.5 By using this epitope, it was possible to confirm the ASCA specificity of the antibody response induced by C albicans infection in rabbits. This epitope was then used to prepare ASCA-specific immunoglobulins from sera from rabbits as well as from patients with CD or candidiasis. An immunogenic candidate for ASCA, other than S cerevisiae molecules, namely a 120-kDa C albicans MP was then identified. Interestingly, this molecule was homologous in both its antigenicity and relative molecular weight to a 120-kDa MP of S cerevisiae prepared by using the same extraction procedure. Previous studies conducted to determine the predominant immunoreactive antigens of S cerevisiae involved in ASCA detection have already identified a 120-kDa glycoprotein and a 200-kDa glycoprotein,13,30 described as a major antigen. No evidence of reactivity against a 200-kDa MP was found in the current study. The reasons for these differences may be because of the different procedures used for antigen extraction. Whether the two 120-kDa MPs are actually homologous is the subject of further proteomic studies. However, the current investigation shows, for the first time, that an ASCA immunogen could reside in a C albicans molecule that shares an oligomannose motif with S cerevisiae.

The oligomannose sequence recognized by ASCAs in S cerevisiae is under the control of an α-1,3 MNT encoded by a gene designated MNN1 or MNT4, its putative homolog. The C albicans genome database shows that this species expresses homologs of mnn1, designated IPF8746/CA1548.31 In S cerevisiae, inactivation of either MNN1 or MNT4 caused a dramatic reduction in the reactivity of 120-kDa MP, confirming the importance of these genes in the expression of the ASCA epitope by this protein. These experiments also showed that, in comparison to S cerevisiae S288C, the parental strain used to construct the mutants, S cervisiae SU1 expressed the 120 kDa MP at a higher level, validating the use of this strain for the development of the original ASCA detection test.4 In contrast to S cerevisiae, C albicans is a diploid organism for which no mutant bank is available. However, both species have the ability to express the same oligomannose sequences on different molecules including MPs and PPM and to modulate their expression depending on the “environmental” growth conditions. Recent data from C albicans differential transcriptional analysis showed that IPF8746 is overexpressed at acidic pH (Mille et al, unpublished data, June 2005). This correlates with the NMR demonstration by Kobayashi et al27 that oligomannose chains containing 1 α-terminal 1,3-linked mannopyranose unit were increased markedly in C albicans PPM under acidic conditions. In the current study, it was observed that PPM extracted from C albicans grown in acidic conditions was recognized by ASCAs and therefore mimicked S cerevisiae PPM in its ability to detect ASCAs. This shows that at low pH, differential triggering of MNT gene regulation can lead C albicans to synthesize ASCA epitopes in a larger quantity and to present them within its PPM. Moreover, this also shows that expression of ASCA major epitopes by C albicans depends on the environmental conditions. Our experimental data show that only an injection of live C albicans cells generating an infection can induce ASCAs, whereas an injection of formalin-killed yeast cells does not. Because the ASCA epitope is polysaccharidic in nature and is not, by contrast to proteins, sensitive to formalin treatment, this shows that the ASCA epitope is primarily expressed during C albicans pathogenic phase. This explains why previous studies exploring antigenic heterogeneity of yeast species in their ability to reveal antibodies associated to CD failed to select C albicans instead of S cerevisiae4,32; at neutral pH used during in vitro experiments, the relevant epitopes are expressed by S cerevisiae but not by C albicans. By contrast, these epitopes are strongly expressed by C albicans cells under environmental conditions imposed by the host during pathogenic development. This modulation was confirmed when expression of the epitope was investigated in infected human tissue by using specific ASCA probes. In this case, labelling of C albicans invading host tissue was higher than that observed with C albicans or even S cerevisiae grown in vitro.

This study shows that C albicans is able to synthesize the same antigen as S cerevisiae and that C albicans infection is able to generate ASCAs. However, it does not prove that C albicans is the only or most important fungal source for the antigen nor does it indicate if the source of antigenic stimulation is dietary or from colonization. C albicans is an obvious potential enteric antigenic source. This species can be found in any segment of the gut from the oral cavity to the rectum.33 Around 50% of healthy individuals are colonized with C albicans in different segments of their gut. This percentage is higher when more sensitive methods of sampling and detection are used.33 At the level of the small intestine, studies have showed that 86% of duodenal aspirates contain Candida.34 C albicans can also infect virtually all host tissues depending on local or general defects in immune response/homeostasis. Candidiasis is the most frequent opportunistic infection (90%) in human immunodeficiency virus–infected patients and is now ranked fourth as an agent of nosocomial bloodstream infections,15 most of which originate from the gut and are favored by antibiotic treatment. Still, little is known about gut fungal flora and whether or not it is altered in patients with CD receiving or not immunosuppressors and/or antibiotics. The α-1,3 linked mannose at the terminal non reducing end of α-1,2 oligomannose chains is a glycan motif that is shared by yeasts other than C albicans and S cerevisiae. By using a basic alignment search tool for yeast sequences (http://www.ncbi.nlm.nih.gov/BLAST/Blast.cgi), we found that several yeast species present high degrees of homology with S cerevisiae MNN1. A subsequent review of the chemical literature revealed that the presence of the oligomannose sequence corresponding to the ASCA epitope has been established for several species using NMR, combined or not with the use of monospecific rabbit antisera. Among these is Candida glabrata, another yeast saprophyte of the human digestive tract that is also an opportunistic pathogen.35 Another source of antigenic stimulation could be from diet. Most strains of S cerevisiae used for food production (bread, beer, and wine) express the ASCA epitope. One clinical study has shown that exclusion of yeasts from the diet or, conversely, their administration reduces or increases the symptoms of CD in ASCA-positive patients, respectively.36 The ASCA epitope is also present in at least 2 species commonly involved in cheese production, Kluyveromyces lactis (soft cheese)37 and Debaryomyces hansenii (cheese rind).38

Whatever the source of stimulation, endogenous and/or exogenous, it is likely that CD patients are mounting, as well as described for bacterial antigens,12,39 an abnormal immune reaction to yeasts. A disturbed cellular response to S cerevisiae mannan has already been observed in ASCA-positive CD patients.2 It can be hypothesized that a lack of tolerance to C albicans could lead to ASCA formation and persistence in a subset of CD patients genetically predisposed. A recent study has shown an agreement in ASCA titers within concordant monozygotic twins, suggesting that the level of response is genetically determined, but no association with CARD15/NOD2 was found.40 Finally, from a basic point of view, experimental models of candidiasis have revealed the importance of CD4+CD25+ regulatory T cells and a role for control of inflammation versus protection.41 Interestingly, some anti–C albicans oligomannose antibodies can affect this balance42; these consist of anti–β-mannoside antibodies. These antibodies are protective, whereas anti–α-mannosides are not.43 Because ASCAs belong to this last category, the epitope shift they reveal could thus represent another level of immune failure in CD.

In conclusion, although the role of endoluminal bacterial antigens in CD has been widely investigated,39 the role of yeast antigens has been poorly explored.44 This study provides molecular clinical and experimental evidence that C albicans is able to induce the production of ASCAs, which recognize the immunodominant antigen of CD. Further clinical and basic studies are needed to assess whether this dynamic relationship between ASCAs and anti–C albicans antibodies, which suggests a “yeast connection” in CD, could help in our understanding and treatment of this disease.


Senaste inlägg

Visa alla

Comentários


bottom of page