|Home | About | Journals | Submit | Contact Us | Français|
In this work, the zebrafish model organism was developed to obtain a minivertebrate host system for a Candida albicans infection study. We demonstrated that C. albicans can colonize and invade zebrafish at multiple anatomical sites and kill the fish in a dose-dependent manner. Inside zebrafish, we monitored the progression of the C. albicans yeast-to-hypha transition by tracking morphogenesis, and we monitored the corresponding gene expression of the pathogen and the early host immune response. We performed a zebrafish survival assay with different C. albicans strains (SC5314, ATCC 10231, an hgc1 mutant, and a cph1/efg1 double mutant) to determine each strain's virulence, and the results were similar to findings reported in previous mouse model studies. Finally, using zebrafish embryos, we monitored C. albicans infection and visualized the interaction between pathogen and host myelomonocytic cells in vivo. Taken together, the results of this work demonstrate that zebrafish can be a useful host model to study C. albicans pathogenesis, and they highlight the advantages of using the zebrafish model in future invasive fungal research.
Candida albicans is an opportunistic fungal pathogen that commonly colonizes various anatomical sites in humans. It can become invasive and cause life-threatening infections in immunocompromised patients (33). Moreover, drug-resistant C. albicans continues to be a serious concern in the clinical setting. Therefore, obtaining a comprehensive understanding of C. albicans pathogenesis should improve medical therapy and facilitate the development of new antifungal drugs (3). Establishing an effective animal model system to study C. albicans-host interactions is critical for achieving these goals. The mouse model is the predominant animal host model used for C. albicans research. However, mice are difficult to use in large-scale studies due to the limited number of offspring produced and excessive experimental costs. To date, several invertebrate model organisms have been used to study C. albicans infection, including the wax moth (Galleria mellonella), the fruit fly (Drosophila melanogaster), and a nematode (Caenorhabditis elegans) (1, 10, 35). These simple minihosts have several advantages, such as conserved innate immunity, well-developed molecular tools, inexpensive care systems, and the possibility of supporting a high-throughput drug screening platform (27). A number of studies have utilized these minihosts to decipher the virulence mechanism of C. albicans infection (6, 27). For example, in the fly model, the ability of C. albicans to transition between the yeast and hyphal forms is critical for C. albicans virulence (7). The Toll mutant fly has been used for large-scale studies of genes involved in the pathogenesis of C. albicans infection, and C. elegans is used for in vivo screening of chemical libraries for antifungal activity (4, 8). However, these minihosts have immune systems with major differences from those of mammals, particularly a lack of adaptive immunity. The limitations of these minihosts highlight the need for other models to reveal the complex mechanisms of fungal pathogenesis.
Recently, the zebrafish (Danio rerio) has been increasingly used for biomedical research due to its high reproductive rate, comprehensive molecular tools, and low maintenance cost (48). The zebrafish is more similar to mammals than the minihosts mentioned above in terms of genetics, physiology, and anatomical structure, and importantly, it has both innate and adaptive immune functions (24). In addition to adult fish, zebrafish embryos have also been used in studies of bacterial infection (24). One advantage of the zebrafish embryo is its optical transparency, which permits real-time visualization of pathogen-host interactions. Furthermore, many transgenic zebrafish lines contain immune cells that are constitutively labeled with fluorescence reporters, so researchers are now able to track the progress of pathogens and responsive immune cells simultaneously in real time (5). Experimental approaches such as N-ethyl-N-nitrosourea (ENU) mutagenesis, morpholino knockdown, and microarray profiling could also help resolve outstanding problems with complex infections. Numerous studies have already utilized the zebrafish system to study the pathogenesis of various human infectious diseases, including diseases caused by bacteria and viruses (24, 38).
In this study, we tested the feasibility of using the zebrafish model system to investigate the pathogenesis of C. albicans. We demonstrated that introduction of C. albicans into adult zebrafish kills this aquatic host in a dose-dependent manner. The yeast-to-hypha transition of C. albicans was analyzed to determine the conserved pathogenic mechanism of C. albicans. Host immune responses to C. albicans infection were also determined. Finally, we examined other virulence factors important for fungal filamentation and the progression to host lethality, and we placed these observations in the context of other models of C. albicans infection.
The zebrafish used in this study were approximately 7 months old and weighed 0.25 to 0.35 g. They were maintained using guidelines described previously (44). Embryos were obtained from the wild-type AB, the TL, or the Tg(lyz:DsRED2)nz50 transgenic line (14, 44). The laboratory protocol was approved by the Institutional Animal Care and Use Committee of National Tsing Hua University.
The C. albicans strains used in this study are listed in Table Table1.1. A single colony from fresh YPD agar plates (1% yeast extract, 2% peptone, 2% dextrose, 1.5% agar) was inoculated into 5 ml of YPD broth and incubated at 30°C for 24 h with shaking at 180 rpm. Cells were harvested by centrifugation, washed once with sterile phosphate-buffered saline (PBS), and resuspended in sterile PBS. C. albicans cells were diluted with PBS and used for injection into zebrafish. For heat inactivation, the C. albicans cell suspensions were boiled at 100°C for 60 min.
Zebrafish were anesthetized by immersion in water containing 0.17 g/ml of tricaine (Sigma) and then intraperitoneally (i.p.) injected with 10 μl of PBS containing heat-inactivated C. albicans cells or C. albicans cells at different concentrations (1 × 108, 1 × 109, and 1 × 1010 CFU/ml) using a 26.5-gauge syringe (Hamilton Syringe 701N). To determine whether living C. albicans cells are required to kill zebrafish, 1 × 107 CFU of heat-inactivated C. albicans was also injected into peritoneal cavities. After infection, the fish were returned to the recovery tanks immediately and kept in separate 10-liter tanks in which the water was changed daily. The tanks were housed in an incubator with a controlled temperature (28.5°C) and with a cycle consisting of 14 h of light and 10 h of darkness. The fish were closely monitored, and mortality was determined every 3 h for 1 week.
Zebrafish injected with 10 μl of PBS or a C. albicans cell suspension (1 × 108 or 1 × 109 CFU/ml) were used to assess the fungal load in peritoneal cavities at 2, 15, and 23 h postinjection (hpi). The infected fish were sacrificed by immersion in ice water, and their organs and tissues were collected with sterile surgical blades and forceps. The samples were then rinsed twice with sterile PBS and homogenized in 0.5 ml sterile PBS using glass tissue grinders. The volumes of the homogenates were adjusted to 1 ml with sterile PBS in 1.5-ml microcentrifuge tubes. The number of CFU of C. albicans in each homogenate was determined by plating serial dilutions on YPD agar containing penicillin (100 U/ml), streptomycin (100 μg/ml), kanamycin (45 μg/ml), and chloramphenicol (30 μg/ml) and incubating the plates at 30°C for 24 h.
To examine the ability of mutant strains to form hyphae in zebrafish, a histological analysis was performed using fish killed with 1 × 108 CFU of C. albicans. The fish were fixed in Bouin's solution, and this was followed by decalcification with 0.5 M EDTA overnight at 42°C. The samples were sequentially dehydrated with alcohol and butanol and embedded in paraffin wax. Serial transverse tissue sections were stained with Periodic acid-Schiff stain and Meyer's hematoxylin. The samples were examined using an Axioskop 2 plus microscope (Carl Zeiss). The images were captured with an AxioCam HRc charge-coupled device camera and were processed with AxioVision 4.7 software (Carl Zeiss).
Control and infected fish (infected with 10 μl of 1 × 1010 CFU/ml C. albicans cells) were sacrificed by immersion in ice water at different time points and homogenized in liquid nitrogen. Total RNA was extracted using TRIzol reagent according to the manufacturer's instructions (Invitrogen). RNA quality was analyzed by gel electrophoresis. For cDNA synthesis, 6.25 μg of total RNA was pretreated with DNase I (Invitrogen) and then reverse transcribed by using the SuperScript III enzyme (Invitrogen) in a 50-μl reaction mixture. Quantitative real-time PCR was carried out using the 7500 real-time PCR system (Applied Biosystems). The primers used in this study are listed in Table Table2.2. Briefly, each 15-μl reaction mixture contained 25 ng cDNA, 7.5 μl SYBR green PCR master mixture (Applied Biosystems), 0.2 μM forward primer, and 0.2 μM reverse primer. The reactions were performed by using 1 cycle of 95°C for 10 min followed by 40 cycles of 95°C for 15 s and 60°C for 1 min.
Infection of zebrafish embryos with C. albicans was performed as previously described (21). Briefly, at 30 h postfertilization (hpf) zebrafish embryos were immobilized in a 1.2% low-melting-point agarose gel containing 0.17 g/ml tricaine. The embryos were microinjected with 4.6 nl of C. albicans OG1 cells (1 × 109 CFU/ml) either in the hindbrain of the AB strain or in the yolk of the Tg(lyz:DsRED2)nz50 strain. Each embryo was then soaked in filtered water at 28.5°C for 18 h. An Eclipse TE-2000-E fluorescence microscope (Nikon) or an A1R confocal microscope (Nikon) was used to capture each image in combination with a stage-top microincubator system (TOKAI Hit, Japan) that maintained 100% humidity and the temperature at 28.5°C.
For the fungal burden and quantitative real-time PCR analyses, two-tailed, unpaired t tests were used to assess the statistical significance of differences between two experimental groups.
The survival rates of infected fish were determined to evaluate the feasibility of using zebrafish as a model host for C. albicans infection. As shown in Fig. Fig.1,1, >95% of zebrafish injected with PBS and a low dose of C. albicans (1 × 106 CFU) survived for >80 hpi. In contrast, only 30% of the fish injected with 1 × 107 CFU of C. albicans were alive at >70 hpi, whereas none of the fish injected with 1 × 108 CFU survived for >50 hpi. Zebrafish injected with high doses of C. albicans exhibited apparent signs of illness, including prolonged bleeding, loss of mobility, and dwelling at the bottom of the tank. These fish gradually lost their upright reflex and the ability to swim and finally stopped moving their gill lids. All of the fish injected with heat-inactivated cells survived for >80 hpi, like the fish injected with the low dose of living cells or PBS (Fig. (Fig.1).1). These results indicate that whole, living cells of C. albicans kill zebrafish in a dose-dependent manner.
The ability to colonize and invade tissues within the host is critical for C. albicans infection. Therefore, colony formation and histological analyses were used to examine C. albicans within the zebrafish. As shown in Fig. Fig.2A,2A, the fungal burden in the group of fish injected with 1 × 107 CFU was higher than that in the fish infected with a lower dose of C. albicans (1 × 106 CFU). The number of C. albicans cells increased from 2 to 23 hpi, which indicated that there was cell proliferation. In particular, the numbers of cells in fish injected with 1 × 107 CFU of C. albicans increased more rapidly between 15 and 23 hpi. These results correlate well with the observation in the killing assay that infected fish started to die at approximately 25 hpi (Fig. (Fig.11).
Tissue sections derived from fish killed with 1 × 108 CFU indicated that C. albicans localized to various anatomical sites, including the liver, gastrointestinal (GI) tract, and muscle (Fig. (Fig.2B2B and and2C).2C). In addition, filamentous growth of C. albicans was observed in the connective tissue, muscles, liver, and GI tract (Fig. (Fig.2B,2B, ,2C).2C). Similarly, fungal filamentation was observed in fish injected with 1 × 107 CFU of C. albicans (data not shown). Similar observations of C. albicans colonization and filamentation at multiple loci have been reported for the mouse model and human patients (33). These results suggest that zebrafish is a promising model for studying virulence factors (such as morphogenesis) that contribute to C. albicans pathogenesis.
C. albicans can transition from a unicellular yeast form to distinct filamentous forms, including hyphae and pseudohyphae. This morphological transition is known to be an important C. albicans virulence factor (23, 46). Studies using molecular genetics and genome-wide analyses have identified several signaling and regulatory networks involved in the control of the hyphal transition (16, 29). During infection with C. albicans, the host immune system recognizes the pathogen-associated molecular patterns (PAMPs) of C. albicans and induces the expression of cytokines (32). To further examine the pathogen-zebrafish interactions, the expression profiles of selected hypha-related genes in C. albicans and host immune response genes were monitored by real-time quantitative PCR at 2, 8, and 15 hpi. ALS3, ECE1, HWP1, and PHR1 were selected as hypha-related genes based on a previous study (40). For correlation with the expression of these genes, the morphological progression of the yeast-to-filament transition was analyzed by performing a histological assay at corresponding time points. As shown in Fig. Fig.3,3, the expression levels of ALS3, ECE1, HWP1, and PHR1 were all upregulated as the time after injection increased. In particular, the expression of ECE1 showed a striking 1,200- to 3,700-fold increase compared to the expression in the control C. albicans culture. At 2 hpi, the yeast form of C. albicans colonized the surface of the zebrafish liver (Fig. (Fig.4A).4A). At 8 hpi, the yeast cells started to transform into the filamentous form and invade the liver (Fig. (Fig.4B).4B). At 15 hpi, the continuous elongated filaments of C. albicans were engaged in deep invasion of the liver (Fig. (Fig.4C).4C). On the other hand, in the infected zebrafish the expression of mRNA for interleukin-1β (IL-1β), tumor necrosis factor alpha (TNF-α), and inducible nitric oxide synthase (iNOS) was upregulated from 2 to 15 hpi. Interestingly, the level of expression of IL-10 peaked at 8 hpi and then decreased at 15 hpi, while the expression of gamma interferon (IFN-γ) was gradually downregulated from 2 to 15 hpi (Fig. (Fig.5),5), indicating that the host's immunoregulatory functions could also be detected. Together, these data demonstrate that the expression of C. albicans hypha-associated genes correlated positively with the progress of invasion in the infected zebrafish and that the host immune responses in the infected fish were indeed activated to defend against the fungal pathogen. These expression data are similar to results obtained for mice (32, 40). Collectively, our results strongly suggest that the zebrafish model system has the potential to be a useful tool for studying the C. albicans-host interaction at the molecular level.
Several stages have been proposed to describe the process of C. albicans infection, and virulence factors such as adhesion, hydrolytic enzyme secretion, morphological dimorphism, and phenotypic switching contribute to the progression of each stage (17, 28). Using the zebrafish model system, we tested whether the yeast-to-hypha transition plays a central role in C. albicans infection. To address this question, the pathogenicities of two mutant strains defective for filamentation (HLC54 [cph1/cph1 efg1/efg1] and WYZ12.2 [hgc1/hgc1]) and two partially reconstituted strains (HLC84 [cph1/cph1 efg1/efg1 EFG1] and WYZ12.1 [hgc1/hgc1 HGC]) were compared with that of the SC5314 strain (23, 47). CPH1 and EFG1 encode two transcription factors controlling C. albicans morphogenesis, possibly through distinct signaling pathways (23). The HGC1 gene product is a novel, hypha-specific G1 cyclin-related protein (47).
As shown in the killing assay (Fig. (Fig.6),6), the mutant strains were attenuated in virulence compared with the wild type, and the reconstituted strains partially restored the ability to infect the host. Around 50% of the zebrafish infected with 1 × 108 CFU of the wild-type strain were killed within 50 hpi, about 90% of these fish were dead at 100 hpi, and all of them were dead at 140 hpi (data not shown). When we decreased the infection dose by 1 order of magnitude, the survival rate of the zebrafish increased to 40%. However, 1 × 108 CFU of the HLC54 (cph1/cph1 efg1/efg1) and WYZ12.2 (hgc1/hgc1) mutant strains killed 50% of the infected zebrafish within 74 and 81 hpi, respectively, and 30% and 40% of the remaining zebrafish survived until 100 hpi, respectively. When zebrafish were infected by 1 × 107 CFU of either HLC54 or WYZ12.2, the survival rate was 90% at all time points. The virulence of the reconstituted strain containing HGC1 (hgc1/hgc1 HGC1) was close to that of the wild type, but the virulence of the reconstituted strain containing EFG1 (cph1/cph1 efg1/efg1 EFG1) was only partial. Around 50% of the zebrafish infected with 1 × 108 CFU of strain HLC84 (cph1/cph1 efg1/efg1 EFG1) survived until 63 hpi. To our surprise, two test doses of the reconstituted strain containing EFG1 resulted in survival rates (25% and 80%) that were not significantly different from those of the cph1/efg1 double mutant-infected groups. In a previous study, reconstitution with only EFG1 allowed the cph1/efg1 double mutant strain to form hyphae under serum stimulation conditions at 37°C, but it was not enough to restore virulence (23). Efg1 induces hypha-related genes to trigger the morphological transition, but it can become a negative autoregulator when it is phosphorylated (29, 39). Therefore, it was necessary to determine whether EFG1 always positively controls formation of hyphae during zebrafish infection.
The filamentous forms of HLC54 (cph1/cph1 efg1/efg1) and HLC84 (cph1/cph1 efg1/efg1 EFG1) could be identified within zebrafish (Fig. (Fig.7),7), and this finding is consistent with the results of previous studies of mice (9, 23, 36). In contrast, WYZ12.2 (hgc1/hgc1) did not form true hyphae within zebrafish; instead, the cells had a pseudohyphal morphology that included short tubes and multiple branches, while hyphal morphology was observed for the reconstituted strain containing HGC1 (Fig. (Fig.88).
In the killing assay described above (Fig. (Fig.6),6), most of the zebrafish infected with 1 × 108 CFU wild-type C. albicans started to die after 18 hpi, but the times of the initial observed death for groups infected with HLC54 (cph1/cph1 efg1/efg1), HLC84 (cph1/cph1 efg1/efg1 EFG1), WYZ12.2 (hgc1/hgc1), and WYZ12.1 (hgc1/hgc1 HGC) were 26, 40, 30, and 25 hpi, respectively. The delays in the time of the initial death may have resulted from hindrance of the morphological transition in zebrafish. Therefore, histological and hypha-related gene expression analyses were used to examine zebrafish infected with 1 × 108 CFU at 15 hpi, which is the early stage of infection. The wild-type cells formed hyphae at 15 hpi, while the cph1/efg1 double mutant remained in the yeast form (Fig. (Fig.9).9). To correlate the delay in the morphological transition at 15 hpi with the expression of several hypha-related genes, we monitored the ALS3, HWP1, ECE1, and PHR1 genes mentioned above, as well as SAP4-6 and the acid-pH-responsive gene PHR2, which is distinct from the neutral-pH-responsive PHR1 gene (26, 29). As shown in Fig. Fig.10,10, the levels of expression of ALS3, HWP1, ECE1, and SAP4-6 in wild-type C. albicans were higher than those in the cph1/efg1 and hgc1 mutant strains, but the levels of expression of both of the pH-responsive genes (PHR1 and PHR2) were not different in the different mutant strains. Previously, Zheng and Wang showed that deletion of HGC1 in C. albicans did not influence the expression of HWP1 or ECE1 with serum stimulation at 37°C (47). Similarly, in our zebrafish model, we observed no significant difference in the expression of HWP1, ECE1, or any of the selected hypha-related genes between the hgc1 mutant and the reconstituted strain (Fig. (Fig.10).10). The levels of expression of ALS3, ECE1, and HWP1 in HLC84 (cph1/cph1 efg1/efg1 EFG1) were much lower than those in HLC54 (cph1/cph1 efg1/efg1), while the levels of expression of PHR1 and PHR2 in HLC84 were higher than those in in HLC54 (Fig. (Fig.1010).
These results reveal that the timing of hyphal development is related to C. albicans virulence. They also indicate that the hgc1 mutant strain has more severe defects in hyphal formation than the cph1/efg1 mutant strain. The low-virulence phenotype of the cph1/efg1 mutant strain in zebrafish might result from delays in the morphological transition and the downregulation of hypha-related genes, including ALS3, ECE1, HWP1, and SAP4-6, as shown in this study. These genes affect virulence factors, such as tissue adhesion, cell wall component modification, and hydrolytic enzyme production. A deficiency in one of these genes could result in weak pathogenicity of C. albicans (30). Collectively, these results clarify the correlation between hyphal development and C. albicans virulence in zebrafish.
To further evaluate the use of zebrafish as a model system for characteristics other than morphogenesis, another clinically isolated strain of C. albicans, ATCC 10231, was included in our study. This strain is less able to invade the mouse liver and pancreas than the SC5314 strain (19). However, the exact molecular mechanism responsible for the difference in virulence between these two clinically isolated strains remains unclear. In the killing assay (Fig. (Fig.6),6), ATCC 10231 showed attenuated virulence in zebrafish compared to the SC5314 strain. In this assay, the zebrafish infected with 1 × 108 CFU of ATCC 10231 began to die after 28 hpi. Fifty percent of the infected zebrafish were dead by 40 hpi, and 15% survived until 100 hpi. When the infection dose was reduced to 1 × 107 CFU, the survival rate of the ATCC 10231-infected fish was 70% at 100 hpi (Fig. (Fig.6).6). The filamentous form of ATCC 10231 invading the liver could be observed in the dead zebrafish (Fig. (Fig.7)7) and in the infected fish at 15 hpi during the early stage of infection (data not shown). In a gene expression profile analysis, at 15 hpi there was not a significant difference in the levels of expression of the hypha-related genes ALS3, PHR1, and PHR2 between the SC5314 and ATCC 10231 strains, but the expression of ECE1, HWP1, and SAP4-6 was downregulated in ATCC 10231 compared to the expression in SC5314 (Fig. (Fig.10).10). Nonetheless, the filamentous transition of ATCC 10231 was not delayed at 15 hpi, suggesting that the morphological transition is not the most critical factor for the low virulence of ATCC 10231.
To evaluate the use of zebrafish embryos for studying C. albicans infection, we injected a green fluorescent protein (GFP)-expressing C. albicans OG1 strain into the hindbrain of a zebrafish embryo. At 18 hpi, the filamentous form of C. albicans could be readily observed in the live embryo. The hyphae extended continuously and eventually extruded out of the embryonic brain at 18 to 34 hpi (Fig. 11A). To detect the host immune response, we also injected the OG1 strain of C. albicans into the yolk of a lyz:DsRed2 transgenic zebrafish embryo in which the host myelomonocytic cells, including monocytes and granulocytes, endogenously expressed the red fluorescent protein DsRed2 (14). At 24 hpi, we found that filamentous C. albicans colocalized with myelomonocytes, which likely resulted from host immune cells that were recruited to the fungal infection site (Fig. 11B). Therefore, when image-based analysis is used, the zebrafish embryo is a useful tool to study host-pathogen interactions at the cellular level. Embryos also provide a window into aspects of disease mechanisms different from those in adult zebrafish.
In this work, we evaluated the use of zebrafish as a host model to investigate the pathogenicity of invasive C. albicans. We found that the C. albicans SC5314 strain killed adult zebrafish in a dose-dependent manner. Histological analysis then revealed that C. albicans colonized multiple anatomical sites in the adult zebrafish. Time-lapse serial sections and real-time quantitative PCR further showed that the progress of C. albicans filamentation was associated with the expression of several known hypha-specific genes and that the host responded to infection by activating immunity-related genes. Taking advantage of their optical transparency, we infected zebrafish embryos with fluorescently labeled C. albicans and obtained striking real-time visualization of C. albicans-myelomonocytic cell interactions. Together, our findings demonstrate that the zebrafish has the potential to be a useful model organism for studying C. albicans pathogenesis and the interactions of this fungus with its host.
To date, zebrafish have been used mostly to study interactions between hosts and bacterial pathogens (24, 38). For example, Neely and colleagues found that Streptococcus pyogenes (a human pathogen) and Streptococcus iniae (a natural pathogen of fish) infect adult zebrafish at similar efficiencies, and the comparable results support the conclusion that the Streptococcus-zebrafish model can be used to study human pathogenic bacteria (31). By employing large-scale, signature-tagged mutagenesis, 1,200 different mutant S. pyogenes strains were screened with the zebrafish infection model to identify virulence genes of the bacterial pathogen, which is important in S. pyogenes pathogenesis (18). The zebrafish model has also been used in studies of other bacteria, including studies of chronic tuberculosis caused by Mycobacterium marinum infection, real-time analysis of Salmonella enterica serovar Typhimurium infection, and analysis of Listeria monocytogenes-phagocyte interactions (22, 25, 37, 41). All of these examples indicate that the zebrafish model is a useful tool for studying infectious disease, and our study expanded its use to fungal pathogens.
Several in vivo studies of mammal models have indicated that deletion of CPH1 and EFG1 can restrict the dimorphic transition of C. albicans under many hypha-inducing conditions but that a corresponding double mutant strain can transition into the filamentous form in the model organism's kidney or tongue (9, 23, 36). To focus on the morphogenic transition and other aspects of C. albicans pathogenesis, we introduced two hypha-defective mutant strains, HLC54 (cph1/efg1) and WYZ12.2 (hgc1), and a noninvasive clinical isolate, ATCC 10231, into zebrafish. We found that the morphological transition of C. albicans correlated with the virulence of each mutant strain. In C. albicans infection of zebrafish, a lack of both CPH1 and EFG1 delayed filament formation and influenced the expression of several virulence genes, including ALS3, ECE1, and HWP1. Similar results have been obtained with the mouse model and in other in vitro studies. In addition, all of the results indicate that the reduced virulence of HLC54 is complicated and not caused simply by defects in morphogenesis (9, 29, 36). In contrast to HLC54, the HGC1 deletion strain did not form true hyphae during C. albicans infection in zebrafish, and it did not directly influence the expression of several of the downstream genes regulated by CPH1 and EFG1. The HGC1 deletion mutant cannot form filaments under various liquid hypha-inducing conditions, and hyphal growth of this mutant in the kidneys of mice is defective (47). Hgc1 is a G1 cyclin-related protein that binds to Cdc28 to induce hyphal development through regulation of apical growth and cell separation. Hgc1 also inhibits Ace2-activated genes, including the cytokinesis-related genes CHT3, SCW11, and DSE1, through the Efg1-dependent pathway (42, 43). Therefore, the HGC1 mutant strain might be a good candidate for studying morphogenesis-related C. albicans pathogenesis. We also found that the lower virulence of ATCC 10231 may be due to SAP4-6 downregulation, which reduces its invasiveness but not hyphal transition. Together, our zebrafish killing assay and histological examination suggest that the yeast-to-hypha transition plays an important role in C. albicans infection of zebrafish and support the hypothesis that the zebrafish can be a useful alternative model system for studying C. albicans virulence and pathogenesis.
The transparency of the zebrafish embryo provides major advantages in research because it permits in vivo image-based analyses that are difficult to perform in other models. As shown in Fig. Fig.11,11, the zebrafish embryo enables direct, real-time observation for analysis of the pathogenicity of C. albicans and the interaction of fungal and host immune cells. Different transgenic lines of zebrafish with tandem fluorescent labels for specific immune cell groups or organs are readily available to the research community (5, 24). In terms of the host immune system, the zebrafish macrophage appears at 26 h postfertilization (hpf), and functional T and B cells emerge at 7 and 21 days postfertilization (dpf), respectively (11, 12, 15). These findings suggest that transparent zebrafish embryos from 26 hpf to 7 dpf could be used to visualize real-time interactions between fungal pathogens and the innate immune cells of the host.
The physiological temperature (28.5°C) of zebrafish is lower than the body temperatures of murine or porcine hosts, which may explain why a higher dose of C. albicans is required in our zebrafish model and why the virulence of the hgc1 mutant strain observed in our assay was lower. Temperature is one of the environmental cues that stimulate the hyphal transition in C. albicans. This fungal pathogen prefers to stay in the yeast form at low temperatures (24 to 28°C), while increasing the temperature to 37°C induces a transition to the hyphal form (20). Therefore, the temperature of the zebrafish system (28.5°C) may affect the virulence of C. albicans. Despite the temperature issue, our results still strongly suggest that C. albicans can colonize, proliferate, undergo morphological transition, and invade deep into the organs of zebrafish. Recently, a transparent adult zebrafish with internal organs easily observable without equipment was constructed (45). With this zebrafish, the C. albicans infection process and the response of the zebrafish host can be monitored in real time. It could improve the study of C. albicans infection and provide new insights into the interaction of these organisms.
In summary, our results reveal that the pathogenicity of C. albicans can indeed be examined using zebrafish. This study validates the conclusion that the zebrafish is a valuable tool for studying this opportunistic fungal pathogen.
We thank Bon-Chu Chung, Jen-Leih Wu, and Sheng-Ping Hwang (Academia Sinica, Taiwan), Yi-Chuan Cheng (Chang Gung University, Taiwan), Wei-Yuan Chow (National Tsing Hua University), Shyh-Jye Lee and Yung-Shu Kuan (National Taiwan University, Taiwan), Yi-Wen Liu (Tunghai University, Taiwan), and Kathy Crosier and Phil Crosier (The University of Auckland, Auckland, New Zealand) for providing zebrafish for this study and Hsiu-Jung Lo (National Health Research Institute, Taiwan) and Yue Wang (Institute of Molecular and Cell Biology, Singapore) for providing C. albicans strains for this work.
This work was supported by grants NSC97-2627-B-007-003 and NSC98-2627-B-007-016 to Y.J.C. and grants NSC97-2627-B-007-002 and NSC98-2627-B-007-015 to C.Y.L. from the National Science Council (Taiwan, Republic of China). We also appreciate support from the NTHU-NHRI Zebrafish Core Facility.
Editor: G. S. Deepe, Jr.
Published ahead of print on 22 March 2010.