Differential Candida albicans lipase gene expression during alimentary tract colonization and infection

Abstract

The human pathogenic fungus Candida albicans, which can reside as a benign commensal of the gut, possesses a large family of lipase encoding genes whose extracellular activity may be important for colonization and subsequent infection. The expression of the C. albicans lipase gene family (LIP1–10) was investigated using a mouse model of mucosal candidiasis during alimentary tract colonization (cecum contents) and orogastric infection. LIPs4–8 were expressed in nearly every sample prepared from the cecum contents and infected mucosal tissues (stomach, hard palate, esophagus and tongue) suggesting a maintenance function for these gene products. In contrast, LIPs1, 3, and 9, which were detected consistently in infected gastric tissues, were essentially undetectable in infected oral tissues. In addition, LIP2 was expressed consistently in cecum contents but was undetectable in infected oral tissues suggesting LIP2 may be important for alimentary tract colonization, but not oral infection. The host responded to a C. albicans infection by significantly increasing expression of the chemokines MIP-2 and KC at the site of infection. Therefore, differential LIP gene expression was observed during colonization, infection and at different infected mucosal sites.

Mucosal candidiasis, Colonization, Virulence factor, Animal model

1 Introduction

Approximately 35–80% of all humans are asymptomatic carriers of Candida species in the oral cavity, gastrointestinal tract and female genital tract [1,,3]. The consequence of such carriage frequency is that Candida albicans is the leading cause of fungal infection in humans with the vast majority of infections derived from endogenous origin [4,,6]. Candidiasis occurs most frequently in immunocompromised hosts, intensive care patients, and patients undergoing chemotherapy; however, invasive candidiasis in non-immunocompromised patients is becoming more prevalent [6]. Candida is the fourth leading cause of nosocomial bloodstream infections in the United States, preceded only by coagulase-negative staphylococci, Staphylococcus aureus, and enterococci [7,8]. Even with current antifungal therapy, systemic candidiasis is associated with an extremely high mortality rate of approximately 40%[9]. While non-life-threatening, mucosal candidiasis is the most common type of C. albicans infection that is frequently encountered in diabetic patients, patients taking antibiotics, organ transplant recipients, and immunocompromised patients. In particular, oropharyngeal candidiasis occurs in 90% of HIV-infected patients at some point during the course of infection [10] and is an important clinical marker predicting increasing immunosuppression [11,12].

Secreted bacterial extracellular enzymes such as lipases, in addition to disrupting tissues and providing nutrients, may also play an important role during infection by modulating the immune response [13] and by inhibiting the chemotactic and phagocytic activity of monocytes, macrophages, and granulocytes [14,,16]. In comparison to bacterial systems, little is known about the role of fungal lipases during infection although a large family of lipase genes (LIP1–10) has recently been identified in C. albicans[17]. Indirect evidence suggests some fungal lipases may be more important than others during infection since some of the lipase genes were expressed differentially during systemic infection and during in vitro infection of reconstituted human epithelial cells [18]. In contrast, much work has been performed delineating the role and function of another large family of C. albicans secreted virulence factors, the secreted aspartyl proteinases (Saps). Combined work on targeted gene disruptions and gene expression studies have shown different roles for Sap-encoding genes during the course of infection, in different environmental niches, and in different infections caused by Candida[19,,,22]. Importantly, Naglik et al. [23,24] also showed that some of these virulence factors are expressed preferentially in patients with candidiasis, rather than asymptomatic Candida carriers.

Herein, we report on the lipase genes that are expressed by C. albicans colonizing the alimentary tract and infecting mucosal tissues. The aims of this study were: (1) to investigate whether LIPs were expressed in both colonizing and tissue invasive Candida cells; (2) to investigate whether LIPs were differentially expressed at various mucosal sites, and (3) whether the host elicited a chemotactic response to C. albicans lipase expression and infection. In summary, we examined the expression of the C. albicans putative virulence LIPs(1–10) gene family during colonization (cecum contents) and from infected-orogastric tissues (tongue, hard palate, esophagus, and stomach). In addition, we also examined the host's chemokine (KC and MIP-2) response at the site of infection.

2 Materials and methods

2.1 Microorganism and media

C. albicans SC5314 was maintained by monthly transfer on Sabouraud dextrose agar (SDA). Stock cultures were maintained at 4 °C after growth for 24 h at 37 °C. Germfree mice were orally inoculated with C. albicans harvested from cultures grown in Sabouraud dextrose broth (SDB) for 24 h at 37 °C.

2.2 Immunodeficient Tgɛ26 mice

All mice used in this study were derived into the germfree state by caesarian section and maintained at the University of Wisconsin Gnotobiotic Research Laboratory. A germfree colony of immunodeficient Tgɛ26 mice was established with mice obtained from Dr. C. Terhorst (Harvard Medical School, Boston). The Tgɛ 26 mice were originally generated by overexpressing the full-length CD3ɛ gene in C57BL/6× CBA/J mice. The transgenic mice are defective in NK cells and T cells [25].

2.3 Animal model of orogastric candidiasis

Under germfree conditions, mice were colonized (alimentary tract) by oral inoculation with a pure culture of C. albicans SC5314 (10⁶ cells/ml for 5 h in the drinking water). The germfree mice became colonized within 24 h as verified by culturing the fecal pellets onto SDA. Mice were euthanized by CO₂ inhalation and infected mucosal tissues (stomach, tongue, palate, and esophagus) or ceca were collected aseptically. Intestinal colonization (stomach and cecum) with C. albicans was quantified by homogenization in phosphate-buffered saline (PBS) using the Stomacher II (Fisher) on a high setting for 120 s. Stomachs, after the contents were removed, were rinsed three times with PBS prior to homogenization thereby ensuring that the number of colony forming units (CFUs) represented the number of cells in the infected stomach tissue. Serial dilutions of the homogenate were prepared and plated on SDA. Colonies were counted after incubation at 37 °C for 24 h and are presented as the log₁₀ viable C. albicans CFU/g (dry weight) of tissue.

2.4 Histopathology

Tissue samples were collected aseptically and tongue, palate, esophagus, stomach, liver, kidney, and spleen were fixed in 10% formaldehyde in PBS. The tissues were processed in graded (100%, 95%, 80%, and 70%) alcohol and xylene solutions and embedded in paraffin. Tissue sections (5 μm) were stained with Periodic Acid-Schiff stain for fungi. Histopathology of Candida-infected and control (germfree) tissues was ranked by a clinical pathologist (Dr. Thomas Warner) as follows: 0, no microorganisms seen in a high-powered field (HPF × 400); 1, 1–10 microorganisms (yeast and hyphae) per HPF; 2, 11–50 microorganisms per HPF; 3, 51–100 microorganisms per HPF; and 4, confluent microorganisms per HPF.

2.5 RNA preparation

Stomach contents were removed prior to homogenization thereby ensuring that the preparation reflected infected tissue only. Cecum contents, infected tissue (stomach, hard palate, esophagus, and tongue) and control germfree tissues were homogenized in ‘RNAwiz’ (Ambion) using a Tekmar tissumizer and subsequent glass bead protocol as described previously [26]. After treatment with DNase1, the quantity and quality of total RNA isolated from each tissue were measured spectrophotometrically at an absorbance of 260 and 280 nm.

2.6 Reverse transcription-polymerase chain reaction (RT-PCR)

Equal amounts of total RNA isolated from C. albicans-infected and germfree control tissues were reverse transcribed into cDNA using the retroscript kit (Ambion). The absence of genomic DNA contamination was tested by performing PCR analysis with samples that lacked reverse transcriptase and also by using primers for two intron-containing genes (EFB1, KC). PCRs were also performed in the absence of template. PCR products were not obtained from these control reactions. PCR analysis was performed with the primers described in Table 1. Primers specific for the mouse glyceraldehyde-3-phosphate dehydrogenase gene (GAPDH) was used as an internal loading control for mouse mRNA while the C. albicans housekeeping EFB1 gene was used as an internal mRNA control for C. albicans[17,19,26]. Primers for EFB1 amplification were designed to span an intron of 365 bp in size [27]. EFB1 transcripts derived from complementary (c) DNA resulted in a RT-PCR product of 242 bp and were reproducibly detected from Candida-infected tissues. The PCR conditions were as follows: after an initial denaturation at 95 °C for 2 min, the samples were subjected to 33 cycles of denaturation at 95 °C for 30 s, annealing at 54 °C for 30 s, and extension at 72 °C for 60 s, with a final extension at 72 °C for 10 min. PCRs contained 2 mM MgCl₂ except for the LIP8 reaction which contained 1.5 mM MgCl₂. The identity of the RT-PCR products were confirmed by DNA sequencing.

1

C. albicans PCR primers used in this study

Gene	Primers (forward and reverse)	PCR product size (bp)
EFB1	5′-GAACGAATTCTTGGCTGAC	cDNA: 242
5′-CATCAGAACCGAACAAGTC	DNA: 607
LIP1	5′-AATTCACTGGGATCAAGAG	540
5′-TAAGTGACATGGACGTTAC
LIP2	5′-GCTGAAAATTGTTTGGCTG	450
5′-GGAAGTACAGGTGAGTAG
LIP3	5′-TCTTGTGGATGACTTCTAC	501
5′-AGCAACTTGGGCATCATC
LIP4	5′-ATTGCTTGGCTGATGGTG	419
5′-GTGGGATGTTTGGATATTC
LIP5	5′-GTGTTTGAGGAATTTGATG	325
5′-ATAGCTTCAGTGAGGTGAC
LIP6	5′-GTTAAACCTGGTGCCAAAG	453
5′-GGTACAAGAAATTCGATGC
LIP7	5′-CCATCATTCGAGACATTTC	456
5′-ATCCAAGTGGTTGTAATGC
LIP8	5′-TACCAACATTACTGCTACC	445
5′-GACCATTAGTACCATCTTC
LIP9	5′-AACAATCACTGCTTGACAG	475
5′-TGTTGTGCAACGACATCC
LIP10	5′-TAATCAGATCCGAAGACTC	607
5′-GAGCTTAACTTCACCATAC

2.7 Competitive RT-PCR

Competitive RT-PCR was performed using primers and methodology as described previously [28].

3 Results

3.1 Immunodeficient Tgɛ26 mouse model of orogastric candidiasis

Germfree immunodeficient Tgɛ26 mice, which are naturally susceptible to lethal orogastric candidiasis [29], were colonized with C. albicans SC5314 to mimic a natural route of infection. The alimentary tracts of the germfree mice were quickly colonized with C. albicans and remained colonized for the duration of the study since cultures of fecal pellets consistently yielded approximately 8 log₁₀ CFU/g. Viable counts from infected stomachs and cecum contents were high (approximately 7–8 log₁₀ CFU/g) 7 days after colonization and did not significantly change during the course of infection [26,30] (data not shown). Tgɛ26 mice generally lost 12–25% of their body weight within 21–28 days and did not survive past 5 weeks after oral inoculation. Death of the infected mice was associated with severe Candida esophagitis [29]. Histopathology of infected tissues revealed severe oroesophageal (tongue, palate, and esophagus) and gastric candidiasis after oral inoculation (Fig. 1). The histopathology scores indicated that the number of C. albicans cells within the infected tissue did not change significantly during the course of infection or between the different sites of infection (data not shown). Histopathology analysis and culture analysis did not detect any dissemination to the internal organs in these immunodeficient mice.

1

Histopathology of C. albicans infected stomach (A), hard palate (B), tongue (C), and esophagus (D) harvested from Tgɛ26 mice 2–4 weeks after oral association. The representative tissues shown were fixed in buffered formalin and stained using a standard Periodic Acid-Schiff reaction to detect the presence of infecting C. albicans (400× magnification).

3.2 Differential LIP expression during colonization and orogastric infection

RT-PCR was used as a specific, sensitive, and discriminating method to examine the expression of the closely related C. albicans LIP gene family during orogastric candidiasis. Expression of the LIP gene family was examined in C. albicans-infected orogastric tissues (stomach, tongue, palate, and esophagus) and during alimentary tract colonization (cecum contents) 21–28 days after oral association. LIP4 (26 out of 26 infected tissues and cecum contents analyzed), LIP5 (25/26), LIP6 (22/26), LIP7 (25/26), and LIP8 (22/26) were expressed in almost every sample analyzed, irrespective of the type or site of infection and irrespective of whether the cells were located in infected tissue or were colonizing the alimentary tract (Fig. 2, Table 2). Stehr et al. [18] has also recently shown constitutive expression of LIP4–8 during in vitro infection of reconstituted human epithelium. In contrast, LIP1 (6/6 stomachs), LIP3 (6/6), LIP9 (4/6) and to a lesser extent, LIP2 (2/6) and LIP10 (2/6), were expressed in gastric tissues while these LIPs were essentially undetectable in oral tissues. Indeed, LIP9 and LIP10 were only detectable in gastric tissues, although expression of LIP10 was sporadic and weak. Interestingly, LIP2 was expressed consistently in colonized samples but was absent in infected oral tissues and expressed only sporadically in infected gastric tissues. Conversely, LIP1 and LIP3 were expressed consistently in infected gastric tissues but to a much lesser extent during colonization (Table 2). Therefore, while LIPs4–8 were expressed in both colonized contents and infected mucosal tissues, LIPs1–3 and LIPs9–10 were expressed differentially during C. albicans orogastric infection and colonization. Importantly, LIP2 transcripts were detected in cecum contents, but were undetectable in infected oral tissues.

2

RT-PCR results for C. albicans EFB1 and LIP1–10 expression in oroesophageal tissues (esophagus, tongue, and palate), gastric tissues (stomach) and in colonized material (cecum contents) during orogastric candidiasis in immunodeficient Tgɛ26 mice. Tissues were harvested 21–28 days after inoculation with C. albicans. Lanes 1–10 represent RT-PCR products for LIP1–10, respectively. Results presented are representative from different sets of tissues as detailed in Table 2. The results from the stomach and cecum contents reflect data obtained from individual tissues; however, due to the smaller sample size, tongue, palate, or esophagus tissues were combined (two per sample) to obtain adequate amounts of starting material.

2

Summary of LIP gene expression during orogastric candidiasis 21–28 days after oral inoculation with C. albicans

Tissue	LIP gene expressiona
	LIP1	LIP2	LIP3	LIP4	LIP5	LIP6	LIP7	LIP8	LIP9	LIP10
Stomachs	6/6	2/6	6/6	6/6	6/6	6/6	6/6	5/6	4/6	2/6
Tongueb	1/5	0/5	0/5	5/5	5/5	3/5	4/5	2/5	0/5	0/5
Hard palateb	1/4	0/4	0/4	4/4	4/4	4/4	4/4	4/4	0/4	0/4
Esophagusb	3/5	0/5	1/5	5/5	4/5	3/5	5/5	5/5	0/5	0/5
Cecum contents	1/6	6/6	3/6	6/6	6/6	6/6	6/6	6/6	0/6	0/6

^aNumber of tissues positive for LIP expression/number of tissues analyzed.

^bIndicates that two tissues were combined for each sample tested in order to obtain adequate amounts of starting material.

3.3 Chemokine response to C. albicans infection

Since bacterial lipases have been shown to inhibit monocyte and granulocyte chemotaxis in vitro [14,15], competitive RT-PCR was used to analyze the chemokine (MIP-2 and KC) response at the site of infection (stomach and tongue) compared to germfree control tissues. There was no difference in MIP-2 or KC expression in infected gastric tissues compared to germfree control tissues 7 days after oral inoculation (Fig. 3A), even though histopathology scores and CFU counts from the stomach were high at this time point (data not shown); however, from 14 to 28 days after oral inoculation, expression of both KC and MIP-2 were greatly increased, ranging from 7- to 27-fold. In contrast, during oral candidiasis, KC expression was quickly and significantly increased at every time point analyzed (Fig. 3A). Relative RT-PCR was used to analyze MIP-2 transcripts from the same tissues since transcripts from the germfree control tongues were undetectable, making it unfeasible to measure using competitive RT-PCR. Both MIP-2 and KC displayed a similar pattern of expression during oral candidiasis since MIP-2 expression was quickly increased and sustained throughout the 28 day study (Fig. 3B). Uninfected kidney tissues harvested from germfree control and Candida-infected mice 28 days after oral inoculation were also examined for MIP-2 and KC expression by competitive RT-PCR; however, there was no significant difference in MIP-2 or KC expression in kidney tissue harvested from Candida-infected mice compared to germfree control mice (data not shown). The chemokine response to orogastric candidiasis appeared to be confined to the site of infection.

3

Time course analysis of chemokine expression in response to orogastric candidiasis in immunodeficient Tgɛ26 mice. Stomachs and tongues were harvested from germfree or Candida-infected mice at 7, 14, 21, and 28 days after oral inoculation with C. albicans. (A) Competitive RT-PCR analysis was used to analyze temporal MIP-2 and KC expression in stomach (St) and tongue (T) tissues compared to germfree control tissues. * indicates P < 0.05 (Student's t test) for tissues from infected mice compared to the tissues from germfree control mice. Values are means of three independent tissues ± the standard error. (B) Relative RT-PCR analysis of MIP-2 expression in tongue tissues. The analysis was performed using three independent tissues for each group. A PCR was also performed in the absence of cDNA template (lane 1). M, 100 bp DNA ladder.

4 Discussion

The Tgɛ26 model facilitates studies on mucosal candidiasis since the mice can be colonized and infected by a natural portal of entry (alimentary tract) and the disease resembles the infection seen in many immunosuppressed, organ transplant and AIDS patients. Since the alimentary tracts of the mice are colonized with a pure culture of Candida, it was also possible to compare the LIP gene expression profile of Candida's colonizing cells to tissue invasive cells. The latter distinction is vital in the identification of genes that may be important for colonization and infection since up to 80% of the population are asymptomatic carriers of C. albicans in the gut and at mucosal surfaces [1,,3]. Since Candida colonization of mucosal surfaces is regarded as a leading risk factor for infection [6], the identification of genes important for colonization may provide novel targets for antifungal prophylaxis. To our knowledge, this is the first report comparing putative virulence factor expression during gastrointestinal tract colonization and infection. LIPs4–8 were expressed during both colonization and infection; therefore, while LIPs4–8 were expressed in vivo, their expression cannot be solely attributed to mucosal infection per se. Simultaneous expression of SAPs1–10 genes, another large family of extracellular virulence factors, was also detected in both cecum contents and infected gastric tissues (data not shown). Nevertheless, LIP1, 3, and 9 were expressed preferentially during gastric candidiasis compared to oral candidiasis suggesting that C. albicans was responding to the distinct environmental challenges encountered in each niche (for example, acid pH of the stomach versus the neutral pH in the mouth). In addition, a role for LIP2 in gastrointestinal colonization rather than oral infection may be inferred since its expression was detected consistently in the cecum contents but was noticeably absent during oral infection. Since we have previously shown that components of the innate immunity are expressed at higher basal levels in conventional mice with a complex microbial flora compared to germfree mice [28] (data not shown), the Tgɛ26 model of candidiasis is also appropriate for studying the hosts innate defense against C. albicans challenge. Bacterial lipases have been shown to inhibit granulocyte and monocyte chemotaxis and phagocytosis in vitro [14,15]. Our data indicate that C. albicans infection and subsequent LIP expression result in significant chemokine expression at the site of infection, suggesting that fungal lipase expression in vivo does not inhibit the host's ability to produce these chemokines.

In summary, basic studies on C. albicans lipase production in vivo could lead to innovative treatments for the prophylaxis and therapy of mucosal candidiasis.

Acknowledgments

We thank Dr. C. Terhorst (Harvard Medical School, Boston) for providing the Tgɛ26 mice and the University of Wisconsin Gnotobiotic Research Laboratory for deriving and maintaining the mice in a germfree state. We also thank Kimberly Bauer, Andrea Boan, Peter Nicholas, Emily Paulling and Philip Werner for technical assistance. This work was supported by MUSC institutional funds (Grant URC-24413) and by the National Institutes of Health (Grant DE-13968).