Cryptococcus neoformans Gene Involved in Mammalian Pathogenesis Identified by a Caenorhabditis elegans Progeny-Based Approach
http://www.100md.com
感染与免疫杂志 2005年第12期
Division of Infectious Diseases, Massachusetts General Hospital, Boston, Massachusetts 02114
Department of Molecular Genetics and Microbiology Division of Infectious Diseases
Department of Medicine Howard Hughes Medical Institute, Duke University Medical Center, Durham, North Carolina 27710
Edward A. Doisy Department of Biochemistry and Molecular Biology
Department of Molecular Microbiology and Immunology, Saint Louis University School of Medicine, Saint Louis, Missouri 63104
Department of Microbiology and Molecular Genetics, Harvard Medical School, Boston, Massachusetts 02114
ABSTRACT
Caenorhabditis elegans can serve as a substitute host for the study of microbial pathogenesis. We found that mutations in genes of the fungal pathogen Cryptococcus neoformans involved in mammalian virulence allow C. elegans to produce greater numbers of progeny than when exposed to wild-type fungus. We used this property to screen a library of C. neoformans mutants for strains that permit larger C. elegans brood sizes. In this screen, we identified a gene homologous to Saccharomyces cerevisiae ROM2. C. neoformans rom2 mutation resulted in a defect in mating and growth defects at elevated temperature or in the presence of cell wall or hyperosmolar stresses. An effect of the C. neoformans rom2 mutation in virulence was confirmed in a murine inhalation infection model. We propose that a screen for progeny-permissive mutants of microorganisms can serve as a high-throughput method for identifying novel loci related to mammalian pathogenesis.
INTRODUCTION
The nematode Caenorhabditis elegans is emerging as a substitute host for elucidating important mechanisms of fungal and bacterial pathogenesis, thanks to genetic tractability, anatomical simplicity, and the availability of extensive genetic and genomic resources (23, 35). The C. elegans killing assay used in previous studies entails observing a synchronized population of nematodes for kinetics of death on a lawn of the microbe being studied. Living and dead worms are counted at 24-h intervals, and the LT50, or time for 50% of the nematodes to die, is calculated (1, 7, 10, 15, 24, 38). This killing assay has been used to study effectively gram-negative (1, 15, 38) and gram-positive (10) bacteria, as well as the human pathogenic fungus Cryptococcus neoformans (24, 25).
Interestingly, C. elegans does not produce a sustainable brood of progeny in the presence of C. neoformans or some other pathogens (10, 25, 38). We hypothesized that a screen of brood size would circumvent the need for labor-intensive manual assessment for living versus dead nematodes. The aim of this study was to investigate the role of the C. neoformans virulence factors (such as capsule and melanin and their regulatory pathways) in C. elegans progeny production. Also, we evaluate the hypothesis that brood size can be used effectively as a facile marker for virulence in the C. elegans-C. neoformans system. Of note is that no progeny-based screen has been reported previously.
MATERIALS AND METHODS
Strains and media. The C. neoformans strains used in these experiments are summarized in Table 1 or described in the text. Yeast cultures were maintained as frozen stocks or on yeast-peptone-dextrose (YPD; Difco) agar plates and grown in YPD. C. neoformans cultures were grown at 30°C unless otherwise specified. Cryptococcus laurentii cultures were grown at 25°C. The standard C. elegans strain N2 Bristol was maintained at 15°C and propagated on Escherichia coli.
C. elegans killing assays were performed as previously described (24, 25) with minor modifications. C. neoformans strains were inoculated into 2 ml of YPD and grown at 30°C for 24 h; 10 μl of the culture was spread on 35-mm tissue-culture plates (Falcon) containing brain heart infusion (BHI) agar (Difco). The plates were incubated at 30°C for 48 h and then at 25°C for an additional 24 h. Of note is that by incubating the plates for an extra day at 30°C, killing of C. elegans was faster than what we reported in the past (24, 25). Ampicillin was added to the medium to prevent growth of E. coli OP50 that might have been carried over on transfer of worms to the yeast-containing plates. C. elegans animals at the L4 developmental stage were transferred from a lawn of E. coli OP50 on nematode growth medium to a lawn of the yeast to be tested on BHI medium, incubated at 25°C, and examined for viability at 24-h intervals with a Nikon SMZ645 dissecting microscope. Three plates with 50 L4 larval C. elegans nematodes per plate were used for each strain. Every 24 h, living and dead worms were counted, with living worms being transferred to a fresh, identical lawn of C. neoformans to allow observation of the original worm without obstruction by progeny.
Quantification of C. elegans progeny. Lawns for the quantification of C. elegans progeny assays were prepared as detailed above for the killing assays. The only difference is that C. laurentii plates were incubated at 25°C throughout the experiment (i.e., were not placed at 30°C as was initially done with C. neoformans plates). The reason for the different temperature is that 25°C is the optimal temperature for C. laurentii growth (25). However, special care was taken to ensure that the C. neoformans and C. laurentii lawns were similar in size and thickness.
Nematodes were placed on lawns of C. neoformans or C. laurentii on BHI agar at 25°C and transferred to a new, identical lawn every 24 h throughout the reproductive period. After removal of the parent nematode, each lawn was spread face down on a 100-mm tissue-culture plate (Falcon) with nematode growth medium agar and E. coli strain OP50. Care was taken not to intermingle OP50 bacteria with C. neoformans or C. laurentii. Numbers reported are the total progeny produced over the first 72 h of egg laying. No unhatched eggs were seen. Approximately 20 to 25 wild-type C. elegans N2 nematodes were used per group, and only nematodes that survived all 3 days of the experiment were included. Viable progeny was defined as a first-stage (or older) larva. Also, to compare progeny production between strain ATCC 208820 (2e-tuc4) that produces laccase and strain ATCC 208819 (2e-tu4) that is laccase negative we grew these strains in the presence of L-DOPA (3,4-dihydroxyphenylalanine) (100 μg/ml) and added L-DOPA (100 μg/ml) to the BHI medium (25).
Identification and construction of mutant strains. Random mutants of C. neoformans were generated using biolistic insertion of a plasmid containing the URA5 gene into the strain F99 (MAT ura5) (9, 40). In mutants permitting significant brood production, DNA flanking the insertion was identified using arbitrary primed PCR and compared to the H99 genome database at Duke University. Gene prediction was performed with the FGENESH algorithm, and predicted genes were compared to the GenBank database. A disruption allele for the ROM2 gene was constructed by overlap PCR to replace most of the coding region with a cassette conferring resistance to nourseothricin. Primers (all 5'-3') JOHE13032 (TCCTCTATCATCCCATCTAG) and JOHE13033 (GCTTATGTGAGTCCTCCCGTTGTTGGATTGCTGCTG) and primers JOHE13034 (CTCGTTTCTACATCTCTTGTCATGACGCACCACCAG) and JOHE13035 (GATGACGAATTTGCTTTACTAG) were used to amplify the flanks of the ROM2 gene from KN99 DNA, and primers JOHE8677 (GAAGAGATGTAGAAACGAG) and JOHE11866 (GGGAGGACTCACATAAGC) were used to amplify the NAT cassette. Equimolar amounts of the three products were mixed, and primers JOHE13032 and JOHE13035 were used to amplify a contiguous fragment; this fragment was transformed into strain KN99 (MAT) by use of a biolistic apparatus. Disruption of ROM2 was confirmed with PCR and Southern blot hybridization. Only strains containing a single insertion were used. The rom2 mutation was crossed to strain KN99a, and basidiospore progeny were selected to identify a rom2 MATa strain. To reconstitute ROM2, a wild-type copy of ROM2 was amplified with primers JOHE13031 (ACCTATCATCTCGCTGATCC) and JOHE13036 (GAGAAGGATACTGAAAAGGC) and cloned into a plasmid conferring resistance to neomycin, which was introduced into the rom2 mutant with Agrobacterium-mediated transformation followed by selection on neomycin-containing media (MediaTech, Inc., Herndon, VA).
In vitro evaluation of cryptococcal strains. Capsule production was assessed by exclusion of India ink following growth for 48 h at 30°C in low-iron media containing 56 μM EDDHA. Measurement of the capsule size (diameter of the edge of the capsule to the cell wall and cell diameter) was evaluated on an Olympus BX51 microscope using Olympus Microsoft software. Melanin production was assessed by growth on L-DOPA-containing minimal medium agar for 7 days at 30°C. For the mating assays, strains were each crossed by mixing strains and incubating on V8 media at 25°C in the dark for 14 days. For the study of the growth on cell wall and osmolar and other stressors (such as caffeine and NaCl), incubation times differed because growth rates differed under different conditions, but incubation was continued until a growth defect appeared or was plainly not forthcoming. All in vitro assays were performed at 25°C, 30°C, and 37°C. In all studies, results at 25°C and 30°C were similar unless otherwise specified.
Murine models of virulence. Experiments with mice were performed as previously described (24). Wild-type 8-week-old female ICR/CD1 mice (Charles River Laboratories, Wilmington, MA) were used in all studies. In brief, C. neoformans strains were grown at 30°C with shaking overnight in YPD to late-log phase. The yeast cells were centrifuged and washed in phosphate-buffered saline. The concentration of yeast cells in the inoculum was evaluated before inoculation by use of a hemocytometer and confirmed by plating serial dilutions and enumerating CFU. Mice were anesthetized by intraperitoneal injection of tribromoethyl alcohol (Aldrich, Milwakee, WI) (400 mg/kg of body weight) and suspended by the incisors on a silk thread. A 50-μl volume of the inoculum (1.0 x 106 yeast cells of either the rom2 mutant the parental strain KN99 or the reconstituted strain KN99 rom2 + ROM2) was slowly pipetted into the nares with continued suspension for 10 min. The murine protocols were approved by the Massachusetts General Hospital Committee on Research, Subcommittee on Research Animal Care, and special attention was given to minimize suffering of the mice.
Evaluation of tissue burden. Following euthanasia, we harvested the lungs from mice 12 days after administration of a cryptococcal inoculum by nasal inhalation. Tissues were weighed and homogenized in sterile phosphate-buffered saline by use of a Tissue Tearor (model 398; Biospec Products Inc., Racine, WI). Then, serial dilutions were plated on YPD agar containing 100 μg/ml ampicillin, 100 μg/ml streptomycin, and 45 μg/ml kanamycin. CFU of C. neoformans were counted after growth at 30°C for 72 h.
Statistical analysis. Survival assays in mice and nematodes were repeated three times, and each experiment independently gave statistically significant results. The figures and the P values provided herein are each from one representative experiment. Murine and C. elegans killing curves were plotted and estimation of differences in survival (log-rank and Wilcoxon tests) analyzed with the Kaplan-Meier method performed using STATA 6 statistical software (Stata, College Station, TX). The same software program was used for the statistical analysis of the CFU from murine organs and comparison of progeny production (Mann-Whitney and Kruskal-Wallis tests). Plotting and statistical evaluation of CFU data were done using Microcal Origin (OriginLab Corporation, Northampton, MA). A P value of less than 0.05 was considered to be significant.
RESULTS
Abrogation of brood production is associated with C. neoformans genes involved in pathogenesis. First we quantified C. elegans progeny produced over 3 days in the presence of wild-type C. neoformans and a ras1 mutant of C. neoformans that was previously found to be attenuated in C. elegans (25) and mammalian (2) killing. The nonpathogenic C. laurentii strain ATCC 76483 was used as a control. Both C. laurentii and the hypovirulent ras1 mutant of C. neoformans allowed C. elegans to produce greater numbers of viable progeny than did the wild-type C. neoformans KN99 or the reconstituted strain H99 ras1 + RAS1 (Table 1; P = 0.0001 and 0.0456, respectively). Levels of progeny production by the strain H99 and the isogenic strain KN99 were similar. In addition to the RAS1 signaling cascade, a second regulatory pathway that involves the G-alpha protein-cyclic AMP-protein kinase A signaling pathway has been described for C. neoformans. Interestingly, a C. neoformans pka1 mutant that displays attenuated virulence in mice (8) and C. elegans killing (25) did not allow progeny production that was statistically different from that by nematodes exposed to wild-type C. neoformans (Table 1).
In order to dissect the role of the major cryptococcal virulence factors (capsule and melanin production) in progeny production, we studied C. elegans brood production in the presence of an acapsular strain with a mutation involving gene CAP59 that is needed for capsule formation and a strain with a mutation involving the enzyme laccase that is essential for melanin production. Progeny production in the presence of the acapsular strain was significantly higher than the progeny production on lawns of wild-type C. neoformans (average ± standard error of 1.68 ± 0.4 for the KN99 group and 5.9 ± 2.0 for the H99 cap59 group; P = 0.0044). However, nematodes exposed to the acapsular strain cap59 made significantly less progeny compared to the nematodes exposed to ras1 (5.9 ± 2.0 for H99 cap59 compared to 17.06 ± 2.28 for KN99 rom2 and 13.5 ± 3.04 for H99 ras1; in both cases, P < 0.01). There was no difference in progeny production between the strain ATCC 208820 (2e-tuc4) that produces laccase and strain ATCC 208819 (2e-tu4) that is laccase negative (data not shown).
A C. elegans progeny-based screen. Based on the finding that some virulence factors of C. neoformans can cause abrogation of brood production in C. elegans, we investigated whether screening for brood production can be used to identify novel factors involved in C. neoformans pathogenesis in mammals. We screened a library of 1,500 random C. neoformans mutants for strains that permitted a C. elegans brood to develop on lawns of these mutants, a phenotype we termed "progeny permissive." We isolated three such mutants, a rate similar to that seen with existing screens utilizing killing of C. elegans by pathogens (10, 21, 24). Genetic analysis of one mutant strain revealed that the disrupted locus encoded a homolog of Saccharomyces cerevisiae Rom2p. S. cerevisiae Rom2p is a guanyl-nucleotide exchange factor for Rho1p and is required for yeast cell integrity under conditions of heat and osmolar stress (5, 31, 32).
Evaluation of a C. neoformans rom2 mutant in vitro. A mutant lacking all but the last 18 codons of the ROM2 locus grew similarly to the wild-type yeast at 30°C, but it demonstrated reduced growth at 37°C (Fig. 1A). All three strains grew similarly at 25°C (data not shown). Of note is that at 37°C, the rom2 mutant demonstrated a temperature-sensitive phenotype, and a delay in growth. Interestingly, the high-temperature growth defect was corrected by the addition of 1 M sorbitol to the media. Of note is that in spite of the delay in growth in YPD media, it finally reaches a concentration similar to that of the wild type even at 37°C (Fig. 1A) and that there was no increased lysis associated with the rom2 mutant, as noted using the BCIP (5-bromo-4-chloro-3-indolyl phosphate; Sigma) colorimetric assay (data not shown).
Melanin and capsule production were not affected by the rom2 deletion (data not shown), while mating was significantly impaired (Fig. 1B). Basidiospore production and hyphal projection were markedly diminished when MAT and MATa rom2 mutants were crossed, but the phenotype of the reconstituted strain KN99 rom2 + ROM2 was similar to that of the wild type. Of note is that the mating defect was less pronounced when we mated the rom2 mutant with the KN99 partner but did not revert to wild-type mating even with the addition of osmotic stabilizer to the media (data not shown).
The finding that a C. neoformans rom2 deletion mutant exhibits a defect in mating and a growth defect that can be suppressed by addition of the osmotic stabilizer sorbitol suggests that in C. neoformans ROM2 may play a role in the formation or structure of the cell wall, similar to findings with S. cerevisiae (32). In order to evaluate this hypothesis we examined the sensitivity of the C. neoformans rom2 mutant to cell wall and osmolar stressors.
First we examined the sensitivity of KN99 rom2 to caffeine, a phosphodiesterase inhibitor that has been used to screen for yeast mutants with altered cell wall proteins and/or structure (6). In S. cerevisiae, deletion of ROM2 renders cells sensitive to caffeine (32). Cells of KN99, KN99 rom2, and KN99 rom2 + ROM2 were grown on solid media containing caffeine. Plates were incubated at 25°C, 30°C, or 37°C. KN99 rom2 demonstrated sensitivity to 7 mM caffeine that was apparent at all temperatures tested (Fig. 2 and data not shown). Of note, at 37°C, 7 mM caffeine entirely inhibited growth of the mutant. The reconstituted strain, KN99 rom2 + ROM2, grew similarly to the wild type under all conditions, and no significant difference between the rom2 mutant, the wild type, and the reconstituted strain was seen on YPD containing 3 mM caffeine.
KN99 rom2 was also hypersensitive to 1.5 M NaCl at 25°C and 30°C, giving rise to very sparse colonies of tiny size (Fig. 2). Growth of the strain KN99 rom2 + ROM2 in 1.5 M NaCl was similar to the wild-type growth. Of note, at 37°C all three strains (the wild type, rom2 mutant, and reconstituted strain) failed to grow in the presence of 1.5 M NaCl; as a consequence, it was not possible to determine the relative sensitivity of KN99 rom2 to 1.5 M NaCl at this temperature.
A C. neoformans rom2 mutant is hypovirulent in C. elegans and mice. Strain KN99 rom2 permitted significantly higher numbers of C. elegans progeny to develop than did the wild-type yeast. More specifically, the KN99 rom2 mutant allowed significantly more progeny than KN99 or KN99 rom2 + ROM2 (P < 0.0001 and P = 0.0004, respectively; Table 1). Moreover, the KN99 rom2 mutant was significantly attenuated in virulence in C. elegans and demonstrated an impaired ability to survive passage through the nematode grinder organ (Fig. 3).
Interestingly, the rom2 mutant was also avirulent in an inhalation infection model in mice (Fig. 4) and was considerably impaired in establishing infection in murine lung tissue (Fig. 5). Of note, experiments were performed in triplicate and a total of 32 mice received KN99 rom2. No death or illness was observed for any of the mice that received KN99 rom2, including a total of 22 mice that were kept for 120 days postinoculation.
DISCUSSION
In this report we present the novel observation that some, but not all, mutations in C. neoformans genes involved in mammalian virulence allow C. elegans to produce greater numbers of progeny than when exposed to wild-type fungus. This progeny-permissive phenotype can be used to screen collections of C. neoformans mutants to identify genes involved in mammalian infection. The first gene identified using this approach is a gene homologous to Saccharomyces cerevisiae ROM2. In C. neoformans a rom2 mutant demonstrated a defect in mating and in growth at 37°C, as well as susceptibility to cell wall and osmolar stressors. Interestingly, ROM2 was essential for full virulence in an inhalation murine model of infection, suggesting that this screen is useful for elucidating pathogenesis in mammalian hosts.
This study extends previous work by describing a new technique, the progeny screen, for using the C. elegans to study host-pathogen interactions and builds upon previous work to create invertebrate model host systems (1, 15, 21-25, 37). The theory that a common, fundamental set of molecular mechanisms is employed by pathogens against a widely divergent array of metazoan hosts has been supported by the discovery of substantial commonality between virulence factors required for disease in mice and in killing of C. elegans (10, 21-24).
A C. elegans screen using the progeny-permissive phenotype can allow identification of fungal genes relevant for both nematode and mammalian pathogenesis. A screen for the ability of microbes to affect C. elegans progeny may be a valuable and economical technique for dissecting mechanisms of host-pathogen interactions. Screening for a progeny-permissive phenotype is much simpler than the existing killing assay—an especially valuable feature for studying pathogens with slower killing kinetics, such as C. neoformans. Fewer worms per microbial lawn are required, and surveillance need only occur once around days 3 and 5. Also, screening for the progeny-permissive phenotype may also facilitate automated screens in C. elegans. Current automated systems of C. elegans screening have limitations in distinguishing living worms from dead, a function that must be performed with very high accuracy and consistency for a killing assay screen to be effective. In addition, some microbial pathogens, such as Enterococcus faecium, fail to kill C. elegans but still inhibit progeny production (10, 22). While the C. elegans killing assay cannot be used to study such organisms, a screen for production of nematode progeny in the presence of pathogen may prove useful.
The progeny screen does have obvious limitations. The first is the unavoidable difference between nematodes and mammals. The use of invertebrate model hosts for modeling microbial pathogenesis is based on the hypothesis that there exists, for some pathogens, a set of fundamental virulence traits whose involvement in pathogenicity is broadly conserved across host phyla. However, it is therefore likely that some C. neoformans factors needed for pathogenesis in humans may not be employed against C. elegans and might consequently be overlooked by screens in C. elegans. Conversely, some C. neoformans traits may be involved in the interaction with C. elegans but not involved in human infection; these traits might come through a screen in the C. elegans pathosystem, such as the progeny screen, but would not be clinically relevant. A second problem is the fact that attenuated-virulence and progeny-permissive phenotypes, despite the aforementioned logical connection, are empirically distinct from one another. For example, PKA1 and melanin production that were involved in killing of the nematode (25) did not affect progeny production. Additionally, the quantification of progeny for the experiment requires some experience. For example, the opaque quality of cryptococcal lawns makes it difficult to quantify brood size.
As noted above, the attenuated killing and the decrease in brood size do not always overlap. On the basis of the finding that among the three mutants that allow significantly higher progeny production (cap59, rom2, and ras1), two (cap59 and rom2) are unable to survive ingestion by the nematode (Fig. 3 and reference 25), it appears that a screen for progeny may provide a unique way to identify genes associated with cryptococcal capsule and cell wall formation and structure. However, further study is necessary in order to determine whether the decrease in brood size is a result only of energy deprivation because the nematodes are unable to grind the yeast cells or whether it is part of a more extensive host-pathogen interaction.
Our findings suggest that in C. neoformans ROM2 plays a significant role in growth at high temperature, in mating, and in maintaining the stability of the cell wall in the presence of cell wall and osmolar stressors. The growth defect cannot account for the phenotype of the rom2 mutant in the C. elegans system, as nematode experiments are conducted at 25°C. Also, the defect in mating also can not account for the hypovirulence of rom2 in the C. elegans pathosystem, as mating is not associated with killing of the nematode in C. neoformans var. grubii (29). It is more likely that the sensitivity of the C. neoformans rom2 mutant to osmotic and other stresses may explain the inability of the strain KN99 rom2 to survive ingestion by C. elegans that involves passage through the grinder organ that disrupts the cells with tooth-like projections (Fig. 3). The temperature-sensitive phenotype is probably more important in the loss of virulence in mammals; however, it is likely that the cell wall defect is also important for the loss of virulence in mammals.
The phenotype of the C. neoformans rom2 mutant reported in this work has interesting similarities to that of Rom2p mutants of S. cerevisiae (detailed in references 6, 17, and 32). In S. cerevisiae, Rom2p is the Rho1 GTP/GDP exchange factor. The Rho1-protein kinase C pathway is known as the cellular integrity pathway for its role in cell wall biosynthesis and mating (32). Pkc1p protein directly activates a mitogen-activated protein kinase (MAP kinase) cascade consisting of the MAP kinase kinase kinase Bck1p, the redundant MAP kinase kinase pair, Mkk1p and Mkk2p, and the MAP kinase Slt2p/Mpk1p. As noted above, the pathway is activated by the GDP/GTP exchange factor Rom2p, via Rho1p (3, 5, 12, 20). This pathway is involved in the maintenance of cell integrity, being responsible for expression of several cell wall genes (FKS1, MNN1, CSD1) and activated by cell wall-perturbing compounds (11, 13). The pathway is also required for growth at elevated temperatures (16-18) and is involved in cell wall rearrangements in response to hyperosmotic conditions (41).
In both S. cerevisiae and C. neoformans ROM2 is associated with growth at higher temperatures, resistance to the cell wall stressor caffeine, and resistance to osmotic stresses caused by NaCl. Additionally, the growth phenotype caused by rom2 mutation in both S. cerevisiae and C. neoformans is corrected by the addition of the osmotic stabilizer sorbitol. In addition to ROM2 reported here, homologs to S. cerevisiae genes of the Rho1-protein kinase C pathway have been identified in C. neoformans. These genes also demonstrated a role in cell wall maintenance and stress response. Such genes include FKS1, which encodes a subunit of glucan synthase (39), and C. neoformans MPK1, which is needed for growing at 37°C, virulence in a murine model, and resistance to inhibitors of enzymes required for cell wall biosynthesis (14).
In addition to its clinical significance (34), C. neoformans is an ideal model organism for other pathogenic fungi, being easily cultured in the laboratory and benefiting from abundant genetic and genomic resources (19). C. elegans is a powerful tool for investigating conserved mechanisms of C. neoformans virulence against metazoans. The fact that C. elegans reduces brood size as a response to interactions with C. neoformans may serve as a useful marker in dissecting these mechanisms. This approach may allow development of better-defined screens of much higher throughput than are currently feasible.
ADDENDUM IN PROOF
The recent observation (K. J. Gerik, M. J. Donlin, C. E. Soto, A. M. Banks, I. R. Banks, M. A. Maligie, C. P. Selitrennikoff, and J. K. Lodge, Mol. Microbiol. 58:393-408, 2005) that a different rom2 mutant shows no temperature or caffeine sensitivity reflects differences in the assay used in their study and the one used in our study, including solid versus liquid medium and the use of 1 mM caffeine versus 7 mM caffeine. We have done a side-by-side comparison of these two independent and distinct mutants in the same conditions and found that they have similar phenotypes, including wild-type growth at 37°C on solid medium, impaired growth and 37°C in liquid medium, sensitivity to 1.5 M NaCl, and importantly, the production of increased Caenorhabditis progeny.
ACKNOWLEDGMENTS
Financial support was provided by a K08 award (AI63084-01) from National Institutes of Health and a New Scholar Award in Global Infectious Diseases of the Ellison Medical Foundation to E.M.
REFERENCES
1. Aballay, A., and F. M. Ausubel. 2002. Caenorhabditis elegans as a host for the study of host-pathogen interactions. Curr. Opin. Microbiol. 5:97-101.
2. Alspaugh, J. A., L. M. Cavallo, J. R. Perfect, and J. Heitman. 2000. RAS1 regulates filamentation, mating and growth at high temperature of Cryptococcus neoformans. Mol. Microbiol. 36:352-365.
3. Angeles de la Torre-Ruiz, M., J. Torres, J. Arino, and E. Herrero. 2002. Sit4 is required for proper modulation of the biological functions mediated by Pkc1 and the cell integrity pathway in Saccharomyces cerevisiae. J. Biol. Chem. 277:33468-33476.
4. Apidianakis, Y., L. G. Rahme, J. Heitman, F. M. Ausubel, S. B. Calderwood, and E. Mylonakis. 2004. Challenge of Drosophila melanogaster with Cryptococcus neoformans and role of the innate immune response. Eukaryot. Cell 3:413-419.
5. Bickle, M., P. A. Delley, A. Schmidt, and M. N. Hall. 1998. Cell wall integrity modulates RHO1 activity via the exchange factor ROM2. EMBO J. 17:2235-2245.
6. Chai, B., J. M. Hsu, J. Du, and B. C. Laurent. 2002. Yeast RSC function is required for organization of the cellular cytoskeleton via an alternative PKC1 pathway. Genetics 161:575-584.
7. Couillault, C., N. Pujol, J. Reboul, L. Sabatier, J.-F. Guichou, Y. Kohara, and J. J. Ewbank. 2004. TLR-independent control of innate immunity in Caenorhabditis elegans by the TIR domain adaptor protein TIR-1, an ortholog of human SARM. Nat. Immunol. 5:488-494.
8. D'Souza, C. A., J. A. Alspaugh, C. Yue, T. Harashima, G. M. Cox, J. R. Perfect, and J. Heitman. 2001. Cyclic AMP-dependent protein kinase controls virulence of the fungal pathogen Cryptococcus neoformans. Mol. Cell. Biol. 21:3179-3191.
9. Edman, J. C., and K. J. Kwon-Chung. 1990. Isolation of the URA5 gene from Cryptococcus neoformans var. neoformans and its use as a selective marker for transformation. Mol. Cell. Biol. 10:4538-4544.
10. Garsin, D. A., C. D. Sifri, E. Mylonakis, X. Qin, K. V. Singh, B. E. Murray, S. B. Calderwood, and F. M. Ausubel. 2001. A simple model host for identifying Gram-positive virulence factors. Proc. Natl. Acad. Sci. USA. 98:10892-10897.
11. Gustin, M. C., J. Albertyn, M. Alexander, and K. Davenport. 1998. MAP kinase pathways in the yeast Saccharomyces cerevisiae. Microbiol. Mol. Biol. Rev. 62:1264-1300.
12. Heinisch, J. J., A. Lorberg, H. P. Schmitz, and J. J. Jacoby. 1999. The protein kinase C-mediated MAP kinase pathway involved in the maintenance of cellular integrity in Saccharomyces cerevisiae. Mol. Microbiol. 32:671-680.
13. Klis, F. M., P. Mol, K. Hellingwerf, and S. Brul. 2002. Dynamics of cell wall structure in Saccharomyces cerevisiae. FEMS Microbiol. Rev. 26:239-256.
14. Kraus, P. R., D. S. Fox, G. M. Cox, and J. Heitman. 2003. The Cryptococcus neoformans MAP kinase Mpk1 regulates cell integrity in response to antifungal drugs and loss of calcineurin function. Mol. Microbiol. 48:1377-1387.
15. Kurz, C. L., S. Chauvet, E. Andres, M. Aurouze, I. Vallet, G. P. Michel, M. Uh, J. Celli, A. Filloux, S. De Bentzmann, I. Steinmetz, J. A. Hoffmann, B. B. Finlay, J. P. Gorvel, D. Ferrandon, and J. J. Ewbank. 2003. Virulence factors of the human opportunistic pathogen Serratia marcescens identified by in vivo screening. EMBO J. 22:1451-1460.
16. Lee, K. S., and D. E. Levin. 1992. Dominant mutations in a gene encoding a putative protein kinase (BCK1) bypass the requirement for a Saccharomyces cerevisiae protein kinase C homolog. Mol. Cell. Biol. 12:172-182.
17. Levin, D. E. 2005. Cell wall integrity signaling in Saccharomyces cerevisiae. Microbiol. Mol. Biol. Rev. 69:262-291.
18. Levin, D. E., F. O. Fields, Kunisawa, J. M. Bishop, and J. Thorner. 1990. A candidate protein kinase C gene, PKC1, is required for the S. cerevisiae cell cycle. Cell 62:213-224.
19. Loftus, B. J., E. Fung, P. Roncaglia, D. Rowley, P. Amedeo, D. Bruno, J. Vamathevan, M. Miranda, I. J. Anderson, J. A. Fraser, J. E. Allen, I. E. Bosdet, M. R. Brent, R. Chiu, T. L Doering, M. J. Donlin, C. A. D'Souza, D. S. Fox, V. Grinberg, J. Fu, M. Fukushima, B. J. Haas, J. C. Huang, G. Janbon, S. J. Jones, H. L. Koo, M. I. Krzywinski, J. K. Kwon-Chung, K. B. Lengeler, R. Maiti, M. A. Marra, R. E. Marra, C. A. Mathewson, T. G. Mitchell, M. Pertea, F. R. Riggs, S. L. Salzberg, J. E. Schein, A. Shvartsbeyn, H. Shin, M. Shumway, C. A. Specht, B. B. Suh, A. Tenney, T. R. Utterback, B. L. Wickes, J. R. Wortman, N. H. Wye, J. W. Kronstad, J. K. Lodge, J. Heitman, R. W. Davis, C. M. Fraser, and R. W. Hyman. 2005. The genome of the basidiomycetous yeast and human pathogen Cryptococcus neoformans. Science 307:1321-1324.
20. Lorberg, A., H. P. Schmitz, U. Gengenbacher, and J. J. Heinisch. 2003. KlROM2 encodes an essential GEF homologue in Kluyveromyces lactis. Yeast 20:611-624.
21. Mahajan-Miklos, S., M. W. Tan, L. G. Rahme, and F. M. Ausubel. 1999. Molecular mechanisms of bacterial virulence elucidated using a Pseudomonas aeruginosa-Caenorhabditis elegans pathogenesis model. Cell 96:47-56.
22. Moy, T., E. Mylonakis, S. B. Calderwood, and F. M. Ausubel. 2004. Cytotoxicity from hydrogen peroxide produced by Enterococcus faecium. Infect. Immun. 72:4512-4520.
23. Mylonakis, E., and A. Aballay. 2005. Worms and flies as genetically tractable animal models to study host-pathogen interactions. Infect. Immun. 73:3833-3841.
24. Mylonakis, E., A. Idnurm, R. Moreno, J. El Khoury, J. B. Rottman, F. M. Ausubel, J. Heitman, and S. B. Calderwood. 2004. Cryptococcus neoformans Kin1 protein kinase homologue, identified through a Caenorhabditis elegans screen, promotes virulence in mammals. Mol. Microbiol. 54:407-419.
25. Mylonakis, E., F. M. Ausubel, J. R. Perfect, J. Heitman, and S. B. Calderwood. 2002. Killing of Caenorhabditis elegans by Cryptococcus neoformans as a model of yeast pathogenesis. Proc. Natl. Acad. Sci. USA 99:15675-15680.
26. Mylonakis, E., R. Moreno, J. El Khoury, A. Idnurm, J. Heitman, S. B. Calderwood, F. M. Ausubel, and A. Diener. 2005. Galleria mellonella as a model system to study Cryptococcus neoformans pathogenesis. Infect. Immun. 73:3842-3850.
27. Nelson, R. T., B. A. Pryor, and J. K. Lodge. 2003. Sequence length required for homologous recombination in Cryptococcus neoformans. Fungal Genet. Biol. 38:1-9.
28. Nelson, R. T., J. Hua, B. Pryor, and J. K. Lodge. 2001. Identification of virulence mutants of the fungal pathogen Cryptococcus neoformans using signature-tagged mutagenesis. Genetics 157:935-947.
29. Nielsen, K., G. M. Cox, A. P. Litvintseva, E. Mylonakis, S. D. Malliaris, D. K. Benjamin, Jr., S. S. Giles, T. G. Mitchell, A. Casadevall, J. R. Perfect, and J. Heitman. 2005. Cryptococcus neoformans strains preferentially disseminate to the central nervous system during infection. Infect. Immun. 73:4922-4933.
30. Nielsen, K., G. M. Cox, P. Wang, D. L. Toffaletti, J. R. Perfect, and J. Heitman. 2003. Sexual cycle of Cryptococcus neoformans var. grubii and virulence of congenic a and isolates. Infect. Immun. 71:4831-4841.
31. Ozaki, K., K. Tanaka, H. Imamura, T. Hihara, T. Kameyama, H. Nonaka, H. Hirano, Y. Matsuura, and Y. Takai. 1996. Rom1p and Rom2p are GDP/GTP exchange proteins (GEPs) for the Rho1p small GTP binding protein in Saccharomyces cerevisiae. EMBO J. 15:2196-2207.
32. Park, J. I., E. J. Collinson, C. M. Grant, and I. W. Dawes. 2005. Rom2p, the Rho1 GTP/GDP exchange factor of Saccharomyces cerevisiae, can mediate stress responses via the Ras-cAMP pathway. J. Biol. Chem. 280:2529-2535.
33. Perfect, J. R., S. D. R. Lang, and D. T. Durack. 1980. Chronic cryptococcal meningitis: a new experimental model in rabbits. Am. J. Pathol. 101:177-194.
34. Perfect, J. R., and A. Casadevall. 2002. Cryptococcosis. Infect. Dis. Clin. N Am. 16:837-874.
35. Pradel, E., and J. J. Ewbank. 2004. Genetic models in pathogenesis. Annu. Rev. Genet. 38:347-363.
36. Salas, S. D., J. E. Bennett, K. J. Kwon-Chung, J. R. Perfect, and P. R. Williamson. 1996. Effect of the laccase gene CNLAC1, on virulence of Cryptococcus neoformans. J. Exp. Med. 184:377-386.
37. Steenbergen, J. N., H. A. Shuman, and A. Casadevall. 2001. Cryptococcus neoformans interactions with amoebae suggest an explanation for its virulence and intracellular pathogenic strategy in macrophages. Proc. Natl. Acad. Sci. USA 98:15245-15250.
38. Tan, M. W., S. Mahajan-Miklos, and F. M. Ausubel. 1999. Killing of Caenorhabditis elegans by Pseudomonas aeruginosa used to model mammalian bacterial pathogenesis. Proc. Natl. Acad. Sci. USA 96:715-720.
39. Thompson, J. R., C. M. Douglas, W. Li, C. K. Jue, B. Pramanik, X. Yuan, T. H. Rude, D. L. Toffaletti, J. R. Perfect, and M. Kurtz. 1999. A glucan synthase FKS1 homolog in Cryptococcus neoformans is single copy and encodes an essential function. J. Bacteriol. 181:444-453.
40. Wang, P., M. E. Cardenas, G. M. Cox, J. R. Perfect, and J. Heitman. 2001. Two cyclophilin A homologs with shared and distinct functions important for growth and virulence of Cryptococcus neoformans. EMBO Rep. 2:511-518.
41. Wojda, I., R. Alonso-Monge, J. P. Bebelman, W. H. Mager, and M. Siderius. 2003. Response to high osmotic conditions and elevated temperature in Saccharomyces cerevisiae is controlled by intracellular glycerol and involves coordinate activity of MAP kinase pathways. Microbiology 149:1193-1204.(Robin J. Tang, Julia Breg)
Department of Molecular Genetics and Microbiology Division of Infectious Diseases
Department of Medicine Howard Hughes Medical Institute, Duke University Medical Center, Durham, North Carolina 27710
Edward A. Doisy Department of Biochemistry and Molecular Biology
Department of Molecular Microbiology and Immunology, Saint Louis University School of Medicine, Saint Louis, Missouri 63104
Department of Microbiology and Molecular Genetics, Harvard Medical School, Boston, Massachusetts 02114
ABSTRACT
Caenorhabditis elegans can serve as a substitute host for the study of microbial pathogenesis. We found that mutations in genes of the fungal pathogen Cryptococcus neoformans involved in mammalian virulence allow C. elegans to produce greater numbers of progeny than when exposed to wild-type fungus. We used this property to screen a library of C. neoformans mutants for strains that permit larger C. elegans brood sizes. In this screen, we identified a gene homologous to Saccharomyces cerevisiae ROM2. C. neoformans rom2 mutation resulted in a defect in mating and growth defects at elevated temperature or in the presence of cell wall or hyperosmolar stresses. An effect of the C. neoformans rom2 mutation in virulence was confirmed in a murine inhalation infection model. We propose that a screen for progeny-permissive mutants of microorganisms can serve as a high-throughput method for identifying novel loci related to mammalian pathogenesis.
INTRODUCTION
The nematode Caenorhabditis elegans is emerging as a substitute host for elucidating important mechanisms of fungal and bacterial pathogenesis, thanks to genetic tractability, anatomical simplicity, and the availability of extensive genetic and genomic resources (23, 35). The C. elegans killing assay used in previous studies entails observing a synchronized population of nematodes for kinetics of death on a lawn of the microbe being studied. Living and dead worms are counted at 24-h intervals, and the LT50, or time for 50% of the nematodes to die, is calculated (1, 7, 10, 15, 24, 38). This killing assay has been used to study effectively gram-negative (1, 15, 38) and gram-positive (10) bacteria, as well as the human pathogenic fungus Cryptococcus neoformans (24, 25).
Interestingly, C. elegans does not produce a sustainable brood of progeny in the presence of C. neoformans or some other pathogens (10, 25, 38). We hypothesized that a screen of brood size would circumvent the need for labor-intensive manual assessment for living versus dead nematodes. The aim of this study was to investigate the role of the C. neoformans virulence factors (such as capsule and melanin and their regulatory pathways) in C. elegans progeny production. Also, we evaluate the hypothesis that brood size can be used effectively as a facile marker for virulence in the C. elegans-C. neoformans system. Of note is that no progeny-based screen has been reported previously.
MATERIALS AND METHODS
Strains and media. The C. neoformans strains used in these experiments are summarized in Table 1 or described in the text. Yeast cultures were maintained as frozen stocks or on yeast-peptone-dextrose (YPD; Difco) agar plates and grown in YPD. C. neoformans cultures were grown at 30°C unless otherwise specified. Cryptococcus laurentii cultures were grown at 25°C. The standard C. elegans strain N2 Bristol was maintained at 15°C and propagated on Escherichia coli.
C. elegans killing assays were performed as previously described (24, 25) with minor modifications. C. neoformans strains were inoculated into 2 ml of YPD and grown at 30°C for 24 h; 10 μl of the culture was spread on 35-mm tissue-culture plates (Falcon) containing brain heart infusion (BHI) agar (Difco). The plates were incubated at 30°C for 48 h and then at 25°C for an additional 24 h. Of note is that by incubating the plates for an extra day at 30°C, killing of C. elegans was faster than what we reported in the past (24, 25). Ampicillin was added to the medium to prevent growth of E. coli OP50 that might have been carried over on transfer of worms to the yeast-containing plates. C. elegans animals at the L4 developmental stage were transferred from a lawn of E. coli OP50 on nematode growth medium to a lawn of the yeast to be tested on BHI medium, incubated at 25°C, and examined for viability at 24-h intervals with a Nikon SMZ645 dissecting microscope. Three plates with 50 L4 larval C. elegans nematodes per plate were used for each strain. Every 24 h, living and dead worms were counted, with living worms being transferred to a fresh, identical lawn of C. neoformans to allow observation of the original worm without obstruction by progeny.
Quantification of C. elegans progeny. Lawns for the quantification of C. elegans progeny assays were prepared as detailed above for the killing assays. The only difference is that C. laurentii plates were incubated at 25°C throughout the experiment (i.e., were not placed at 30°C as was initially done with C. neoformans plates). The reason for the different temperature is that 25°C is the optimal temperature for C. laurentii growth (25). However, special care was taken to ensure that the C. neoformans and C. laurentii lawns were similar in size and thickness.
Nematodes were placed on lawns of C. neoformans or C. laurentii on BHI agar at 25°C and transferred to a new, identical lawn every 24 h throughout the reproductive period. After removal of the parent nematode, each lawn was spread face down on a 100-mm tissue-culture plate (Falcon) with nematode growth medium agar and E. coli strain OP50. Care was taken not to intermingle OP50 bacteria with C. neoformans or C. laurentii. Numbers reported are the total progeny produced over the first 72 h of egg laying. No unhatched eggs were seen. Approximately 20 to 25 wild-type C. elegans N2 nematodes were used per group, and only nematodes that survived all 3 days of the experiment were included. Viable progeny was defined as a first-stage (or older) larva. Also, to compare progeny production between strain ATCC 208820 (2e-tuc4) that produces laccase and strain ATCC 208819 (2e-tu4) that is laccase negative we grew these strains in the presence of L-DOPA (3,4-dihydroxyphenylalanine) (100 μg/ml) and added L-DOPA (100 μg/ml) to the BHI medium (25).
Identification and construction of mutant strains. Random mutants of C. neoformans were generated using biolistic insertion of a plasmid containing the URA5 gene into the strain F99 (MAT ura5) (9, 40). In mutants permitting significant brood production, DNA flanking the insertion was identified using arbitrary primed PCR and compared to the H99 genome database at Duke University. Gene prediction was performed with the FGENESH algorithm, and predicted genes were compared to the GenBank database. A disruption allele for the ROM2 gene was constructed by overlap PCR to replace most of the coding region with a cassette conferring resistance to nourseothricin. Primers (all 5'-3') JOHE13032 (TCCTCTATCATCCCATCTAG) and JOHE13033 (GCTTATGTGAGTCCTCCCGTTGTTGGATTGCTGCTG) and primers JOHE13034 (CTCGTTTCTACATCTCTTGTCATGACGCACCACCAG) and JOHE13035 (GATGACGAATTTGCTTTACTAG) were used to amplify the flanks of the ROM2 gene from KN99 DNA, and primers JOHE8677 (GAAGAGATGTAGAAACGAG) and JOHE11866 (GGGAGGACTCACATAAGC) were used to amplify the NAT cassette. Equimolar amounts of the three products were mixed, and primers JOHE13032 and JOHE13035 were used to amplify a contiguous fragment; this fragment was transformed into strain KN99 (MAT) by use of a biolistic apparatus. Disruption of ROM2 was confirmed with PCR and Southern blot hybridization. Only strains containing a single insertion were used. The rom2 mutation was crossed to strain KN99a, and basidiospore progeny were selected to identify a rom2 MATa strain. To reconstitute ROM2, a wild-type copy of ROM2 was amplified with primers JOHE13031 (ACCTATCATCTCGCTGATCC) and JOHE13036 (GAGAAGGATACTGAAAAGGC) and cloned into a plasmid conferring resistance to neomycin, which was introduced into the rom2 mutant with Agrobacterium-mediated transformation followed by selection on neomycin-containing media (MediaTech, Inc., Herndon, VA).
In vitro evaluation of cryptococcal strains. Capsule production was assessed by exclusion of India ink following growth for 48 h at 30°C in low-iron media containing 56 μM EDDHA. Measurement of the capsule size (diameter of the edge of the capsule to the cell wall and cell diameter) was evaluated on an Olympus BX51 microscope using Olympus Microsoft software. Melanin production was assessed by growth on L-DOPA-containing minimal medium agar for 7 days at 30°C. For the mating assays, strains were each crossed by mixing strains and incubating on V8 media at 25°C in the dark for 14 days. For the study of the growth on cell wall and osmolar and other stressors (such as caffeine and NaCl), incubation times differed because growth rates differed under different conditions, but incubation was continued until a growth defect appeared or was plainly not forthcoming. All in vitro assays were performed at 25°C, 30°C, and 37°C. In all studies, results at 25°C and 30°C were similar unless otherwise specified.
Murine models of virulence. Experiments with mice were performed as previously described (24). Wild-type 8-week-old female ICR/CD1 mice (Charles River Laboratories, Wilmington, MA) were used in all studies. In brief, C. neoformans strains were grown at 30°C with shaking overnight in YPD to late-log phase. The yeast cells were centrifuged and washed in phosphate-buffered saline. The concentration of yeast cells in the inoculum was evaluated before inoculation by use of a hemocytometer and confirmed by plating serial dilutions and enumerating CFU. Mice were anesthetized by intraperitoneal injection of tribromoethyl alcohol (Aldrich, Milwakee, WI) (400 mg/kg of body weight) and suspended by the incisors on a silk thread. A 50-μl volume of the inoculum (1.0 x 106 yeast cells of either the rom2 mutant the parental strain KN99 or the reconstituted strain KN99 rom2 + ROM2) was slowly pipetted into the nares with continued suspension for 10 min. The murine protocols were approved by the Massachusetts General Hospital Committee on Research, Subcommittee on Research Animal Care, and special attention was given to minimize suffering of the mice.
Evaluation of tissue burden. Following euthanasia, we harvested the lungs from mice 12 days after administration of a cryptococcal inoculum by nasal inhalation. Tissues were weighed and homogenized in sterile phosphate-buffered saline by use of a Tissue Tearor (model 398; Biospec Products Inc., Racine, WI). Then, serial dilutions were plated on YPD agar containing 100 μg/ml ampicillin, 100 μg/ml streptomycin, and 45 μg/ml kanamycin. CFU of C. neoformans were counted after growth at 30°C for 72 h.
Statistical analysis. Survival assays in mice and nematodes were repeated three times, and each experiment independently gave statistically significant results. The figures and the P values provided herein are each from one representative experiment. Murine and C. elegans killing curves were plotted and estimation of differences in survival (log-rank and Wilcoxon tests) analyzed with the Kaplan-Meier method performed using STATA 6 statistical software (Stata, College Station, TX). The same software program was used for the statistical analysis of the CFU from murine organs and comparison of progeny production (Mann-Whitney and Kruskal-Wallis tests). Plotting and statistical evaluation of CFU data were done using Microcal Origin (OriginLab Corporation, Northampton, MA). A P value of less than 0.05 was considered to be significant.
RESULTS
Abrogation of brood production is associated with C. neoformans genes involved in pathogenesis. First we quantified C. elegans progeny produced over 3 days in the presence of wild-type C. neoformans and a ras1 mutant of C. neoformans that was previously found to be attenuated in C. elegans (25) and mammalian (2) killing. The nonpathogenic C. laurentii strain ATCC 76483 was used as a control. Both C. laurentii and the hypovirulent ras1 mutant of C. neoformans allowed C. elegans to produce greater numbers of viable progeny than did the wild-type C. neoformans KN99 or the reconstituted strain H99 ras1 + RAS1 (Table 1; P = 0.0001 and 0.0456, respectively). Levels of progeny production by the strain H99 and the isogenic strain KN99 were similar. In addition to the RAS1 signaling cascade, a second regulatory pathway that involves the G-alpha protein-cyclic AMP-protein kinase A signaling pathway has been described for C. neoformans. Interestingly, a C. neoformans pka1 mutant that displays attenuated virulence in mice (8) and C. elegans killing (25) did not allow progeny production that was statistically different from that by nematodes exposed to wild-type C. neoformans (Table 1).
In order to dissect the role of the major cryptococcal virulence factors (capsule and melanin production) in progeny production, we studied C. elegans brood production in the presence of an acapsular strain with a mutation involving gene CAP59 that is needed for capsule formation and a strain with a mutation involving the enzyme laccase that is essential for melanin production. Progeny production in the presence of the acapsular strain was significantly higher than the progeny production on lawns of wild-type C. neoformans (average ± standard error of 1.68 ± 0.4 for the KN99 group and 5.9 ± 2.0 for the H99 cap59 group; P = 0.0044). However, nematodes exposed to the acapsular strain cap59 made significantly less progeny compared to the nematodes exposed to ras1 (5.9 ± 2.0 for H99 cap59 compared to 17.06 ± 2.28 for KN99 rom2 and 13.5 ± 3.04 for H99 ras1; in both cases, P < 0.01). There was no difference in progeny production between the strain ATCC 208820 (2e-tuc4) that produces laccase and strain ATCC 208819 (2e-tu4) that is laccase negative (data not shown).
A C. elegans progeny-based screen. Based on the finding that some virulence factors of C. neoformans can cause abrogation of brood production in C. elegans, we investigated whether screening for brood production can be used to identify novel factors involved in C. neoformans pathogenesis in mammals. We screened a library of 1,500 random C. neoformans mutants for strains that permitted a C. elegans brood to develop on lawns of these mutants, a phenotype we termed "progeny permissive." We isolated three such mutants, a rate similar to that seen with existing screens utilizing killing of C. elegans by pathogens (10, 21, 24). Genetic analysis of one mutant strain revealed that the disrupted locus encoded a homolog of Saccharomyces cerevisiae Rom2p. S. cerevisiae Rom2p is a guanyl-nucleotide exchange factor for Rho1p and is required for yeast cell integrity under conditions of heat and osmolar stress (5, 31, 32).
Evaluation of a C. neoformans rom2 mutant in vitro. A mutant lacking all but the last 18 codons of the ROM2 locus grew similarly to the wild-type yeast at 30°C, but it demonstrated reduced growth at 37°C (Fig. 1A). All three strains grew similarly at 25°C (data not shown). Of note is that at 37°C, the rom2 mutant demonstrated a temperature-sensitive phenotype, and a delay in growth. Interestingly, the high-temperature growth defect was corrected by the addition of 1 M sorbitol to the media. Of note is that in spite of the delay in growth in YPD media, it finally reaches a concentration similar to that of the wild type even at 37°C (Fig. 1A) and that there was no increased lysis associated with the rom2 mutant, as noted using the BCIP (5-bromo-4-chloro-3-indolyl phosphate; Sigma) colorimetric assay (data not shown).
Melanin and capsule production were not affected by the rom2 deletion (data not shown), while mating was significantly impaired (Fig. 1B). Basidiospore production and hyphal projection were markedly diminished when MAT and MATa rom2 mutants were crossed, but the phenotype of the reconstituted strain KN99 rom2 + ROM2 was similar to that of the wild type. Of note is that the mating defect was less pronounced when we mated the rom2 mutant with the KN99 partner but did not revert to wild-type mating even with the addition of osmotic stabilizer to the media (data not shown).
The finding that a C. neoformans rom2 deletion mutant exhibits a defect in mating and a growth defect that can be suppressed by addition of the osmotic stabilizer sorbitol suggests that in C. neoformans ROM2 may play a role in the formation or structure of the cell wall, similar to findings with S. cerevisiae (32). In order to evaluate this hypothesis we examined the sensitivity of the C. neoformans rom2 mutant to cell wall and osmolar stressors.
First we examined the sensitivity of KN99 rom2 to caffeine, a phosphodiesterase inhibitor that has been used to screen for yeast mutants with altered cell wall proteins and/or structure (6). In S. cerevisiae, deletion of ROM2 renders cells sensitive to caffeine (32). Cells of KN99, KN99 rom2, and KN99 rom2 + ROM2 were grown on solid media containing caffeine. Plates were incubated at 25°C, 30°C, or 37°C. KN99 rom2 demonstrated sensitivity to 7 mM caffeine that was apparent at all temperatures tested (Fig. 2 and data not shown). Of note, at 37°C, 7 mM caffeine entirely inhibited growth of the mutant. The reconstituted strain, KN99 rom2 + ROM2, grew similarly to the wild type under all conditions, and no significant difference between the rom2 mutant, the wild type, and the reconstituted strain was seen on YPD containing 3 mM caffeine.
KN99 rom2 was also hypersensitive to 1.5 M NaCl at 25°C and 30°C, giving rise to very sparse colonies of tiny size (Fig. 2). Growth of the strain KN99 rom2 + ROM2 in 1.5 M NaCl was similar to the wild-type growth. Of note, at 37°C all three strains (the wild type, rom2 mutant, and reconstituted strain) failed to grow in the presence of 1.5 M NaCl; as a consequence, it was not possible to determine the relative sensitivity of KN99 rom2 to 1.5 M NaCl at this temperature.
A C. neoformans rom2 mutant is hypovirulent in C. elegans and mice. Strain KN99 rom2 permitted significantly higher numbers of C. elegans progeny to develop than did the wild-type yeast. More specifically, the KN99 rom2 mutant allowed significantly more progeny than KN99 or KN99 rom2 + ROM2 (P < 0.0001 and P = 0.0004, respectively; Table 1). Moreover, the KN99 rom2 mutant was significantly attenuated in virulence in C. elegans and demonstrated an impaired ability to survive passage through the nematode grinder organ (Fig. 3).
Interestingly, the rom2 mutant was also avirulent in an inhalation infection model in mice (Fig. 4) and was considerably impaired in establishing infection in murine lung tissue (Fig. 5). Of note, experiments were performed in triplicate and a total of 32 mice received KN99 rom2. No death or illness was observed for any of the mice that received KN99 rom2, including a total of 22 mice that were kept for 120 days postinoculation.
DISCUSSION
In this report we present the novel observation that some, but not all, mutations in C. neoformans genes involved in mammalian virulence allow C. elegans to produce greater numbers of progeny than when exposed to wild-type fungus. This progeny-permissive phenotype can be used to screen collections of C. neoformans mutants to identify genes involved in mammalian infection. The first gene identified using this approach is a gene homologous to Saccharomyces cerevisiae ROM2. In C. neoformans a rom2 mutant demonstrated a defect in mating and in growth at 37°C, as well as susceptibility to cell wall and osmolar stressors. Interestingly, ROM2 was essential for full virulence in an inhalation murine model of infection, suggesting that this screen is useful for elucidating pathogenesis in mammalian hosts.
This study extends previous work by describing a new technique, the progeny screen, for using the C. elegans to study host-pathogen interactions and builds upon previous work to create invertebrate model host systems (1, 15, 21-25, 37). The theory that a common, fundamental set of molecular mechanisms is employed by pathogens against a widely divergent array of metazoan hosts has been supported by the discovery of substantial commonality between virulence factors required for disease in mice and in killing of C. elegans (10, 21-24).
A C. elegans screen using the progeny-permissive phenotype can allow identification of fungal genes relevant for both nematode and mammalian pathogenesis. A screen for the ability of microbes to affect C. elegans progeny may be a valuable and economical technique for dissecting mechanisms of host-pathogen interactions. Screening for a progeny-permissive phenotype is much simpler than the existing killing assay—an especially valuable feature for studying pathogens with slower killing kinetics, such as C. neoformans. Fewer worms per microbial lawn are required, and surveillance need only occur once around days 3 and 5. Also, screening for the progeny-permissive phenotype may also facilitate automated screens in C. elegans. Current automated systems of C. elegans screening have limitations in distinguishing living worms from dead, a function that must be performed with very high accuracy and consistency for a killing assay screen to be effective. In addition, some microbial pathogens, such as Enterococcus faecium, fail to kill C. elegans but still inhibit progeny production (10, 22). While the C. elegans killing assay cannot be used to study such organisms, a screen for production of nematode progeny in the presence of pathogen may prove useful.
The progeny screen does have obvious limitations. The first is the unavoidable difference between nematodes and mammals. The use of invertebrate model hosts for modeling microbial pathogenesis is based on the hypothesis that there exists, for some pathogens, a set of fundamental virulence traits whose involvement in pathogenicity is broadly conserved across host phyla. However, it is therefore likely that some C. neoformans factors needed for pathogenesis in humans may not be employed against C. elegans and might consequently be overlooked by screens in C. elegans. Conversely, some C. neoformans traits may be involved in the interaction with C. elegans but not involved in human infection; these traits might come through a screen in the C. elegans pathosystem, such as the progeny screen, but would not be clinically relevant. A second problem is the fact that attenuated-virulence and progeny-permissive phenotypes, despite the aforementioned logical connection, are empirically distinct from one another. For example, PKA1 and melanin production that were involved in killing of the nematode (25) did not affect progeny production. Additionally, the quantification of progeny for the experiment requires some experience. For example, the opaque quality of cryptococcal lawns makes it difficult to quantify brood size.
As noted above, the attenuated killing and the decrease in brood size do not always overlap. On the basis of the finding that among the three mutants that allow significantly higher progeny production (cap59, rom2, and ras1), two (cap59 and rom2) are unable to survive ingestion by the nematode (Fig. 3 and reference 25), it appears that a screen for progeny may provide a unique way to identify genes associated with cryptococcal capsule and cell wall formation and structure. However, further study is necessary in order to determine whether the decrease in brood size is a result only of energy deprivation because the nematodes are unable to grind the yeast cells or whether it is part of a more extensive host-pathogen interaction.
Our findings suggest that in C. neoformans ROM2 plays a significant role in growth at high temperature, in mating, and in maintaining the stability of the cell wall in the presence of cell wall and osmolar stressors. The growth defect cannot account for the phenotype of the rom2 mutant in the C. elegans system, as nematode experiments are conducted at 25°C. Also, the defect in mating also can not account for the hypovirulence of rom2 in the C. elegans pathosystem, as mating is not associated with killing of the nematode in C. neoformans var. grubii (29). It is more likely that the sensitivity of the C. neoformans rom2 mutant to osmotic and other stresses may explain the inability of the strain KN99 rom2 to survive ingestion by C. elegans that involves passage through the grinder organ that disrupts the cells with tooth-like projections (Fig. 3). The temperature-sensitive phenotype is probably more important in the loss of virulence in mammals; however, it is likely that the cell wall defect is also important for the loss of virulence in mammals.
The phenotype of the C. neoformans rom2 mutant reported in this work has interesting similarities to that of Rom2p mutants of S. cerevisiae (detailed in references 6, 17, and 32). In S. cerevisiae, Rom2p is the Rho1 GTP/GDP exchange factor. The Rho1-protein kinase C pathway is known as the cellular integrity pathway for its role in cell wall biosynthesis and mating (32). Pkc1p protein directly activates a mitogen-activated protein kinase (MAP kinase) cascade consisting of the MAP kinase kinase kinase Bck1p, the redundant MAP kinase kinase pair, Mkk1p and Mkk2p, and the MAP kinase Slt2p/Mpk1p. As noted above, the pathway is activated by the GDP/GTP exchange factor Rom2p, via Rho1p (3, 5, 12, 20). This pathway is involved in the maintenance of cell integrity, being responsible for expression of several cell wall genes (FKS1, MNN1, CSD1) and activated by cell wall-perturbing compounds (11, 13). The pathway is also required for growth at elevated temperatures (16-18) and is involved in cell wall rearrangements in response to hyperosmotic conditions (41).
In both S. cerevisiae and C. neoformans ROM2 is associated with growth at higher temperatures, resistance to the cell wall stressor caffeine, and resistance to osmotic stresses caused by NaCl. Additionally, the growth phenotype caused by rom2 mutation in both S. cerevisiae and C. neoformans is corrected by the addition of the osmotic stabilizer sorbitol. In addition to ROM2 reported here, homologs to S. cerevisiae genes of the Rho1-protein kinase C pathway have been identified in C. neoformans. These genes also demonstrated a role in cell wall maintenance and stress response. Such genes include FKS1, which encodes a subunit of glucan synthase (39), and C. neoformans MPK1, which is needed for growing at 37°C, virulence in a murine model, and resistance to inhibitors of enzymes required for cell wall biosynthesis (14).
In addition to its clinical significance (34), C. neoformans is an ideal model organism for other pathogenic fungi, being easily cultured in the laboratory and benefiting from abundant genetic and genomic resources (19). C. elegans is a powerful tool for investigating conserved mechanisms of C. neoformans virulence against metazoans. The fact that C. elegans reduces brood size as a response to interactions with C. neoformans may serve as a useful marker in dissecting these mechanisms. This approach may allow development of better-defined screens of much higher throughput than are currently feasible.
ADDENDUM IN PROOF
The recent observation (K. J. Gerik, M. J. Donlin, C. E. Soto, A. M. Banks, I. R. Banks, M. A. Maligie, C. P. Selitrennikoff, and J. K. Lodge, Mol. Microbiol. 58:393-408, 2005) that a different rom2 mutant shows no temperature or caffeine sensitivity reflects differences in the assay used in their study and the one used in our study, including solid versus liquid medium and the use of 1 mM caffeine versus 7 mM caffeine. We have done a side-by-side comparison of these two independent and distinct mutants in the same conditions and found that they have similar phenotypes, including wild-type growth at 37°C on solid medium, impaired growth and 37°C in liquid medium, sensitivity to 1.5 M NaCl, and importantly, the production of increased Caenorhabditis progeny.
ACKNOWLEDGMENTS
Financial support was provided by a K08 award (AI63084-01) from National Institutes of Health and a New Scholar Award in Global Infectious Diseases of the Ellison Medical Foundation to E.M.
REFERENCES
1. Aballay, A., and F. M. Ausubel. 2002. Caenorhabditis elegans as a host for the study of host-pathogen interactions. Curr. Opin. Microbiol. 5:97-101.
2. Alspaugh, J. A., L. M. Cavallo, J. R. Perfect, and J. Heitman. 2000. RAS1 regulates filamentation, mating and growth at high temperature of Cryptococcus neoformans. Mol. Microbiol. 36:352-365.
3. Angeles de la Torre-Ruiz, M., J. Torres, J. Arino, and E. Herrero. 2002. Sit4 is required for proper modulation of the biological functions mediated by Pkc1 and the cell integrity pathway in Saccharomyces cerevisiae. J. Biol. Chem. 277:33468-33476.
4. Apidianakis, Y., L. G. Rahme, J. Heitman, F. M. Ausubel, S. B. Calderwood, and E. Mylonakis. 2004. Challenge of Drosophila melanogaster with Cryptococcus neoformans and role of the innate immune response. Eukaryot. Cell 3:413-419.
5. Bickle, M., P. A. Delley, A. Schmidt, and M. N. Hall. 1998. Cell wall integrity modulates RHO1 activity via the exchange factor ROM2. EMBO J. 17:2235-2245.
6. Chai, B., J. M. Hsu, J. Du, and B. C. Laurent. 2002. Yeast RSC function is required for organization of the cellular cytoskeleton via an alternative PKC1 pathway. Genetics 161:575-584.
7. Couillault, C., N. Pujol, J. Reboul, L. Sabatier, J.-F. Guichou, Y. Kohara, and J. J. Ewbank. 2004. TLR-independent control of innate immunity in Caenorhabditis elegans by the TIR domain adaptor protein TIR-1, an ortholog of human SARM. Nat. Immunol. 5:488-494.
8. D'Souza, C. A., J. A. Alspaugh, C. Yue, T. Harashima, G. M. Cox, J. R. Perfect, and J. Heitman. 2001. Cyclic AMP-dependent protein kinase controls virulence of the fungal pathogen Cryptococcus neoformans. Mol. Cell. Biol. 21:3179-3191.
9. Edman, J. C., and K. J. Kwon-Chung. 1990. Isolation of the URA5 gene from Cryptococcus neoformans var. neoformans and its use as a selective marker for transformation. Mol. Cell. Biol. 10:4538-4544.
10. Garsin, D. A., C. D. Sifri, E. Mylonakis, X. Qin, K. V. Singh, B. E. Murray, S. B. Calderwood, and F. M. Ausubel. 2001. A simple model host for identifying Gram-positive virulence factors. Proc. Natl. Acad. Sci. USA. 98:10892-10897.
11. Gustin, M. C., J. Albertyn, M. Alexander, and K. Davenport. 1998. MAP kinase pathways in the yeast Saccharomyces cerevisiae. Microbiol. Mol. Biol. Rev. 62:1264-1300.
12. Heinisch, J. J., A. Lorberg, H. P. Schmitz, and J. J. Jacoby. 1999. The protein kinase C-mediated MAP kinase pathway involved in the maintenance of cellular integrity in Saccharomyces cerevisiae. Mol. Microbiol. 32:671-680.
13. Klis, F. M., P. Mol, K. Hellingwerf, and S. Brul. 2002. Dynamics of cell wall structure in Saccharomyces cerevisiae. FEMS Microbiol. Rev. 26:239-256.
14. Kraus, P. R., D. S. Fox, G. M. Cox, and J. Heitman. 2003. The Cryptococcus neoformans MAP kinase Mpk1 regulates cell integrity in response to antifungal drugs and loss of calcineurin function. Mol. Microbiol. 48:1377-1387.
15. Kurz, C. L., S. Chauvet, E. Andres, M. Aurouze, I. Vallet, G. P. Michel, M. Uh, J. Celli, A. Filloux, S. De Bentzmann, I. Steinmetz, J. A. Hoffmann, B. B. Finlay, J. P. Gorvel, D. Ferrandon, and J. J. Ewbank. 2003. Virulence factors of the human opportunistic pathogen Serratia marcescens identified by in vivo screening. EMBO J. 22:1451-1460.
16. Lee, K. S., and D. E. Levin. 1992. Dominant mutations in a gene encoding a putative protein kinase (BCK1) bypass the requirement for a Saccharomyces cerevisiae protein kinase C homolog. Mol. Cell. Biol. 12:172-182.
17. Levin, D. E. 2005. Cell wall integrity signaling in Saccharomyces cerevisiae. Microbiol. Mol. Biol. Rev. 69:262-291.
18. Levin, D. E., F. O. Fields, Kunisawa, J. M. Bishop, and J. Thorner. 1990. A candidate protein kinase C gene, PKC1, is required for the S. cerevisiae cell cycle. Cell 62:213-224.
19. Loftus, B. J., E. Fung, P. Roncaglia, D. Rowley, P. Amedeo, D. Bruno, J. Vamathevan, M. Miranda, I. J. Anderson, J. A. Fraser, J. E. Allen, I. E. Bosdet, M. R. Brent, R. Chiu, T. L Doering, M. J. Donlin, C. A. D'Souza, D. S. Fox, V. Grinberg, J. Fu, M. Fukushima, B. J. Haas, J. C. Huang, G. Janbon, S. J. Jones, H. L. Koo, M. I. Krzywinski, J. K. Kwon-Chung, K. B. Lengeler, R. Maiti, M. A. Marra, R. E. Marra, C. A. Mathewson, T. G. Mitchell, M. Pertea, F. R. Riggs, S. L. Salzberg, J. E. Schein, A. Shvartsbeyn, H. Shin, M. Shumway, C. A. Specht, B. B. Suh, A. Tenney, T. R. Utterback, B. L. Wickes, J. R. Wortman, N. H. Wye, J. W. Kronstad, J. K. Lodge, J. Heitman, R. W. Davis, C. M. Fraser, and R. W. Hyman. 2005. The genome of the basidiomycetous yeast and human pathogen Cryptococcus neoformans. Science 307:1321-1324.
20. Lorberg, A., H. P. Schmitz, U. Gengenbacher, and J. J. Heinisch. 2003. KlROM2 encodes an essential GEF homologue in Kluyveromyces lactis. Yeast 20:611-624.
21. Mahajan-Miklos, S., M. W. Tan, L. G. Rahme, and F. M. Ausubel. 1999. Molecular mechanisms of bacterial virulence elucidated using a Pseudomonas aeruginosa-Caenorhabditis elegans pathogenesis model. Cell 96:47-56.
22. Moy, T., E. Mylonakis, S. B. Calderwood, and F. M. Ausubel. 2004. Cytotoxicity from hydrogen peroxide produced by Enterococcus faecium. Infect. Immun. 72:4512-4520.
23. Mylonakis, E., and A. Aballay. 2005. Worms and flies as genetically tractable animal models to study host-pathogen interactions. Infect. Immun. 73:3833-3841.
24. Mylonakis, E., A. Idnurm, R. Moreno, J. El Khoury, J. B. Rottman, F. M. Ausubel, J. Heitman, and S. B. Calderwood. 2004. Cryptococcus neoformans Kin1 protein kinase homologue, identified through a Caenorhabditis elegans screen, promotes virulence in mammals. Mol. Microbiol. 54:407-419.
25. Mylonakis, E., F. M. Ausubel, J. R. Perfect, J. Heitman, and S. B. Calderwood. 2002. Killing of Caenorhabditis elegans by Cryptococcus neoformans as a model of yeast pathogenesis. Proc. Natl. Acad. Sci. USA 99:15675-15680.
26. Mylonakis, E., R. Moreno, J. El Khoury, A. Idnurm, J. Heitman, S. B. Calderwood, F. M. Ausubel, and A. Diener. 2005. Galleria mellonella as a model system to study Cryptococcus neoformans pathogenesis. Infect. Immun. 73:3842-3850.
27. Nelson, R. T., B. A. Pryor, and J. K. Lodge. 2003. Sequence length required for homologous recombination in Cryptococcus neoformans. Fungal Genet. Biol. 38:1-9.
28. Nelson, R. T., J. Hua, B. Pryor, and J. K. Lodge. 2001. Identification of virulence mutants of the fungal pathogen Cryptococcus neoformans using signature-tagged mutagenesis. Genetics 157:935-947.
29. Nielsen, K., G. M. Cox, A. P. Litvintseva, E. Mylonakis, S. D. Malliaris, D. K. Benjamin, Jr., S. S. Giles, T. G. Mitchell, A. Casadevall, J. R. Perfect, and J. Heitman. 2005. Cryptococcus neoformans strains preferentially disseminate to the central nervous system during infection. Infect. Immun. 73:4922-4933.
30. Nielsen, K., G. M. Cox, P. Wang, D. L. Toffaletti, J. R. Perfect, and J. Heitman. 2003. Sexual cycle of Cryptococcus neoformans var. grubii and virulence of congenic a and isolates. Infect. Immun. 71:4831-4841.
31. Ozaki, K., K. Tanaka, H. Imamura, T. Hihara, T. Kameyama, H. Nonaka, H. Hirano, Y. Matsuura, and Y. Takai. 1996. Rom1p and Rom2p are GDP/GTP exchange proteins (GEPs) for the Rho1p small GTP binding protein in Saccharomyces cerevisiae. EMBO J. 15:2196-2207.
32. Park, J. I., E. J. Collinson, C. M. Grant, and I. W. Dawes. 2005. Rom2p, the Rho1 GTP/GDP exchange factor of Saccharomyces cerevisiae, can mediate stress responses via the Ras-cAMP pathway. J. Biol. Chem. 280:2529-2535.
33. Perfect, J. R., S. D. R. Lang, and D. T. Durack. 1980. Chronic cryptococcal meningitis: a new experimental model in rabbits. Am. J. Pathol. 101:177-194.
34. Perfect, J. R., and A. Casadevall. 2002. Cryptococcosis. Infect. Dis. Clin. N Am. 16:837-874.
35. Pradel, E., and J. J. Ewbank. 2004. Genetic models in pathogenesis. Annu. Rev. Genet. 38:347-363.
36. Salas, S. D., J. E. Bennett, K. J. Kwon-Chung, J. R. Perfect, and P. R. Williamson. 1996. Effect of the laccase gene CNLAC1, on virulence of Cryptococcus neoformans. J. Exp. Med. 184:377-386.
37. Steenbergen, J. N., H. A. Shuman, and A. Casadevall. 2001. Cryptococcus neoformans interactions with amoebae suggest an explanation for its virulence and intracellular pathogenic strategy in macrophages. Proc. Natl. Acad. Sci. USA 98:15245-15250.
38. Tan, M. W., S. Mahajan-Miklos, and F. M. Ausubel. 1999. Killing of Caenorhabditis elegans by Pseudomonas aeruginosa used to model mammalian bacterial pathogenesis. Proc. Natl. Acad. Sci. USA 96:715-720.
39. Thompson, J. R., C. M. Douglas, W. Li, C. K. Jue, B. Pramanik, X. Yuan, T. H. Rude, D. L. Toffaletti, J. R. Perfect, and M. Kurtz. 1999. A glucan synthase FKS1 homolog in Cryptococcus neoformans is single copy and encodes an essential function. J. Bacteriol. 181:444-453.
40. Wang, P., M. E. Cardenas, G. M. Cox, J. R. Perfect, and J. Heitman. 2001. Two cyclophilin A homologs with shared and distinct functions important for growth and virulence of Cryptococcus neoformans. EMBO Rep. 2:511-518.
41. Wojda, I., R. Alonso-Monge, J. P. Bebelman, W. H. Mager, and M. Siderius. 2003. Response to high osmotic conditions and elevated temperature in Saccharomyces cerevisiae is controlled by intracellular glycerol and involves coordinate activity of MAP kinase pathways. Microbiology 149:1193-1204.(Robin J. Tang, Julia Breg)